Chapter 3

Previous Related Literature: Enlisting the Major

Findings and Identifying the Gaps
3.0 Background

Educational attainment is an important determinant of one’s job opportunities and

relative well-being. One influence on the level of education children attain is the level
of education attained by parents. A supportive learning environment at home, expected
in families where parents have high level of education, is likely to be reflected in a
higher educational attainment of children. Parents contribute to this environment when
they impart a positive attitude towards education, when they can give financial support
to their children’s continued education and when they share in their children’s
intellectual pursuits. It has been found that the correlation between the socioeconomic
status of parents and their children is positive and significant, in virtually every society
for which estimates are available and for many different measures of status. These
correlations quantify the rate of transmission of interpersonal inequality from one
generation to the next and are often interpreted as measures of a society’s failure to
provide equality of opportunity to children from differing family backgrounds. Equality
of opportunity in education could be understood in relation to equal opportunities in
access to different levels of education irrespective to one’s socio-economic background.

Too much influence of parental status may give rise to persistence in inequalities
especially in highly diverse societies like India. It means that an individual’s success
(education or economic) is largely predictable in terms of family background and less
on individual’s effort. Presence of high level of association between parents and child’s
educational level is a violation of the principle of equality of opportunities, as observed
by many theorists. Although there are numerous public policies aimed at promoting
‘equality of opportunity’ in education like, education finance, tax exemptions on
savings earmarked for educational purposes, subsidized student loans etc. but still there
exist inequalities in educational opportunities across countries and over time. This
situation will pose an important problem, as due to the presence of substantial returns
to education estimated for most economies, barriers in access to the same for socially
disadvantaged individuals may widen the social and economic gap between groups
within societies (Herbst & Rok, 2011). Though some family background factors are
32
beyond the control of an individual but the effect of the same could be minimized
through proper formulation of educational policies.

With this background, the present chapter explore various notions of intergenerational
mobility. The chapter first deals with various concepts and measures of
intergenerational mobility and later followed by linking the concept with various other
notions such as equality of opportunity, equality of outcome including ‘nature’ versus
‘nurture’ effect on intergenerational mobility. The last section of the chapter provides
an account of international and Indian studies on mobility. Finally, the chapter list out
the gaps identified from the present studies. As put forward in the previous Chapter 2,
this chapter is an extension of literature review pertaining to the present study.

3.1 Concepts and Measures of Intergenerational Mobility

Intergenerational mobility deals with the relationship between socio-economic status of

parents and socio-economic outcomes of their children as an adult. Generally, it is
measured in terms of, income or earnings, social class or occupation and education. The
statistical measure of intergenerational transmission of inequality was proposed for the
first time by Galton (1889) as described in Chapter 2 who was interested in analyzing
the transmission of height from parents to children. In doing so, he developed the idea
of ‘regression towards mean’ which could be used to study the phenomenon of
transmission of socio-economic status. In case of socio-economic status, this implies
that most children from well-off families would be less-off with respect to their parents
but still will stay at a level above than the average. On the contrary children from less-
off families would do better than their parents but will remain below the average.
Therefore, the approach of Galton measures the speed of convergence of a given
process towards the mean. The three most widely used indicator of intergenerational
mobility namely income, occupation and education have all used the approach
propounded by Galton.

Income mobility has remained the most explored dimension among all the indicators of
mobility for the simple reason that it reflects equality of opportunity in a society.
Parental income is one of the best predictors of future life chances of children which
have both a direct and an indirect impact on a child’s attainments as adults. Directly it

33
enhances a child’s life chances through gifts and bequests along with investment in
children’s education while indirectly it will lead to better care, better lifestyle etc.
Besides this, the transfer of economic status in terms of income or earnings
characterizes a country’s income inequality (Solon, 2002) and there exists possibility
that rising income inequality will have long-run effect on reducing equality of
opportunity and intergenerational mobility (Smeeding, 2013). But the choice of type of
income is crucial to the measurement of intergenerational income mobility. Most
research studies have considered personal annual earnings for both generations as it
reflects an individual’s success in the labour market (Corak, 2004; Solon, 2002; Corak
& Heisz, 1999; Bjorklund & Jantti, 1997). All these studies are based on the assumption
that a child’s earning is solely dependent on father’s earnings and that it has no
relationship with the total income (earnings as well as other income components such
as benefits) of the entire family which is not entirely true (Moonen & Brakel, 2011).
For example, Goldberger (1989) showed that inherited endowments lead to
intergenerational persistence and Solon (2004) depicted the effect of parental
investments on intergenerational mobility. Although the results imply that the
measurement of intergenerational mobility should be based on permanent income of
both generations but, datasets mostly deals with short-term measures.

Intergenerational income mobility is measured by a function that mirrors the theoretical

assumptions of Becker and Tomes model (1979;1986) outlined in Chapter 2. In this
framework, child’s earnings/income is expressed as a linear function of parental
earnings/income. The most common approach is to measure intergenerational
earnings/income elasticity through applying least squares to the regression of a
logarithmic measure of son’s earnings/income on a logarithmic measure of father’s
earnings/income including parental age and child’s age as controls (Björklund & Jäntti,
2009; Blanden, 2009; Solon, 2002). The key parameter of interest is the coefficient
associated with the log of parental long-term earnings/income in the Galtonian
regression model termed as ‘intergenerational earnings/income elasticity’. The
coefficient indicates the extent to which child’s earnings is related to those of the
parents’ and is a summary measure of the degree of earnings persistence across
generations. Its complement, one minus the elasticity is often interpreted as the measure
of intergenerational mobility (Black & Devereux, 2010) and measures the extent to
which incomes “regress” to the mean similar as Galtonian regression approach. The

34
elasticity indicates the fraction of earnings that is on average transmitted across
generations associated with a one percent change in parental earnings. For example, a
value of the elasticity of 0.5 implies that half of the relative difference in parental
incomes is transmitted, on average, to the children. In general, the empirical estimates
of the elasticity tend to lie between 0 and 1 indicating that initial income advantage will
be wiped out over several generations (Blanden, 2009). Zero intergenerational elasticity
corresponds to complete mobility where the earnings of parents and children are
unrelated. On the contrary, intergenerational elasticity equating one indicates complete
immobility and hence perfect association between parent’s and child’s earnings. Higher
intergenerational income elasticity implies larger parental income effects and lower
intergenerational mobility. This model has now been augmented by adding controls for
variables like educational level for both parents and children (Blanden, et. al., 2004).

Another measure of interest is the intergenerational correlation co-efficient which is a

standardised measure and is widely used in literature. The correlation between log
earnings of parent’s and child equals intergenerational elasticity only when, the
standard deviation of log earnings is same for both generations (Black & Devereux,
2010). It is given as a product of ratio of the standard deviation of child’s earnings and
parent’s earnings along with intergenerational elasticity. Therefore, the correlation
factors out the cross-sectional dispersion earnings of the two generations and ranges
between 0 and 1, same as intergenerational elasticity (Torche, 2013). In other words,
the expression of intergenerational correlation proposes that the coefficients
simultaneously depicting intergenerational elasticity and intergeneration correlation
may differ when inequality changes. For example, “when inequality increases the
correlation coefficient may be the same but earnings may regress to the mean at a slower
rate” (Aaronson & Mazumder, 2005). As per Black & Devereux, (2010), there is no
reason to consider that one measure is better than the other, but one should be aware of
the difference between the two measures especially while comparing across time and
space. One major advantage of using intergenerational elasticity over correlation is that
the former is unbiased to any measurement errors in child’s income which is the
dependent variable and hence estimation becomes easy with real world data (Javed &
Irfan, 2014).

The early assessments of intergenerational earnings elasticity during 1980s considered

single-year measure of earnings for both the generations and applied simple least square

35
regression on the available data. As per Becker & Tomes (1986), the parent-son
intergenerational elasticities were generally estimated at 0.2 leading to the conclusion
that earnings were not strongly transmitted between generations. According to them,
“aside from families victimized by discrimination almost all earnings advantages and
disadvantages of ancestors are wiped out in three generations” (p.32). But in 1990s,
intergenerational earnings elasticity stood at 0.40 (Solon, 1999) and a higher value of
around 0.5 was estimated for 2000s (Hertz, 2005; Mazumder, 2005). These updates
were due to better measurement of earnings in either generation or due to the use of
nationally representative dataset.

3.2 Intergenerational Mobility and Equality of opportunity

Intergenerational socio-economic association is often considered as a measure of

equality of opportunity by the social scientists. A strong intergenerational association
means that socio-economic status of parents and their adult off-springs are closely
replicated while a weak association means that individual outcomes are relatively
independent of social origins, meaning thereby all individuals have the same chance of
success or failure. In the latter case, it could be said that society provides equal
opportunity to its members and individual’s success is independent of family
background. The concept of equality of opportunity stemmed from this viewpoint that
an individual should not be held responsible for circumstances beyond her control and
is possibly the most supported conception of justice worldwide. There are however
varied interpretations of this term and the concept has evolved over time as traditionally
equality of opportunity means eradicating any kind of legal barriers to education and
occupation.

In the social choice theory, egalitarianism means equality of welfare or utility. This
view of egalitarianism has faced criticism as this kind of equality fails to hold a person
responsible for their choices or preferences which would ultimately affect the outcome.
Keeping this criticism in mind, political philosophers starting with John Rawls (1971)
developed a new concept of egalitarianism which inserted personal responsibility as an
important qualifier of the degree of equality that is ethically desirable. Therefore, since
Rawls, the development of egalitarianism theory placed much more importance on
equality of opportunities rather than equality of outcomes where opportunities are

36
interpreted in varied ways. Equality of opportunity is understood through metaphors
like ‘level the playing field’ and ‘starting gate equality’. Both these metaphors
emphasize the fact that equal opportunity policy should create a level playing field
initially or provision of equal initial resources so that outcomes will reflect individual
effort exclusively minus the influence of factors which are beyond the control of an
individual. In this sense, inequality of outcomes is ethically desirable if the playing field
was initially at level.

After Rawls, the major philosophical contributions towards this concept were given by
Amartya Sen (1980), Ronald Dworkin (1981a, 1981b), Richard Arneson (1989) and
G.A. Cohen (1989). All these contributions supplemented Rawl’s (1971) notion of
equality of fundamental rights and equal access to primary goods. More recent studies
have dealt with either equity in terms of equalization of capabilities (Sen, 1992) or
opportunities (Roemer, 1993, 1995, 1996, 1998). Roemer conceptualizes the equality
of opportunity concept by translating the existing concepts into a coherent framework
which revolves around five components and formalizes the “level the playing field”
approach. This approach is defined by five words which are i) objective ii)
circumstances iii) type iv) effort and v) instrument. In Roemer's view, “policies to
equalize opportunity should make the attainment of an objective independent of the
individuals' circumstances and dependent only on their efforts” (Roemer 2004). Roemer
(2005) “also stresses that equality of opportunity does not necessarily imply equality of
outcomes but rather level(ling) the playing field so that all have the potential to achieve
the same outcomes; whether or not, in the event, they do, depends upon individual
choice”. Roemer's theory has been extensively discussed (Hurley, 2002; Fleurbaey,
2001; Kolm, 2001; Roemer, 1995, 2003, 2006) and has been largely applied in
empirical works (Betts & Roemer, 2004; Roemer, et al., 2003; Llavador & Roemer,
2001;). Therefore, equality of opportunity is achieved in a society when individuals can
achieve their desired objective, solely determined by their choices and personal effort
rather than determined by circumstances beyond their control (Arneson, 1989; Cohen,
1989; Roemer, 1993, 1998). This indicates that differences in values of the objective
are ethically desirable in equality of opportunity perspective, if they are achieved due
to differential efforts and not due to differential circumstances.

One obvious application of the idea of equality of opportunity is in intergenerational

context as the relationship between the circumstances beyond an individual’s control

37
and the objective can be described in terms of intergenerational correlation. The
objective could be income, education or occupation of a child in adulthood. In terms of
transition matrix, intergenerational transition matrix is a table whose ijth entry is the
fraction of children whose parent or parents achieved an objective i.e. education,
income or occupation level i, and who in turn achieved objective j. The implicit social
goal is to achieve a transition matrix whose rows are identical that is, the distribution
of the outcome among children is independent of the parent’s outcome value. Roemer
(2004) has identified four channels through which parents can influence children’s
achievement of objectives and these are; a) provision of social connections which may
facilitate access to jobs, admission to particular schools or colleges, or access to other
sources of human capital b) formation of beliefs and skills in children through family
culture and monetary and non-monetary investments which shapes skills, aptitude,
beliefs and behavior of a child c) genetic transmission of ability, personality and some
aspects of health which are valued in the labour market and d) formation of preferences
and aspirations in children. Therefore, equality of opportunity policy instrument should
be framed in such a way so that the achievement of the objective of individuals should
be function of their efforts only and not their circumstances.

The major issue is to finalize what constitutes ‘effort’ and ‘circumstances’ and Roemer
(1998) puts forward that this choice must be made by society through ethical and
political processes. Social origin variables like sex, race, parental education or wealth,
region of birth, ascribed caste or ethnic group are considered as the most agreed form
of circumstances beyond individual’s control. Hence these variables provide a direct
link between intergenerational transmission and equality of opportunity besides the four
channels mentioned above through which circumstances affect achievement of
objective in an intergenerational context. Therefore, the interaction between families,
labour markets, education system and policies all structure a child’s opportunities and
determine the extent to which achievement as adult are related with family background.

Besides, transition matrix, intergenerational mobility is also measured using regression

model where β coefficient called intergenerational elasticity measures the degree of
outcome persistence across generations. While the intergenerational elasticity reveals
the statistical association in economic status across representatives of two generations,
empirical measures of inequality of opportunity quantify the extent to which total
inequality can be explained by a set of ‘circumstances’. Therefore, as discussed above,

38
the distinction between individual efforts and pre-determined circumstances is the
conceptual basis for the definition of inequality of opportunity. This indicates that
intergenerational elasticity cannot be used as a direct measure of inequality of
opportunity and the same have been emphasized in the studies of Roemer (2014, 2012)
and Jencks and Tach (2006). According to them, intergenerational elasticity of zero is
not optimum and as clarified by Jencks and Tach (2006), “equal opportunity does not
imply eliminating all sources of economic resemblance between parents and children”
(p. 4). In fact, to draw some inference about equality of opportunity from
intergenerational elasticity one needs to distinguish between differences in
circumstances over which individual has no control and efforts for which an individual
should be responsible.

Although equality of opportunity in a society could not be used as a direct measure of

intergenerational mobility of that society but both these measures are related with one
another. Brunori, et. al. (2013) offer a particularly clear overview and find that indices
of inequality of opportunity are in fact strongly correlated with indicators of
generational mobility, be it in earnings or in education. As these authors put it,
“[I]nequality of opportunity is the missing link between the concepts of income
inequality and social mobility; if higher inequality makes intergenerational mobility
more difficult, it is likely because opportunities for economic advancement are more
unequally distributed among children” (p.17). In their study, they show evidences of a
positive correlation between the index of inequality of opportunity and the
intergenerational elasticity across countries.

3.3 Intergenerational Mobility and Equality of Outcome

Income inequality if rises would make family background as an important determinant

in achieving desired outcome by an adult with the role of their own efforts being
minimized. According to OECD (2011), rising income inequality “can stifle upward
social mobility, making it harder for talented and hard-working people to get the
rewards they deserve. Intergenerational earnings mobility is low in countries with high
inequality such as Italy, the United Kingdom, and the United States, and much higher
in the Nordic countries, where income is distributed more evenly” (p.10). Therefore,
higher inequality is related to skewed opportunity which reduces intergenerational

39
mobility, or in other words, increase in income inequality will probably interfere with
the economic mobility for the next generation of young adults (Corak, 2013). This
indicates that countries with greater income inequality are also the countries where
greater amount of both economic advantages and disadvantages are transferred from
parents’ generation to child’s generation. This positive relationship between income
inequality and intergenerational mobility is portrayed using the ‘Great Gatsby Curve’
which was introduced by Allan Kruger in a speech, “The Rise and Consequences of
Inequality,” to the Centre for American Progress on January 12, 2012. The curve ranks
countries along the horizontal and vertical axes with income inequality in a country
measured by Gini coefficient in the horizontal axis and the vertical axis depicting
intergenerational economic mobility measured as the elasticity between paternal
earnings and a son’s adult earnings. The upward sloping regression line demonstrates
the ‘Great Gatsby Curve’ with Kruger (2012) using it to predict that “persistence in the
advantages and disadvantages of income passed from parents to children ….. [will] rise
by about a quarter for the next generation as a result of the rise in inequality that the US
has seen in the last 25 years” (p. 4).

One of the main reasons to be interested in the relation between income inequality and
intergenerational mobility is that it can be observed cross-nationally, economically
unequal countries are also the least mobile (Jerrim & Macmillan, 2014). A study by
Corak (2013) dealing with cross-national data suggested that during early to mid-1980s,
Finland, Sweden, Norway and Denmark were the most equal while the United Kingdom
and the United States, the least. His findings suggested that in countries like Finland,
Norway and Denmark the relation between parental economic status and the adult
earnings of children is weakest. In these countries, less than one-fifth of the economic
advantage or disadvantage which a father might have during his lifetime is transferred
to a son in their adulthood. But in Italy, the United Kingdom and the United States
roughly 50 percent of any advantage or disadvantage is passed on. Another study by
Smeeding (2013) suggested a moderately positive relationship between inequality in
income and intergenerational elasticity of the same of eleven industrialized countries
with higher levels of inequality associated with lower mobility rates; the rank order
correlation being 0.62. Their findings also revealed that Scandinavian countries are
least unequal with high mobility rates while Italy, the United States, France etc. having
high levels of inequality and lower levels of intergenerational mobility.

40
Although the ‘Great Gatsby Curve’ does not imply any causal relation between
inequality and mobility but theories of child development and economic mobility
advocated towards combining inequality and mobility as a starting point for
understanding the causal process and its policy implications (Corak, 2013). Studies
revealed that societies with more income inequality exercises greater disparity in
investment of resources in children by rich and poor families (Corak, 2013; Solon,
2004; Becker, 1986). Socio-economic status influences children even when they are in
utero (e.g. quality of pre-natal care) which in turn influences early cognitive and social
development. Consequently, large socio-economic differences in cognitive functioning
emerge even before compulsory schooling has begun (Cunha, et. al., 2006). Eventually
both access to good schools and jobs gets influenced by family resources and
connections. The degree of inequality in labour market affects parental resources and
the investment in children and ultimately the return to education children receives as an
adult. This entire process has an impact on children’s earnings in adulthood. Besides
family influence, Solon (2004) mentioned that the reasons for the changes in the
intergenerational elasticity across countries is due to the different balances policies
between the labour market and public policy towards determining the life chances of
children. These institutions together determine the degree, to which parental traits
important in the labour market are transferred to children, efficiency of public and
private investment in generating human capital along with the labour market returns to
human capital. All these taken together determine the degree of intergenerational
income mobility. Therefore “The Great Gatsby Curve is a summary of all of these
underlying gradients, reflecting the outcome of a whole host of ways that inequality of
incomes affects children” (Corak, 2013; p. 7).

Studies has derived various perspectives on the Great Gatsby Curve, one of which is
heritability of traits between parents and their children as mentioned by Becker &
Tomes, 1979. According to them “endowments of capital among children are
determined by the reputation and connections of their families” (p. 1153) and if the
traits are too strongly related and if they are valued in labour market over time then also
there will be intergenerational association in incomes. Instead of this, significant
demographic diversity also, across countries will generate an upward sloping Great
Gatsby Curve even if the society is meritocratic (Becker, 2013; Roemer, 2012).
Therefore, Great Gatsby Curve only reveals the underlying reasons for differences

41
across countries and assesses the extent to which they are relevant for public policy
instead of showing the outcomes. Alternative perspective of the Great Gatsby Curve
could be derived from the returns to education (Solon, 2004). The rate of return to
schooling is taken as an indicator of the degree of inequality in labour market as parents
with more income will invest more and hence those societies will be less generationally
mobile. Solon also suggested that public policies play a major role in reducing labour
market inequality if the policies are of relatively more beneficial to the relatively less
well-off.

Therefore, it could be said that in the literature of economic inequality, both equality of
outcomes and equality of opportunities are associated with different views of social
justice. While, equality of outcomes deals with the distribution of the product of the
efforts and the circumstances under which this effort is made, equality of opportunities
depends on the notion that individuals should have equal opportunities to start with, so
that they can achieve desired outcome as an adult. Advocates of equal opportunities
argued that inequality arising from differences in effort of individuals for which they
are responsible need not to be compensated for. Both the notions of inequality be it
outcome or opportunities is related with the concept of intergenerational mobility as
described above. Although the theorists have majorly dealt with intergenerational
mobility in income while discussing its relationship either with equality of opportunity
or with equality in income but the same could be advanced to describe the relationship
between other types of mobility and both the concepts of equality.

3.4 Relating Intergenerational Educational Mobility with the Two Notions of

Equality

A person’s educational achievement is positively correlated with that of her parents’

and this holds true for virtually every society of the world for which estimates are
available. This correlation is positive and significant in both statistical and practical
senses and quantifies the rate of transmission of interpersonal inequality from one
generation to another. It is often interpreted as measure of a society’s failure to provide
equal educational opportunities to individuals from different socio-economic
backgrounds. Equality of opportunity in education means that circumstances of an

42
individual which is beyond her control such as family background should not affect her
educational achievement. The educational policies should be framed in such a way in
order to compensate for the effects of pre-determined circumstances on an individual’s
educational outcomes. The topic has been raised in policy debates and most societies
have adopted policies aimed at reducing the impact of family background, thereby
equalizing educational opportunities.

In a perfect world every individual irrespective of their races, socio-economic

backgrounds, gender would not only have the same opportunity to receive their desired
level of education but also, they would be able to take full advantage of these
opportunities. But this does not happen, and educational attainment of individuals
varies with their backgrounds identified as ‘circumstances’ beyond individual’s control
by equal opportunity theorists. Family background is relevant as the individual has no
control in selection of his or her family and thus could not be held responsible for any
effect of it on his/her status during adulthood. Therefore, the more important the family
background is, for instance as measured by parental education, for a person’s
educational achievement the less is equality of educational opportunity (Bjorklund &
Salvanes, 2010). In other words, intergenerational transmission of education would
persist strongly across generations if parental education is too much related with child’s
educational attainments. If circumstance variables include parental characteristics like
parental education then it is obvious that equality of opportunity could be analyzed in
intergenerational framework as depicted in various studies (Brunori, et. al., 2013;
Woessman, 2004; Bourguignon, et. al. 2003) For instance, if parental education is
included in circumstance variables and individual’s schooling is a part of effort then
part of educational attainments by an individual describes intergenerational educational
mobility. In fact, the degree of intergenerational educational persistence, partly
determines the share of current educational inequality which is due to differences in
opportunities or circumstances. “Influence of parental background variables in
educational achievements as well as other economic outcomes reveals marked
problems of intergenerational transmission of poverty” where less parental education
shapes opportunities and is related with less education of an individual (Barros, et. al.,
2009). A branch of intergenerational mobility literature focusing directly on educational
transmission have estimated that part of schooling inequality explained due to
characteristics of parents, used to quantify the inequality of educational opportunities

43
and the rest is ascribed to heterogeneous individual’s efforts. Studies by Barros and
Lam, (2003), Lam (1999) for Brazil and Behrman, et. al. (2000) for Latin American
countries followed the similar approach. A study by Bourguignon, et. al. 2003, applied
the same approach in Brazil and argued that attributing the decline in influence of
parental education over time to general rise in mean education is incorrect as the
conditional distribution of educational opportunities have almost remained constant
over time.

Studies by Schutz, et. al. (2008) and Macdonald et al. (2010) examined the question of
intergenerational persistence in educational achievement linked with inequality of
opportunity and provide empirical evidence for the same. The former study assessed
the effect of various education policies such as ability tracking, pre-school education,
length of the school day and educational spending on educational achievements of a
child measured in terms of the International Association for the Evaluation of
Educational Achievement’s Trends in International Mathematics and Science Study
(TIMSS). Their theoretical and empirical model suggests that comprehensive school
systems and extensive early-childhood education can increase the equality of
educational opportunity for children from different family backgrounds, thereby
increasing intergenerational educational mobility. The latter study used OECD’s
Program of International Student Assessment (PISA) score as a measure of ability of a
student and found that the same is positively related with household wealth indicating
students from poorer households are less able to accumulate cognitive skills than
students from wealthier households. The presence of this inequality in circumstance
beyond the student’s control may lead to intergenerational transmission of poverty.
Another study by Brunori, et. al. (2013) documented the association between
inequalities of opportunity and intergenerational correlations of education for 23
countries and the correlation coefficient came out to be 0.60 (p-value: 0.0021). This
means that the measure of inequality of opportunity is strongly positively correlated
with the correlation coefficient of parent and child schooling attainment. Although this
kind of relationship is only possible when the observed set of circumstances beyond an
individual’s control includes only parental education assumed to be enough statistic for
all observed circumstances and the same have been pointed out by research studies
(Barros, et. al. 2009; Bourguignon, et. al. 2007).

44
Therefore, while the literature on intergenerational mobility is now increasingly being
related with studies on inequality of opportunity, as family background is a major
determinant of opportunities, but still they are not perfect substitutes. This is simply
because the former seeks to find out the transmission of one specific indicator
(education, earnings, occupation etc.) from parental generation to child’s generation,
thereby separating the effect of other background circumstances such as race, gender or
geography. On the other hand, the latter seeks to measure the aggregate effect of all
observed circumstances including family background on current inequalities related
with an individual. To this end it could be concluded that parental background is the
most important pre-determined circumstance for an individual which varies across
countries and over time periods and could not be measured ex-ante.

3.5 ‘Nature’ vs ‘Nurture’ Effects on Intergenerational Mobility

Both the conceptions of equality in some way or other are related with the effect of
family background on child’s attainments as adults and in this sense inequality and
mobility are intrinsically related. In countries with greater inequality, both economic
advantages and disadvantages are transferred from parents to child; which may give
rise to persistence between parents and children’s outcomes in terms of education or
income. Understanding the degree of this persistence has been a longstanding goal of
social science where studies are trying to look beyond mere intergenerational
correlations and elasticities. Research studies have now placed increased importance on
the causal mechanisms that underlie this relationship which focuses on disentangling
the respective contributions of the transmission of innate abilities termed as nature and
family background called nurture. These studies typically exploit natural experiments
by contrasting identical and fraternal twins, their offspring, relatives and siblings, or
biological and adopted children and most concluded that more than half of observed
differences in socio-economic outcomes could be attributed to differences in genes
(Sacerdote, 2008; Bjorklund et al., 2005; Plug & Vijverberg, 2003).

The theoretical literature in economics, on understanding the causal process of

association between parents’ and children’s socio-economic outcomes started with
Becker & Tomes model (1979; 1986) where they linked parental resources to income
of children through human capital investments. Several other empirical findings

45
showed that more family income earned on average by more educated parents stimulate
income of children through further schooling of children. Besides parental income,
parental education also has a profound effect on educational attainments of a child
which in turn influences their income in adulthood. All these findings have led the
economists to put emphasis on nurture towards transmission of socio-economic status
from parents to children. But studies by psychologists and behavioral geneticists have
revealed that genetic factors explain about 50 per cent to 60 per cent of variation in
adult IQ while family environment explains little of the variation (Harris, 1998; Devlin,
et.al., 1997; Cardon & Cherny, 1994). Hernnstein and Murray (1994) in the Bell Curve
argued that ability measured in terms of IQ matters most and highly educated parents
are more able than less educated parents. According to them, if ability is transmitted
from generation to generation then education would be persistent across generations
which are influenced by family income; therefore, ability affects income persistence
too. Studies involving adopted children found that the genetic influence of family
background is more important than the influence of family environment in explaining
the ability of children (Loehlin, et. al., 1994; Scarr & Weinberg, 1978). Besides this,
two studies involving only 65 and 95 pairs of identical twins reared apart found that
about 70 percent of ability is inherited (Bouchard, et. al., 1990; Bouchard & McGue,
1981).

Plug & Vijverberg, 2003 have identified four possible strategies through which
heritability of ability takes place. The first one is effect of environment on twins which
works under the assumption that in the same family environment the differences
between fraternal and identical twins will be attributed to genetic contribution.
However, Plomin, et. al. (1990c) found out that study of nature and nurture outcomes
using twins may be biased as identical twins are treated more similarly than fraternal
twins. In addition to these twin studies generally rely on small samples or on data where
twins volunteer to participate, all these may give a bias result. The second strategy takes
into consideration relatives who are either raised together or apart. Here the differences
among relatives raised in different families determine the environmental impact after
controlling for genetic structure among relatives. The major problem in this strategy is
the fact that in reality children who are relatives share more than just genes and since it
is not clear from the data how family environments of siblings in different families are
related, resulting nature estimates are biased and in most situations too high

46
(Goldberger, 1979). The third strategy compares between children who are their
parent’s offspring and adopted children. After controlling for family environment in
which both adopted and biological children are raised together, this strategy identifies
contribution of genetic component as adopted children are genetically unrelated to
families. Small sample size, non-randomness of the placement of adopted children in
adopting families, parents’ emotional and material differentiation between biological
and adopted children and missing relevant information on the biological background
and adoption history of adopted children are some problems of this strategy (Plug &
Vijverberg, 2003). The fourth strategy identifies genetic contribution with data of
identical twins raised in different families. Plomin & Petrill (1997) summarize many of
these family, twin and adoption studies. Their main concern deals with heredity and
cognitive ability measured as IQ (or IQ related) test scores and they revealed that about
half of the variation in IQ test scores is explained by genetic factors. According to them
family environment accounts for about a quarter of the variation in IQ scores when
children are young but its influence decreases as children grows up.

Economists have mostly deviated from the viewpoint of psychologists and behavioral
geneticists that IQ is a dominant factor in predicting economic and social success
(Korenman & Winship, 2000; Goldberger & Manski, 1995). But frequent finding of a
non-declining association between children’s socio-economic status especially
education and earnings has strengthened the idea of some sort of genetic link underlying
these educational and earnings choices. Therefore, studies tried to find empirically the
role of inherited ability in economic models of investment in schooling. For example,
Behrman, Taubman and Wales (1977) did a study that use correlations between
identical and fraternal twins to separate nature from nurture effects. They observed that
if parents constitute genetically random matches, about 44 percent of the variance in
educational attainment would be attributable to genetic differences. Sacerdote (2002)
using data on biological and adopted children, found only small effects of parents’
education and income on the adoptees’ IQ, but large effects on child’s schooling. With
respect to the effect of mother’s and father’s years of education on schooling of child
about 36 and 43 percent is genetically determined, respectively. In fact, he concluded
that for years of schooling and college education nurture seems to be the dominating
factor. Indeed, several researchers studying intergenerational associations in education
or income of adoptive families suggested a positive association between children’s and

47
parents’ education and income in adoptive families and the association is statistically
significant when the sample sizes are large enough. Plug & Vijverberg (2003) using an
intergenerational sample of families from Wisconsin Longitudinal Survey, US
estimated based on a comparison of biological and adopted children and found that
about 55 to 60 percent of the parental ability is genetically transmitted. Sacerdote (2007)
studied a adoptees sub-population within which the placements were approximately
random and found a significantly positive association in the education of the adoptees
and their adoptive mothers. Björklund, et. al. (2005) using Swedish data on adopted
children's biological and adoptive parents estimated intergenerational mobility in
earnings and education and argued that both pre-birth and post-birth factors contribute
towards intergenerational transmission in education and earnings. Another study by
Bjorklund, et. al., (2007) also used Swedish data to explore the sources of the
intergenerational transmission of socioeconomic status. Their analysis focusing on
children reared up in six different family circumstances namely raised by both
biological parents, raised by the biological mother without a stepfather, raised by the
biological mother with a stepfather, raised by the biological father without a stepmother,
raised by the biological father with a stepmother and raised by two adoptive parents
also revealed substantial roles for both pre-birth and post-birth factors.

Therefore, by using empirical strategies in order to identify causality links studies like
these, lean towards suggesting that both inherited ability and family characteristics are
crucial in shaping the links between parental and child’s status. However, precise tests
of the “nature vs. nurture” hypothesis, based on data on IQ tests, reveal that the
comparative effect of cognitive abilities is limited and cannot provide explanation for
the entire effect of parental background (Bowles & Gintis, 2002; Arrow, et. al. 2000).
A study by Blanden, et. al. (2007) for UK revealed that when intergenerational
correlations of income are further decomposed into ability, education and labour market
attachment then abilities describe a very small fraction of social immobility while most
of effect is still due to educational attainment of the present generation. Although, there
is no consensual answer but still most of the explanation mainly in the economic and
cultural resources of parents rather than on genetic transmission (Chechhi, et. al. 2008)

The debate between the importance of nature and nurture is important from a policy
perspective. If the observed intergenerational correlations are primarily due to genetic
transmission of abilities and preferences, then there is little role for policy interventions

48
towards promoting economic and social mobility in the society. However, if the family
background variables along with other contextual characteristics of the country are
important then government can frame policies in order to reduce persistent inequality
in opportunities.

3.6 Intergenerational Mobility: An account of International Studies

With the background of previous sections in mind the present section attempts to give
an account of international studies and Indian studies on intergenerational educational
mobility. This is because examining temporal changes and their comparison across
countries enables the contemplation of how institutional changes affect
intergenerational mobility. There are different institutions that may simultaneously
affect intergenerational mobility in education. For example, the labor market, by
determining in a broad sense ‘the return to education’, influences the incentives to
invest in a child’s human capital. Moreover, the educational system can affect the cost
of this investment by modifying the general availability and the quantity of educational
resources. Although there has been incongruence in findings among the nations but still
there is no doubt that contextual characteristics of different countries and cohorts play
a key role in determining the intergenerational transmission of education. However, the
question is how these contextual characteristics interact with the family background
and shape patterns of intergenerational educational mobility in different societies and
time periods. The section is arranged in the following order. First it gives an account of
multi-country studies followed by studies related to European Nations. Third section
deals with studies of Latin American countries followed by Scandinavian Countries.
Lastly it deals with Australia and Asian nations.

3.6.1 Multi-Country Studies

A study by Chevalier et. al (2009) examines the educational mobility between fathers
and their children using the International Adult Literacy Survey for 20 countries. The
authors using Coefficient of Variation, Gini Coefficient, Transition Matrices and
Multivariate Analysis found answers to the questions of the study, “a) What is the
relationship between educational inequality and educational mobility? b) Does the

49
above question tell us anything labour market incentives? c) Is educational mobility
higher for males or females? d) Has educational mobility increased over time? (p.14)”
Using correlation matrix, the authors found that inequality is negatively associated with
mobility. The results show that countries with egalitarian educational systems in terms
of the outcome of schooling completed, also enjoy greater educational mobility.
Educational mobility is marginally higher for males while inequality in educational
achievements is almost the same, if measured by eigen value index. So, this mean that
men have been less tied down by their background in making educational transitions.

The authors found that the most egalitarian countries are Chile, Italy and Sweden and
the most unequal are Canada, United States (US) and Switzerland (but not the French
speaking part). Except for Chile, this conforms to expectations and previous evidence
in the literature. Other blocks of countries are of interest. Nordic countries appear to
be rather equal except for Denmark, having a father with secondary education is
associated with a penalty lower than 10 percentage points. The Central and Eastern
European Countries are less homogeneous. The expectation here is that their
communist past should have made the distribution of education in these countries
relatively egalitarian. The bottom of this ranking is also surprising with the US,
Denmark and Norway being the countries where the intergenerational link is the
strongest. The reduced link between father’s and daughters’ educational attainment may
be responsible for some of the surprising results in this ranking.

Another study by Broucker & Underwood (1998) deals with the same International
Adult Survey Literacy (IALS) data of 11 countries to make an international comparison
with a focus on post-secondary education. The correlation between parent’s and child’s
educational attainments reveal that in all countries, what mattered was the higher level
of education of the two parents, not which parent had the higher level. For example, in
Canada, the correlation between highest parental level of education and the education
of the respondent was 0.41, higher than the correlation for father and respondent or
mother and respondent. The correlation for Canada was about average compared to
other countries. For all countries, the correlation was stronger between the level of
education of the respondents and that of the parent with the highest level of education,
than for either just the mother or just the father. This was the case regardless of age or
gender of the respondent. The higher the level of education of the parent, the higher
was the proportion of respondents who attained a postsecondary degree or diploma in

50
all countries studied. The odds of attaining a post-secondary degree whose parents have
postsecondary education vs. individuals whose parents have not completed secondary
school reveal that the odds ranged from a low of 2.0 in Australia to a high of 5.8 in
Poland. That is, in Australia individuals whose parents have postsecondary education
had a 2.0 times higher chance of completing postsecondary education than those whose
parents did not complete secondary school, compared to 5.8 times better chances in
Poland. The odds in Canada, at 2.4, were closer to that of Australia.

An empirical study by Hertz et. al (2007) evaluated trends in intergenerational

persistence of educational attainment across 50 years for a sample of 42 nations around
the globe. The data are drawn from World Bank Living Standards Measurement
Surveys (LSMS), similar household surveys conducted by national statistical agencies,
and country surveys affiliated with the European Social Survey (ESS), the International
Social Survey Program (ISSP), and the International Adult Literacy Survey (IALS).
The survey years are the most recent available that include information on parental
education; they fall between 1985 and 2004, with most having been conducted between
1994 and 2004. The age criterion was 20- 69 year old which is between 1916 and 1984.
Sample sizes were from 1,49,477 for Brazil to 1,047 for the Philippines and in total
around 3,90,000 individual parent-child pairs were represented.

The authors used intergenerational education correlations and regressions to compare

between different countries. The seven highest intergenerational schooling correlations
are found in seven Latin American countries: the regional average is 0.60, compared to
values between 0.36 and 0.41 for the other four regions. Among Asian nations,
Indonesia represents a high-outlier (at 0.55) while rural China’s average value across
cohorts was just 0.20. The high-income Western nations and the communist countries
vary between 0.54 (Italy) and 0.28 (Kyrgyzstan) and Italy. The four Nordic nations
(Denmark, Finland, Norway and Sweden) had an average parent-child schooling
correlation of 0.34, which was significantly lower than the non-Nordic average (0.41,
p<0.01), and lower than the average for the Eastern Bloc (0.41, p<0.01). Globally, a
fall of 30 points in the estimated mean regression coefficient over 60 years could be
seen from 0.80 in 1920 to 0.50 in 1980. In regional terms in Latin America, long-run
average persistence is high, still, every country except Nicaragua showed a significant
reduction in either grade or standardized persistence, or both, over time, and there were
no significant positive trends for either measure for any country. These declines have

51
lowered the average regression coefficient for the two most recent cohorts (the 20 to 29
year olds) to 0.60, and the correlation to 0.56, numbers that are still high by international
standards. Average education in the ten Asian surveys grew from about one to four
years for the parents and from 4.7 to 7.5 years for the children, although there was
considerable variation across countries in both starting and final values. Five of the ten
trend estimates for grade persistence in the 25-69 age group are negative and significant
at the ten percent level or better; one is significantly positive. However, only two
nations (Rural China and Sri Lanka) show any significant evidence of a negative trend
in standardized persistence, while for five countries this trend was positive. Indonesia
and Vietnam provide the clearest examples of differences in the signs of these two
trends. In Indonesia, for example, the regression coefficient fell by about 0.04 per five
years, while the correlation coefficient rose by about 0.01 per five years.

Three of the four African surveys paint fairly similar pictures: in Egypt, the KwaZulu-
Natal province of South Africa, and in Ghana, average parental education grew from
below one year to between 3.4 and 4.8 years, over similar time periods, while average
education in the second generation grew from around two to around eight years. In all
three countries the regression coefficient linking parental to child schooling started
quite high (above unity for Egypt and Ghana) but fell significantly for those born after
about 1950 or 1955. In Egypt the correlation coefficient also fell, by about two points
per five years, or from 0.62 to 0.40 over 50 years, indicating a reduction in standardized
persistence. In South Africa and Ghana, however, the correlation coefficient did not
move, hovering at around 0.40. The fourth country, Ethiopia, produces erratic initial
results: in the earliest cohorts nearly 98 percent of parents had no schooling, rendering
the estimated slope of the regression line essentially meaningless. From 1952 on, there
is enough dispersion in parental education to generate more stable estimates. For this
period, the average regression coefficient stands at 0.92, but the correlation is only on
the order of 0.18. In other words, small differences in parental education make a large
difference (nearly one-for-one) in the expected value of children’s education, but
parental education does not vary a great deal across households and can explain only a
modest share of the overall variation in child schooling. In case of South Africa there
exists unchanging grade persistence for those born before 1950 but falling coefficients
thereafter.

52
With the apparent exception of Kyrgyzstan, the eight Eastern Bloc countries surveyed
all had higher average initial levels of education than were found in Latin America,
Asia, or Africa. Five of eight former communist countries fall below the full sample’s
median in terms of their long-run average intergenerational correlation coefficients,
while three nations, Slovenia, Hungary, and Poland, have above-median correlations
(of 0.52, 0.49, and 0.43). Results using the ESS surveys instead of the IALS were
similar. There exists grade persistence for Slovenia, Kyrgyzstan, Poland and Hungary.
However, only Slovenia saw a reduction in standardized persistence, and this statistic
rose significantly in Hungary, the Ukraine and the Czech Republic. The average of the
cohort-specific intergenerational schooling correlations for Denmark, Finland, Norway
and Sweden is 0.34, compared to 0.41 for the non-Nordic high-income Western nations;
this difference is statistically significant at the one-percent level. The Nordic nations
also display significantly lower levels of standardized educational persistence, on
average, than do the Eastern Bloc countries.

In summary it could be said that for most countries and for the sample as a whole, the
regression coefficient of parents’ education as a predictor of schooling in the next
generation fell substantially over 50 years, indicating a long-run decrease in this basic
measure of the intergenerational persistence of educational inequality. However, the
authors do not observe any such trend in the correlation between parent and child
schooling, which is a standardized measure of persistence and which has risen in as
many countries as it has fallen. It thus appears that a one-year difference in parents’
education now corresponds to a smaller difference, on average, in the expected value
of their children’s schooling than it previously did, this being a statement about a
weakened statistical association, not a diminished causal connection. On average for
the countries in the sample, a one-standard-deviation difference in parental education
corresponds to a schooling difference of about 0.4 standard deviations in the next
generation and this figure has held steady for half a century. Otherwise put, around the
world, parents’ schooling, by itself, explains as much of the variance of children’s
schooling as ever. Strikingly, the seven Latin American nations in the sample occupied
the top seven positions among the 42 countries considered when ranked by their parent-
child schooling correlations. These seven had an average parental schooling correlation,
for adults between the ages of 20 and 69, of 0.60, compared to 0.41 for eight Eastern
Bloc nations, 0.39 for ten Asian nations, 0.39 again for 13 Western nations and 0.36 for

53
a small sample four African countries. The Nordic nations stood out for displaying less
persistence, on average, than the non-Nordic high-income nations and less than the
formerly communist countries.

3.6.2 Studies Related to European Nations

A study by Chevalier et. al (2013) in the United Kingdom (UK) addresses an important
issue in the existing literature; the causal effect of parental education on children and
dealing with the separate effects of mother’s and father’s education along with the
causal effect of household income after controlling for education. Using a British
cross-section dataset of Labour Force Survey (LFS) the authors confirmed through
regression analysis that parental education levels is positively correlated to child’s
education attainments. Least squares analysis showed stronger effects of mother’s
education than father’s education and the effects are stronger on sons than on
daughters.

When controlling for paternal income, the instrumental variable results reinforce the
role of maternal education, especially for daughters, where the estimates increase
almost ten times. One year of maternal education, for mothers affected by the
instruments, increases the probability of her daughter staying on by 20 percentage
points. The effects on sons are only half of this and just on the border of statistical
significance. In contrast, paternal education has no statistical effect on the probability
of remaining in education for either son or daughter. Accounting for the endogeneity
of paternal income also increases the elasticity of income on schooling decisions,
however depending on the set of instruments the effects is imprecisely estimated,
and the preferred specification becomes insignificant, even if a large point estimate
is still found for sons.

Chevalier (2004), using the UK Family Resources Survey cross-section data, finds
that for both parents, OLS estimates of the effect of one year of parental education on
the probability of post-compulsory education is about 4 per cent, with the effects
slightly larger for sons than daughters. The author after considering, the legislation
related to change in the school leaving age as an instrument for parental education,
found that the impact of parental education on the child of the same gender h a s
increased substantially, for a sample of biological parents. Similarly, Galindo-Rueda

54
(2003) using the earlier 1947 reform and regression discontinuity, found significant
causal effects but only for fathers.

In case of Austria, several studies have documented the educational expansion in the
country in 20th century but few studies have dealt with measuring the degree of
persistence of educational attainment over generations. Some studies refer to
intergenerational persistence in educational attainments in Austrian scenario e.g.
Bacher (2003) and Landler (1997) analyzed the influence of social background on the
educational achievements of university students. Unlike these studies, a study by
Spielauer (2004) finds that there exists considerable intergenerational persistence of
educational careers contrasting the strong educational expansion over the past decades.
A more recent study by Fesser, et. al (2011) using the Austrian Household Survey on
Housing Wealth have revealed internationally comparable persistence measures of
educational attainment available for Austria. The study using Markovian as well as
univariate and multivariate econometric approaches reveal significant persistence for
intergenerational transmission of educational attainment in Austria for all cohorts born
between 1916 and 1984 but the persistence has decreased over time, implying that
mobility increased. However, Austria is among those countries with the highest
persistence that is lowest mobility. Just Italy and Slovenia exhibit higher levels of
standardized persistence. Therefore, the study concluded that there is a positive and
significant correlation between educational attainment of parents and descendants. The
level of correlation seems to be higher than northern European countries as Netherlands,
Finland or Sweden and closer to southern European countries like Italy or Slovenia.
The dependence of educational outcome of descendants on the education of parents is
decreasing over time. As per the study, educational expansion in Austria has failed to
increase educational mobility between generations.

In case of France, the fast expansion of education system may be thought as increase
in access to education institutes by students irrespective of their social origins but, this
‘democratization’ of education is questionable and is a matter of debate in French
society. A study by Chevalier et.al (2009) suggested that in France if aggregate of three
qualification levels, elementary, secondary and tertiary are taken then the most skewed
distribution of all nations could be found as a result of the very strong selectively
institutionalized French school system. Müller (1991) shows that the differences in
educational distributions are quite marked among the nine CASMIN nations including

55
France. Only two aspects were common to the countries studied which are; i) more
people take advantage of elementary qualification than any other qualification and ii)
all countries have a similar proportion of graduates with the highest level of
qualifications. Using PISA data from the 2000 and 2006 waves a study by Oppedisano
& Turati (2015) showed that educational inequalities are increasing in France. The
study highlights that besides parental background, schools’ characteristics are also
important determinants of inequalities in achievements among students. Comi (2004)
compares earnings and educational mobility in Europe, using the data from the young
sample of the ECHP. The author found lower intergenerational mobility Southern
European countries, France and Ireland, while Nordic countries, the Netherlands and
Austria depicted higher intergenerational mobility and an intermediate position for
Belgium and Germany.

The German education system experienced numerous reforms over the last few decades
for improving "equality of opportunity", i.e. to assure all students equal access to
education. However, during the same time comparative evidence proves that Germany
has low intergenerational educational mobility. Couch and Dunn (1997) compared
intergenerational education correlation between US and Germany and found higher
education mobility in Germany than in the US. Lauer (2003) compares German and
French cohorts born between 1929 and 1968 and finds the two countries to be
surprisingly similar. Study by Dustmann (2004) dealt with correlations between
parental characteristics and child schooling and earnings for birth cohorts between 1920
and 1966 and concluded that parental background affects child outcomes. Also, from
the sociological perspective, the question of educational mobility and intergenerational
status transmission was always there. Blossfeld (1993) examined the birth cohorts
between 1916 and 1965 and found that there is no alteration in the impact of parental
background over time. However, Müller and Haun (1994) analyzed educational
outcomes and transitions for the birth cohorts between 1910 and 1969 and concluded
opposite results. According to them, the significance of parental social class for child
educational outcomes declined over time and their findings were corroborated by Henz
and Maas (1995).

Another study by Riphahn & Trubswetter (2013) study the development of educational
mobility after the fall of the iron curtain in East Germany. The study compared the
importance of parental educational on secondary schooling outcomes of children in East

56
and West Germany. In order to find the probability of educational attainments of child
conditional on the educational attainments of parent the authors used transition
matrices. The transition matrix for East and West Germany across all survey years
confirms strong intergenerational educational correlation in East and West Germany.
In case of West Germany, probability for a 17 year old to attend Advanced School
increases by 200 per cent even if one parent holds an Advanced School rather than a
basic school degree while in East Germany the probability increases to 243 per cent.
Through using multivariate Probit regressions the study revealed that correlations are
significantly different for East and West Germany. In East Germany, children of parents
with only middle school education are at disadvantaged position compared to West
Germany. Surprisingly, men differ more strongly from women in the East and urban
residence seems to be less helpful for educational enrolment in the East than in the
West. Children with one sibling have the highest probability of attending Advanced
School in both the countries with somewhat higher hindrances for children from larger
families in East Germany than in the West.

The Italian school system can be characterized as a prevalently centralized and public
system financed by the government through taxation, which provides the same quality
of education to everybody. Given this characterization, an Italian family at a low level
of income should have the same level of education available as a higher income family.
In fact, Italy is characterized by less income inequality. But a study by Chechhi, et.al
(1999) which looks jointly at the issues of income distribution and intergenerational
mobility in Italy found that standard measures of intergenerational mobility between
occupations and between education levels indicate that poor and non-educated families
are less likely to invest in the education of their children and to move up along the
occupational ladder.

With the help of transition matrices the authors found that the probability that the son
of a graduate is a graduate is higher than in the US (65.1 per cent vs. 61.0 per cent);
vice versa the probability that the son of a non-graduate reaches a college degree is
substantially lower in Italy than in the US (7.1 per cent vs. 20.8 per cent). These
transition probabilities suggest that chances of obtaining a college degree are less in
Italy than in the US, even if Italy experiences a more substantial increase of the
proportion of college graduates from one generation to the other. The odds ratios for
the two transition matrices, show that the odds of obtaining a college degree in Italy are

57
almost 25 times higher if the father has a college degree, while in the US the odds
increases by only 6 times if a student has a graduate father. The chances of attaining a
higher educational category are more unequally distributed in Italy than in the US.
Hence the authors conclude that Italy has low intergenerational mobility, given the
strong association existing between the socio-economic outcomes of parents and their
children as adults.

Many other studies conform the findings of this study e.g. using matrices of transitions
between social classes defined according to the prestige of occupations, Erickson &
Goldthorpe (1992) find that Italy displays less mobility. In a study of educational
attainment across cohorts, Shavit & Blossfeld (1993) find a decline of the impact of
fathers’ occupational status on sons’ educational achievements in the US, whereas the
opposite trend is observed in Italy.

3.6.3 Studies Related to Latin America

Inequality in income in Latin America is very high as compared to other developing as
well as high income countries (OECD, 2008). In principle, it could be contended that
this type of static income inequality across individuals is not bad, as the dispersion in
earnings could act as a strong incentive for parents to invest in their children’s human
capital. However, for poor households to be able to grasp these opportunities, they
should have access to well-functioning credit markets. With this background a study by
Daude (2011) analyses the extent of intergenerational transmission of educational
achievements in 18 Latin American economies. The paper used extensively the data
from test scores and socio-economic and cultural background available through the
OECD’s Programme for International Student Assessment (PISA) study. It allows
analyzing the impact on cognitive skills and the quality of education received by
students rather than just the quantity. Before this Behrman et al. (2001) use an
estimation approach for four countries in the region by using alternative estimates,
based on the performance of children still in the education system.

The baseline regression results reveal that parental education has a statistically
significant impact for all variations considered. Estimated regression coefficient
revealed that, an additional year of parental education increases on average child’s
education by 0.65 years and the results are same for female and male children. The

58
correlation coefficient between parental and child’s educational attainment is around
0.6. Interestingly, this average correlation coefficient for the 18 countries in the sample
is very much in line with evidence provided by Hertz et al. (2007) based on household
surveys for 7 countries in the region. The estimates for both measures of
intergenerational persistence by country show that there is considerable variation in the
region in both measures. With the help of Social Mobility Index derived for 18 Latin
American countries it has been found that in 11 out of the 16 countries,
intergenerational mobility has increased, while the same has declined significantly only
in Colombia and Uruguay.

The author shows that intergenerational persistence in education outcomes is

significantly associated with static income inequality as measured by the Gini
coefficient. Less mobile societies also tend to exhibit high levels of inequality.
However, Costa Rica and Honduras were outliers, wherein social mobility was found
to be much higher than expected given the distribution of income. The author explained
that limited access to credit or savings for disadvantaged and middle-sector households
can also be a significant hurdle to investment in human capital. In fact, in Latin America
access to credit is limited to such an extent that it is possibly holding children back from
pursuing further studies. Further in Latin America there exists social exclusion and
discrimination in terms of rich students going to private school of good quality while
the poor students are missing this opportunity. Therefore, the current education system
in the region endorses selection for only those who can afford it. By itself selection
tends to depress overall educational outcomes and the region’s private schools’
compound this by failing to make the most of their privileged intake.

3.6.4 Studies Related to Scandinavian Countries

In principle one would anticipate the association between social origin and educational
outcomes to be significantly weaker in the Scandinavian countries than in other
Western, industrialized countries because here, welfare states and public social security,
redistribution of incomes, and educational policies were established with the aim of
providing a high level of social security and equal opportunity with respect to
educational choices (Kautto et al., 2001; Erikson & Hansen, 1987). However, the
empirical literature tells a different story. While the Scandinavian countries display

59
comparatively high levels of absolute educational mobility, the relative inequalities in
educational opportunity have shown relatively little change over time. Thus, studies on
the effect of social class background on educational attainment in Sweden (Breen &
Jonsson, 2000; Dryler, 1998; Erikson & Jonsson, 1996); Norway (Lindbekk, 1998;
Hansen, 1997), Finland (Kivinen & Rinne, 1996), and Denmark (Jæger & Holm, 2006;
Davies et al., 2002) all confirm the perseverance of strong educational inequalities in
the Scandinavian countries. Holzer (2006) studies the evolution of the association
between college attendance and parental income over different age cohorts in Sweden,
pointing out that new opening of local colleges has not improved the degree of
intergenerational mobility. In the Scandinavian context of low-income inequality,
comprehensive public social security, and where the direct costs of education are borne
almost exclusively by the state, it is not theoretically plausible that parents’ economic
resources and occupational status, as implied by social class are the main generators of
educational inequalities.

For example, a study by Jaegar & Holm 2006 deals with non-monetary social
background resources in the Scandinavian context. Furthermore, the authors
“disaggregate” social class and investigated the impact of four types of parental
resources: economic, cultural, social, and cognitive “capital”, on children’s educational
outcomes. Using an extremely rich Danish longitudinal data on the educational careers
of cohorts born in the late 1970s, the study developed and tested several hypotheses
relating to the impact of each of the four types of parental resources on children’s
educational attainment within a country. In Denmark, public social expenditure on
social security ranges around 30 per cent of GDP and very high compared to the rest of
the OECD area, and levels of income inequality and poverty are very low compared to
other OECD countries (Burniaux, et al., 1998). Cash benefits and social services are
comprehensive and generous by international standards, and access is based on
universalism and social rights rather than income and means testing (Hansen, 2002;
Kautto, et al., 2001).

Out of the four types of capital the authors found that economic capital of parents has
less effect on educational attainments of children in Denmark. Since education and
education grants are funded almost exclusively by the state in Denmark and the other
Scandinavian countries, the parents would find it hard to make any direct monetary
investments in order to promote their children’s educational chances. The same is true

60
for Sweden and Norway where studies have identified a weak effect of parental income
(Hansen, 1997; Erikson & Jonsson, 1996). The insignificant effect of parental income
persists for other levels of education also. Hence there are good reasons to expect
parents’ non-monetary resources to be of comparatively greater significance in
explaining the persistent intergenerational inequality in educational attainment in
Denmark. This is because non-monetary parental resources like cultural, social, and
cognitive capital are arguably particularly powerful background assets in Denmark and
in the Scandinavian mobility regime compared to elsewhere. The study applying Latent
Class Multinomial Logit Model on the data set from Youth Longitudinal Survey found
that among fathers there are no significant effects of economic capital on children’s
educational outcomes. On the other hand, among mothers the result is less clear since
the authors found that economic capital to be a significant predictor of acquiring
vocational education. With respect to vocational education, the findings then suggest
that the child has a higher probability of obtaining this type of education over no
education beyond elementary school if mothers possess high economic capital. This
result indicates that among mothers, economic resources are related to children’s
educational attainment. Further parents’ cultural capital is a highly important predictor
of children’s educational attainment in Denmark, and especially for mothers. Among
mothers, cultural capital is significantly related to obtaining all types of post-elementary
education, while for fathers this trend is only significant with respect to university-level
education. The reason for stronger effects of cultural capital for mothers than fathers
could be that the transmission of mothers’ cultural capital is stronger since she is likely
to spend more time with the child, and hence her level of cultural capital has a stronger
influence on the initial cognitive and aspiration development of the child. Social capital
is not significant in any of the models when the other forms of capital are controlled.
Finally, with respect to cognitive capital, the study found that mothers’ cognitive capital
has a negative impact on the probability of children choosing particularly vocational
education. Furthermore, for father’s high cognitive capital is negatively associated with
children achieving upper secondary education.

Overall the paper by decomposing social class effect into four types of social origin
resources: economic, cultural, social and cognitive “capitals” have argued that the
traditional focus on social class is misleading in Scandinavian context. Because of
comprehensive public social security, redistribution of incomes, and egalitarian

61
educational policies, it is not evident that stratification based on income and
occupational prestige, as implied by the concept of social class, are the major sources
of inequality preservation in this Scandinavian “mobility regime”. The authors’ argued
that in order to explain the persistent intergenerational inequalities in the Scandinavian
countries, one needs to pay explicit attention to non-monetary aspects of social origins
that affect educational attainment.

3.6.5 Studies Related to Australia

A study by Chesters (2010) drawing on Maximally Maintained Inequality theory and
Relative Risk Aversion theory, used logistic regressions to analyse data collected by
three nationally representative, cross-sectional surveys conducted between 1987 and
2005 (N= 4463) to study the relationship between parental and child’s education. The
result shows that women were less likely to have a university degree than men. In each
year respondents with a university-educated father were more likely to have a university
degree themselves. The effect of father’s education lessened somewhat over time with
the odds ratio declining from 5.9 in 1987 to 2.7 in 1994 and 2005. This suggests that
although there continued to be a statistically significant association between father’s
education and respondent’s education, the magnitude of this association declined over
time. Mother’s education is not significant in 1987 or 1994 but is highly significant in
2005 when having a university-educated mother increased the respondent’s odds of
having a university degree by 2.4, net of other factors.

Other studies also found that a disproportionately low proportion of students from low
socioeconomic backgrounds participate in higher education (Bradley et al. 2008;
Chapman & Ryan, 2003). In other words, ‘the privileged classes manage to maintain
their advantage over time’ (Arum, et al., 2007, p. 5). Marks (2009a) analysed the first
six waves of the Household, Income and Labour Dynamics in Australia (HILDA) data
concluding that students from higher socioeconomic backgrounds were twice as likely
as those from low socioeconomic backgrounds to hold a university qualification. His
analysis showed that although respondents in the younger cohort were more likely to
have a university degree due the expansion of the higher education sector, the effect of
socioeconomic background had not diminished.

62
James, et. al. (2008) also found that students from the lowest socioeconomic quartile
continued to be under-represented in the higher education sector accounting for around
just 15 per cent of university students each year during the period 1989 to 2006.
Chapman and Ryan (2003) examined data from the Youth in Transition Survey and the
Longitudinal Surveys of Australian Youth found that although there had been little
change in the likelihood of males from low wealth backgrounds attending university,
‘females across the entire socioeconomic distribution’ were more likely to attend
university in 1999 than in 1988. Using various inequality measures, Chapman and Ryan
(2003) found that university participation was more unequally distributed in 1993 than
in 1988 but that by 1999 the distribution recovered and was like that of the pre-HECS
era.

3.6.6 Studies Related to Asian Nations

During the last six decades China has made great achievements in raising national
average standards of educational quality and attainment. These achievements have,
however, been distributed unevenly across the country. According to a study, by late
2000s, although more than 90 per cent students of primary school in rural China went
to junior high, but less than 20 per cent transitioned to high school and the transition
rate to tertiary level of education was only 1.3 per cent. As a result, children born in
Beijing, Shanghai or Tianjin were 35 times more likely to attend college than children
born in rural areas (REAP Brief No-107). While labour market reforms have raised the
returns to education across China (Liu, et al., 2010; Zhang, et al., 2002), the
incompleteness of these reforms in particular, ongoing constraints to permanent rural-
to-urban migration (Golley & Meng, 2011; Cai & Wang, 2010) and segmentation in
urban labour markets (Chen, et al., 2007; Appleton, et al., 2004) means that the
incentives for pursuing higher educational outcomes in pursuit of higher wage returns
are uneven across urban, rural and migrant populations. The combination of uneven
development in education and urban biased labour market reforms has undoubtedly
contributed to the gap between urban and rural incomes in China, which is large and
rising over time (Sicular, et al., 2005).

Most of the literature relating to intergenerational patterns of educational attainment in

China has been confined to the rural or urban populations and much of it has focused

63
on the impact of the Cultural Revolution (1966–76) on different members of society at
the time or on subsequent generations. Sato and Li’s (2008) study on the impact of
family origin on educational attainment in rural China basically repeats those of Meng
and Gregory (2002) and Zhou et al. (1998), by showing that class-based discrimination
during the Cultural Revolution did not last long enough to have a permanent effect on
educational attainment across generations instead, those higher up the social strata have
continued to ‘out-educate’ those below them. A recent string of papers indicates that
shortcomings in China’s rural education system from early childhood onwards
seriously impinge on the educational opportunities for rural and migrant children,
which in turn impacts on their ultimate levels of educational attainment (Luo, et al.,
2012; Wang, et al., 2010; Teng, 2005). These papers emphasizes on the low incomes
of poor rural households relative to rising educational costs and the poor quality of rural
education as the major impediments to upward mobility of the poorest members of
Chinese society. A more recent study by Golley & Kong (2012) have found that urban
children are more likely to attend higher levels of education than their rural and migrant
counterparts. Dividing the data into 10 equal 4-year birth cohorts (starting from 1941-
45 to 1986-90) the authors found that in the ninth cohort, 93 per cent of urban children
attained either high school or college degrees, compared with 40 per cent of migrants
and 43 per cent of rural children. As expected, the greatest divergence is at the college
level, with urban children in this cohort more than six and five times more likely to
attend college than their migrant and rural counterparts. Another point worth noting is
that a higher proportion of rural children in the three youngest cohorts attain college
degrees than their migrant counterparts, and yet a higher proportion is also more likely
to complete only primary education. Migrants, on the other hand, are more likely to
have high school degrees, a point that is true across all cohorts apart from the fourth
one. While there are clear differences in the educational attainments of China’s migrant
and rural populations, these are dwarfed by the urban–non-urban divide. These findings
combine to depict a possible future of rising intra-rural, rural–urban and intra-urban
inequalities in educational opportunities and outcomes, in which a positive and possibly
strengthening relationship between the educational achievements of parents and their
children will be a prominent feature.

In case of Indonesia, a study by Mare & Maralani (2006) proposed alternative models
that provide improved estimates of intergenerational effects. To elucidate these models,

64
the study focused on the effects of women’s educational attainment on the education of
the subsequent generation in Indonesia because Indonesian women have historically
obtained relatively low levels of schooling. Indonesia has undergone huge demographic
changes during the past 30 years, including massive declines in fertility and mortality
rates and substantial rural to urban migration. During this period, the average
educational attainment levels and sex differentials have changed dramatically as well.
For example, among men born in 1930-34, 27 per cent had no formal schooling and 92
per cent had no more than primary school. For women in this cohort 56 per cent had
no formal schooling at all and 97 per cent had no more than primary school. For men
born in 1960-64 in contrast, only 5 per cent had no formal schooling and 36 per cent
had more than primary schooling. For women, these corresponding percentages were
10 per cent and 23 per cent respectively (Cobbe & Boediono, 1993). More recent
cohorts show still higher levels of educational attainment and smaller differences
between men and women.

The study shows how to estimate the effects of changes in women’s educational
attainments in the maternal generation and how these effects work through marriage,
fertility, and intergenerational transmission. This enables the researchers to go beyond
most other research, which relies on estimates of the effects of mothers’ schooling that
ignore the effects that accrue through changes in family formation and family size.
Fertility in Indonesia varies by women’s educational attainment, although it does not
follow a simple inverse relationship. Among Indonesian women in the 1970s, fertility
was highest for the women with a primary education, lowest for the small proportion
of women with postsecondary schooling, and at an intermediate level for the women
with no schooling or secondary schooling (Hirschman & Guest, 1990). This non-
monotonic pattern has persisted in more recent years.

The study focused on how a population of women with varying amounts of schooling
produces a generation of offspring who also varies in their educational attainments.
Using Indonesia Family Life Survey (IFLS) the study used the statistical models for
marriage, fertility, and offspring’s educational attainment by maximum likelihood,
applied to each equation separately

The results reveal that educational attainments of both mothers and fathers have strong
positive effects of approximately equal size on the attainments of their children.

65
Moreover, the effects are relatively constant across parents’ educational categories. On
the average, the odds that children are in a higher rather than a lower category of
schooling approximately double for each higher category of their mother’s or father’s
schooling. The predicted probabilities of sons’ and daughters’ schooling by levels of
mother’s schooling, with father’s schooling held constant, increase substantially with
each successive level of mother’s attainment. The estimates of the effects of parents’
schooling on the number of children ever born follow the non-monotonic pattern of
differential fertility found in other research on Indonesia. Expected fertility is constant
for the first three education categories but decreases substantially for the two highest
education categories. With husband’s education held constant at the elementary level,
women’s expected number of children is approximately 5.0 for women with no
schooling, primary education, or junior secondary education, and declines sharply to
3.7 for women with postsecondary education. The effect of husband’s education on
fertility is much smaller than the effect of wife’s education. With women’s educational
attainment held constant, the estimates show that husbands with primary, secondary, or
postsecondary schooling all have approximately 5.3 children, whereas men at the
bottom of the education distribution average somewhat fewer children.

Indonesian couples show extremely strong evidence of positive assortative mating. The
coefficients show that the odds of a woman marrying into the next highest husband’s
educational category are more than seven times greater both for women with elementary
schooling compared to those with no schooling and for women with junior secondary
schooling compared to those with elementary schooling. The odds of a woman
marrying into the next highest husband’s education category are about four times
greater for women with senior secondary schooling than for those with junior secondary
schooling and for women with postsecondary schooling than for those with senior
secondary schooling. The authors combined these factors in order to find the effects of
these factors on intergenerational transmission of education in a combined way. The
patterns show that marriage, fertility and intergenerational transmission combine to
affect the next generation in a complex way. At the upper end of the women’s education
distribution, increases in attainment bode well for the next generation because more of
these women will marry highly educated husbands and these improvements in both
mother’s and father’s education will benefit their children. These effects in the next
generation, however, are offset by the reduced fertility of highly educated women. The

66
net impact of the change in women’s education is positive, but not nearly as great as an
analysis of intergenerational transmission alone would imply.

3.7 An Account of Indian Studies on Intergenerational Educational Mobility

In Indian context one of the earliest works has been done by Driver (1962) who by
using data of one percent of male households in Nagpur District showed that
intergenerational educational mobility is recurrent among rural and urban castes. This
implies that mobility has inconsequential effect on the traditional relation between
positions in the caste and occupational hierarchies. As per the author, this relation was
due to the differences in educational attainment among castes. Kumar et al. (2002a,
2002b) by analyzing National Election Study data of 1971 and 1996 from 80 and 108
parliamentary constituencies respectively, across India, concluded that though there has
been a net upward movement of occupational classes across generations, substantial
class inequalities also persist alongside. As per them such inequalities are primarily
caused by differences in the financial, educational and social resources possessed by
different classes and should not be ascribed to caste alone.

Jalan and Murgai (2008) investigated the educational mobility among the age group 15-
19 using 1992-93 and 1998-99 National Family Health Survey (NFHS) data. The
analysis was conducted on 5-year cohorts, for individuals born between 1969-73 to
those born between 1979-83, to examine how the role of parental education may have
changed over time. As per the study, “mobility has increased across cohorts, with gains
occurring mainly in the mid-80s when the youngest cohort was entering school. There
was little change in mobility between the two older cohorts. This is true in both urban
and rural areas, and with respect to both father’s and mother’s education. For urban men
born in the early 1970s, a one-year difference in the schooling of their parents resulted
in a difference of approximately 0.22-0.24 years in their own schooling. For the younger
cohort that was entering school in the mid-80s, the same difference in parental
education results in a 0.09-0.15 year difference in schooling. Mobility has risen among
rural men as well, with the coefficient on father’s education declining by one-third from
0.30 to 0.20, and the coefficient on mother’s education declining from 0.19 to 0.12
across the three cohorts”. The study also found that after controlling for other attributes
the education gaps between backward and forward castes are not that large.

67
A study by Maitra and Sharma (2009) make use of the India Human Development
Survey, 2005 (IHDS) in order to explain change in educational attainments of adult
male (20 and above) who are born in 1985 or before, across birth cohorts. Specifically,
the study examined the role of parental education on two aspects of child’s educational
attainment i) years of schooling attained and ii) progression across different schooling
levels. As per the authors that there has been a significant increase in educational
attainment of individuals over the last 70 years, with women gaining the most in terms
of increases in educational attainment. Restricting the analysis to adults (those more
than 20 years old at the time of the survey), and accounting for the potential endogeneity
of parental educational attainment, the study revealed that the total effect of parental
education on years of schooling of their children is not statistically significant indicating
an increase in intergenerational educational mobility. The analysis of school
progression was conducted using a sample of 15 − 24 year olds at the time of the survey
and using sequential probit model. It showed that father’s educational attainment has a
positive and statistically significant effect on the probability of continuing to
postsecondary school/college. Private investment in education which is dependent
primarily on father’s education and income, remains to be critical for children in order
to benefit from various educational opportunities.

Besides this a study by Ray & Majumder (2010) on the extent of intergenerational
mobility in both educational and occupational attainments for diverse ethnic groups in
India revealed strong intergenerational immobility in both educational achievement and
occupational distribution among the Scheduled Castes (SCs) and Scheduled Tribes
(STs) who have been discriminated against historically. Using five rounds of National
Sample Survey (NSS) and covering the period 1983-2005, Hnatkovskay, et.al (2013)
investigated intergenerational mobility in educational attainment, occupation and
wages. Their results indicate that the gains during the past two decades have not been
restricted to limited sections of SC/STs. Intergenerational education mobility has been
experienced by both low and relatively highly educated SC/ST households while at the
same time intergenerational income mobility increased for both low and high-income
households amongst SC/STs. Additionally, increase in mobility for SC/STs is faster
than non-SC/STs, on an average Indeed, it has now become far more likely that the son
of a poor illiterate SC/ST cobbler would become a machine worker with middle or

68
secondary school education having a much higher rank in his generation’s income
distribution than his father did in his generation.

Azam & Bhatt (2015) used IHDS Survey of 2005 and found significant improvements
in intergenerational educational mobility in India, at the aggregate level, across social
groups and across states. Although, intergenerational educational mobility in majority
of the Indian states have made noteworthy improvement over time, still, there remains
significant disparity across states. Another study by Emran and Shilpi (2015) used
1992/93 and 2006 rounds of the National Family Health Survey (NFHS) in India to
understand educational mobility in post-reform India. It examined the trends and
patterns of intergenerational mobility in education among new entrants in the labor
force (16-27 year olds) between 1992/93 which is a year immediately following the
economic liberalization and 2006, nearly 15 years after liberalization. The authors
focused on the role of family background in educational attainment of the youth (16 to
27 year olds at the time of the survey) who constituted the bulk of the new entrants into
the labor market. Using intergenerational and sibling correlations, the study found that
educational mobility remained effectively unchanged for a large proportion of Indian
youth after a decade and a half of high economic growth. Asher, et. al. (2018)
documented the trends in educational mobility from the 1950s to the 1980s birth cohorts
and focusing exclusively on marginalized groups found that upward mobility gains in
India over recent decades have accrued to Scheduled Castes and Tribes and Muslims
have lowest intergenerational mobility. Further, none of those gains have come at the
expense of higher caste groups.

3.8 Gaps Identified from Previous Research Studies

From the literature review it can be concluded that many international studies have been
done on the topic of intergenerational mobility in education. A variety of methods and
measures of intergenerational mobility has also been used to find the degree of mobility
across generations. In case of India also, almost all the studies have agreed that over the
years, effect of parental education has reduced indicating that societies have become
more mobile. Though, there are gaps and gradients in mobility in terms of marginalized
groups. The studies attributed various reasons to the process ranging from structural
changes because of liberalization to success of positive discrimination policies.

69
However, limited empirical studies and research have been done in India using
nationalized sources of data to find out the determinants of intergenerational
educational mobility. The determinants would help to quantify the size of the influence
of family background variables on intergenerational educational mobility of an
individual. Knowledge of the magnitude of determinants may inform the policy makers
about the extent to which the various policies have helped towards upward educational
mobility by removing the influence of family background variables. Further, very few
studies have shown together educational inequality and educational mobility across
generations and the relationship between the two. As already mentioned, knowledge
about this relationship is important as theories of child development and economic
mobility advocated towards combining inequality and mobility as a starting point for
understanding the causal process and its policy implications (Corak, 2013). Finally, the
study has dealt with intergenerational mobility in education based on location, gender
and social group of an individual, all at the same time, which also very few studies have
done previously.

70