SERIES

IZA DP No. 10047

PAPER

Intergenerational Mobility in Income and

Economic Status in Ethiopia

Getinet Astatike Haile

DISCUSSION

July 2016

Forschungsinstitut
zur Zukunft der Arbeit
Institute for the Study
of Labor
 Intergenerational Mobility in Income
and Economic Status in Ethiopia

Getinet Astatike Haile

University Nottingham
and IZA

Discussion Paper No. 10047

July 2016

IZA

P.O. Box 7240

53072 Bonn
Germany

Phone: +49-228-3894-0
Fax: +49-228-3894-180
E-mail: iza@iza.org

Any opinions expressed here are those of the author(s) and not those of IZA. Research published in
this series may include views on policy, but the institute itself takes no institutional policy positions.
The IZA research network is committed to the IZA Guiding Principles of Research Integrity.

The Institute for the Study of Labor (IZA) in Bonn is a local and virtual international research center
and a place of communication between science, politics and business. IZA is an independent nonprofit
organization supported by Deutsche Post Foundation. The center is associated with the University of
Bonn and offers a stimulating research environment through its international network, workshops and
conferences, data service, project support, research visits and doctoral program. IZA engages in (i)
original and internationally competitive research in all fields of labor economics, (ii) development of
policy concepts, and (iii) dissemination of research results and concepts to the interested public.

IZA Discussion Papers often represent preliminary work and are circulated to encourage discussion.
Citation of such a paper should account for its provisional character. A revised version may be
available directly from the author.
IZA Discussion Paper No. 10047
July 2016

ABSTRACT

Intergenerational Mobility in Income and Economic Status

in Ethiopia*

Using data from two comprehensive national labour force surveys conducted in 2005 and
2013, this paper examines the extent of intergenerational mobility in Ethiopia using monetary
and non-monetary measures. Quantile regression and OLS based results suggest there is
moderate level of “stickiness” in income mobility across generations. Sons are found to be
more mobile than daughters both in monetary and non-monetary terms, although the mobility
gap appears to have narrowed recently. There is virtually no evidence on intergenerational
mobility in the context of low income countries in general and Sub-Saharan Africa in
particular. The paper thus provides valuable insights into issues of intergenerational mobility
in a low income country setting. The mixed approach used addresses possible measurement
error in income, as well as offering a broader scope in examining intergenerational mobility.

JEL Classification: J62, D31

Keywords: intergenerational mobility, income, education, occupation, Ethiopia

Corresponding author:

Getinet Astatike Haile

Division of Industrial Economics
Nottingham University Business School
South Building, Jubilee Campus
Wollaton Road
Nottingham, NG8 1BB
United Kingdom
E-mail: getinet.haile@nottingham.ac.uk

*
The author would like to thank the Central Statistical Agency (CSA) of Ethiopia for making the data
used in this paper available. The usual disclaimer applies.
1. Introduction

Intergenerational mobility is said to happen when children ‘occupy different positions in

their generation’s distribution of economic status than did their parents in their
generation’s distribution’ (Dearden et al. 1997, p.47). The question of how parental or
family background influences children’s lifetime economic status has been the subject of
1
considerable research particularly since the seminal work of Becker and Tomes (1986).
The central hypothesis of this literature lies on parental altruism towards the welfare of
their children generally and investments in their human capital, which shape the fortunes
of children beyond the heredity of cognitive ability and other genetic traits.

The level of intergenerational mobility in a society is thought to reflect the degree of

equality in economic opportunities in the society. Although inequality in income and
economic status is a common feature across countries, the rate of inequality in low-
income countries is reported to be much higher than that in high-income ones (UNDP
2013). Understanding the drivers of inequality in income and economic status is vital to
address issues of inequality. Examining the distributions of income and economic status
between successive generations plays an important role in this regard. This paper aims to
shed some light on intergenerational mobility in income and economic status in Ethiopia
and investigates the extent to which inequities in income and economic status are
transmitted across generations.

The bulk of the literature on intergenerational mobility relies on income or earnings to

measure mobility. This is justified on the ground that such a measure provides a simple
metric in examining mobility through the correlation between the income levels of two
successive generations. In cross-sectional data of the sort used in this paper, this is often
realised through a linear regression of children’s income on the income of their parents,
both log-transformed, to yield a measure of intergenerational mobility. An alternative to
using income or earnings to measure economic status is to use non-monetary measures,
such as occupational status or educational attainment (See, for example, Carmichael

1
Some of the recent studies include Black et al. (2015), Blandan and Macmillan (2014), Corak (2013), Black and
Devereux (2010), Aaronson and Mazumder (2008), Currie (2008), Jenkins and Siedler (2007), Ferreira and
Veloso (2006), Blandan (2005), Dustmann (2005), Nguyen et al. (2005).

2
2000; Johnson, 2002; Erikson and Goldthorpe 2002). The non-monetary approach may
offer a broader account of intergenerational mobility as it depicts mobility in both
economic and social status. Investigating intergenerational mobility on a range of
different measures, such as educational attainment and occupational status in addition to
income, may thus provide a more comprehensive picture of intergenerational
transmission in economic status. In fact, some, for example Goldberger (1989), warn that
exclusive focus on monetary measures such as income or earnings may ‘understate the
influence of family background on inequality’ (p. 513). On the other hand, focusing
entirely on non-monetary measures runs the risk of misclassification and hence of
obtaining biased estimates of intergenerational mobility.

Taking these issues into account, this paper combines the monetary and non-monetary
approaches to examine the level of intergenerational mobility in income and economic
status among a sample of young adults between 25 and 35 years of age and their parents
in Ethiopia. To this end, the paper uses data from two comprehensive and nationally
representative labour force surveys conducted in 2005 and 2013. It also examines if
systematic differences in mobility exists between sons and daughters. Due to the
challenges posed by the labour market histories of women, which are often interrupted
for family and child care reasons, most previous studies have focused almost exclusively
on the intergenerational linkages between fathers and sons.2 This paper follows the same
approach and relies on fathers’ income; but it examines intergenerational mobility
between both fathers and sons and fathers and daughters.

Although there has been extensive research on intergenerational mobility in economic

status, the focus of the literature has almost exclusively been on advanced economies.3 In
the context of Sub-Saharan Africa, there is in fact virtually no evidence on
intergenerational mobility to date. This paper thus has the potential to provide important
insights into intergenerational mobility generally and potential differences between sons

2
Hotchkiss and Pitts (2007), Phipps et al. (2001), Blau and Kahn (2000) dwell on such interruptions rigorously.
In the low income country context considered here, the expectation is that interruptions of this nature are likely to
have far more serious repercussions given weak labour market institutions, which may entail longer or permanent
interruptions. Importantly, this compounds the measurement error problems for mothers’ income.
3
The few exceptions to this include Piraino (2014), Brunori et al. (2013) and Thomas (1996) on South Africa;
Gong et al. (2012) on China; Ferreira et al. (2011) on Turkey; Ferreira and Gignoux (2011) on Latin America;
Ferreira and Veloso (2006) and Dunn (2007) on Brazil, and Blinder and Woodruff (2002) on Mexico.

3
and daughters in the context of a low income country, which may be of considerable
interest for researchers and policy makers alike. The paper is organized as follows.
Section 2 describes the LFS data used, providing some background information on
income levels as well as the educational and occupational statuses of parents and adult
children in the Ethiopian dataset. Section 3 previews the empirical framework adopted to
estimate the intergenerational mobility in income, educational attainment and
occupational status. Section 4 discusses the results of the statistical analysis and section 5
concludes the paper.

2. Data and variables

The data used in this paper come from the 2005 and 2013 Labour Force Surveys of
Ethiopia. The surveys constitute the most recent two sweeps of the three nationally
representative and comprehensive household surveys collected by the Central Statistical
Agency (CSA) of Ethiopia, which began in 1999.4 The surveys gather extensive labour
market information with the aim of monitoring the economic and social situation of the
economically active population in the country. As detailed in Kolev and Robles (2010)
the surveys gathered extensive individual information on demographic characteristics
such as age and gender; household composition and individual respondents relationship
to the head and spouse of the household; educational qualification; employment status;
job characteristics including earnings, among others. This paper uses data on young
adults (children) between 25 and 35 years of age, inclusive, and their fathers to examine
intergenerational mobility.

The 2005 and 2013 LFSs monitored 230,680 and 240,660 individuals in total,
respectively, nationally. Of these, 114,827 and 120,709 individuals were children of
either both or one of the household head and their spouse. The important parts of the data
setup involved: (i) identifying the father-child pair(s) in each household and (ii)
generating new variables that record the relevant data on fathers’ education, employment
and earnings across all children between 25 and 35 years of age within each household

4
http://www.csa.gov.et/. The 1999 LFS did not gather data on earnings however, which is the reason for
excluding data from the survey.

4
monitored by the two LFSs. In both waves, fathers were identified based on whether a
respondent was male and had a head or spouse status within the household. Once the
relevant information on fathers has been copied across to each young adult in each of the
households surveyed, only the young adults have been retained for the analyses
conducted. This has yielded 5,493 and 7,759 young adults in 2005 and 2013 respectively.
All retained young adults have information on their and their fathers’ educational and
occupational status and thus form the basis for the empirical analysis examining
intergenerational mobility in educational and occupational mobility. The top panel of
Table A1 in the Appendix provides descriptive statistics on demographic characteristics,
educational qualification, occupation status, household and parental size as well as
geographic region of the retained sample of young adults, disaggregated by survey year.
The bottom panel of Table A1 provides descriptive statistics on the earnings information
of young adult children and their father, which has been symmetrically winsorized at
2%.5The earnings data come from individual responses to the question ‘what was the
total amount paid in your main occupation during the last month?’ The Table shows a
much reduced sample size of father-child pairs with valid information on earnings, which
is caused largely by missing data on fathers’ earnings.

3. Framework of analysis

The paper adopts two main empirical strategies to examine intergenerational mobility in
income and economic status.

3.1 Intergenerational Income Mobility

The traditional framework for measuring intergenerational income mobility is based on a

regression-to-the-mean model, involving OLS regression of the log of children’s income
on the log of their parent’s (often father’s) income.6 However, parental “permanent”

5
Symmetric winsorization has been chosen instead of trimming to minimise loss in observations on father-child
pairs with valid earnings information. See Lien and Balakrishnan (2005) for details on trimming versus
winsorization.
6
The typical model estimated takes the form yic    yip   i or yi,t    yi,t 1   i where y stands for
‘permanent’ income and superscripts (subscripts) c and p (t and t-1) represent child(ren) and their parent(s),
respectively, in household i.

5
income is often unobserved. Instead, researchers rely on some transitory income in one or
several periods. This has potential measurement error problem, which biases the
estimated income elasticity downwards (see for e.g. Zimmerman, 1992; Dearden et al,
1997; Naga, 2002, Black and Devereux 2011).7 Several approaches have been suggested
to overcome the measurement error problem, each with its own potential limitations.
Solon (1992) and Zimmerman (1992) suggest the use of “long-run” average parental
income8 while others (Behrman and Taubman 1990; Mulligan 1999) suggest a variant of
this approach where the “long-run” average of child’s income is regressed on the “long-
run” average of parental income. Naga (2002) argues that the latter approach is more
efficient than using only the “long-run” average of parental income. As noted by
Mazumder (2001), however, most of the applied work in the literature relies on short
time-series due to data limitations. This is likely to lead to biased estimates caused by
transitory income shocks, which may leads to high serial correlation in the income
variable.9 Solon (1992) makes another suggestion involving the use of instrumental
variables technique with parental education serving as an instrument. This approach may
yield an upward-inconsistent estimate; since the child’s long-term economic status or
“permanent” income is determined not only by parental income but also by their level of
education. However, he argues that the IV approach yields an upper bound of the true
intergenerational income mobility. Dearden et al. (1997) and Naga (2002) use predicted
parental income as a proxy for parental “permanent” income. They argue that although
“permanent” income may not be observed, a model of the determination of parents’
income is known to the researcher, which can be used to estimate parental “permanent”
income.

The measurement error problem alluded to in the preceding paragraph is likely to be

more of a concern in the context of a low income country such as Ethiopia for two main
reasons. First, time-series data on income are unavailable in such countries generally.
Even where available, however, they are bound to be much shorter than what have been
7
Income is often measured for a particular year, which is too short since the transitory variance of measured
income may lead to bias in  .
8
This is done by regressing child’s income on a time-series average of parents’ income. The inconsistency of this
estimator diminishes as the length of the time-series increases (Solon 1992).
9
Mazumder (2001) also recommends the use of income instead of earnings, as the former is likely to be a less
noisy measure of economic status than earnings.

6
used in the literature in the context of developed countries, thus ruling out the notion of
having long-run average or permanent income information. Secondly, it is safe to assume
that transitory income shocks are more of a concern in the context of low income
countries than in their developed counterparts, thus making the problem of high serial
correlation noted earlier even more of an issue

Taking such concerns into account, this paper implements the approach suggested by
Dearden et al. (1997) and Naga (2002); and uses predicted parental income as a proxy for
“permanent” income, thus estimating the following model:

(1) yic, j   c  yˆ jp   ic

where, y stands for actual reported income; ŷ is predicted income proxying ‘permanent’
income; superscripts c and p index children and their parent(s), respectively; i and j index
children and households.10 The estimated coefficient  yields the intergenerational
elasticity, while (1   ) gives a measure of the intergenerational mobility.11 The two
extreme values for the coefficient are:   0 , which signifies complete intergenerational
mobility (regression to the mean) with no correlation between children’s and parents’
income, and   1 , which suggests complete intergenerational immobility where, ruling
out  i , children’s income is completely determined by their parents’.

Notwithstanding the measurement issues discussed earlier, equation (1) does not handle
possible nonlinearities in the incomes of children and their parents. To address this, two
approaches have been implemented. First, transition matrix of the quantiles of children’s
and parental income are used to yield: (i) the proportion of all children with earnings
quantiles below the earnings quantiles of their parent (i.e., the sum of the proportions
below the main diagonal of the transition matrix), (ii) the proportion of all children with

10
In the context of developing countries in particular, there is a real possibility that there are more than one
children per household, which the empirical analysis carried out in this paper takes into account.
11
If yic, j and y p are measured in logarithms, the coefficient  corresponds to the elasticity of the child’s
j
income with respect to the parents’ income. In case of equal variances across generations,  represents the
intergenerational correlation coefficient. In case of differing variances, the correlation coefficient can be
estimated as    ( jp /  ic, j ) (Osterberg 2000, Bowles and Gintis 2002, Black and Devereux 2011).

7
earnings quantiles same as the earnings quantiles of their parent (i.e., the sum of
proportions on the main diagonal) and (iii) the proportion of all children with earnings
quantiles above the earnings quantiles of their parent (i.e. the sum of the proportions
above the main diagonal). These proportions and differences in them between sons and
daughters are used to examine intergenerational mobility in earnings. Secondly, a quantile
regression procedure, which estimates quantiles of children’s income conditional on
parental income, has also been implemented in this paper. Quantile regression rules out
the implicit assumption in equation (1) that the effect of parental income does not change

across the entire distribution of children’s income. Suppose that Q ( yic, j | yˆ jp ) denotes the

 th quantile of the income of a young adult in family j conditional on their parental

income. The  th quantile of the conditional distribution of yic, j given ŷ jp is then defined

as Q ( yic, j | yˆ jp )   ( ) yˆ jp ,   (0,1). Quantile regression in this case involves

minimization of the following objective function (Koenker and Bassett 1978):

(2) Min Q(  )    | yic, j   yˆ jp |   (1   ) | yic, j   yˆ jp |

 R k

i i: yic, j   yˆ jp  
i i: yic, j   yˆ jp 

where, as before, yic, j is the income of young adult(s) in family j, ŷ jp is parental income,

 represents the coefficients of interest,  is the particular quantile of interest.12 The

estimated  parameters may differ depending on the particular quantile being
estimated. This allows the analysis of: (i) differential effects of parental income on
children’s income across the income distribution of the young adults in general and (ii)
differential effects of parental income between different groups of young adults,
specifically between sons and daughters, at different quantiles of the conditional
distribution of children’s income.

12
Equation (2) shows that the minimization problem attaches asymmetric penalties of ( 1   ) and  for
overprediction and underprediction, respectively, and is solved using linear programming methods (Buchinsky,
1998).

8
3.2 Intergenerational Mobility in Economic Status

To examine intergenerational mobility in non-monetary economic status, the paper

follows Goldberger (1989) and Nguyen et al. (2005), which involves comparing the
occupational and educational status of children and that of their parents’. The non-
monetary indicators of economic status are likely to provide less noisy measures of long-
term economic status than income as they are less sensitive to transitory shocks (Nickell,
1982; Ermisch and Francesconi, 2002). Measurement errors are also much less of an
issue with educational and occupational status than earnings (Black and Devereux 2011).
Moreover, educational attainment and occupational status are highly correlated with
income (Nickell, 1982; Johnson, 2002). The nonmonetary measures of intergenerational
mobility are thus likely to provide suitable complementary measures to the income
measures of mobility discussed earlier.

The educational attainment rankings of children and their father used in the empirical
analysis are as follows: (i) no education or can’t read and write (y = 0); (ii) grades 1 to 6
(y = 1); (iii) grades 7 to 10 (y = 2), and (iv) grade 11 and higher (y = 3). On the other
hand, the occupational status rankings are: (i) no occupation (y = 0); (ii) elementary
occupations (y = 1); (iii) skilled agriculture and forestry (y = 2); (iv) services and sales (y
= 3); (v) machine operator and crafts (y = 4); (vi) managerial and professional (y = 5).13

These rankings are analysed in two ways. First, similar to the approach used in the
income mobility analysis, transition matrices involving the educational and occupational
status of children and their parents have been used to generate: (i) the proportion of sons
and daughters with an educational attainment (or occupational status) level that is lower
than that achieved by their parent; (ii) the proportion of sons and daughters with the same
educational attainment (or occupational status) as their parent’s, and (iii) the proportion
of sons and daughters with an educational attainment (or occupational status) higher than
that achieved by their parents. Once again, these proportions and differences between

13
Unlike the income measure, here the highest status of either the father or the mother has been taken as the
highest estatusIt is important to note, however, that the ordering of these categories does have some arbitrariness;
and one cannot completely rule out some overlaps/misclassifications as pointed out in Erikson and Goldthorpe
(2002).

9
sons and daughters allow examining intergenerational mobility in non-monetary
economic status.

Secondly, ordered probit model is used to estimate the influence of parental educational
or occupational status on the educational attainment or occupational status of their sons
and daughters. This model based analysis has the advantage of allowing other family-
related factors, such as family structure and the number of siblings, to be taken into
account in examining intergenerational mobility in non-monetary economic status; and
the role played by family background factors in this respect. There is some evidence
highlighting the importance of family background factors in the process of
intergenerational mobility. For example, a two-parent family may have more resources
and may, consequently, be in a better position to invest more in their children’s education
than a single parent family. Also, it may be essential that the child quality-quantity trade-
off (Becker 1991; Hanusheck 1992) is taken into account. Large family/sibling size may
lead to resources being spread more thinly within the family particularly where siblings,
or at least some of them, are not fully engaged economically. This is likely to adversely
impact the educational and occupational status of children in large families with many
siblings.

The ordered probit model used to examine intergenerational mobility in non-monetary

economic status has the following general format:

(3) yic, j   c  y jp  xi, j  zj    ic

where, this time y represents ordinally measured educational and occupational statuses
of young adults and their parents in household j and x ij is the vector of characteristics of

the young adult i in household j, and z j represents the vector of family characteristics.

4. Results and discussion

4.1 Intergenerational mobility in income

10
Table 1 reports the unconditional probabilities obtained from transition matrices
involving quantiles of children’s and their father’s earnings, which reveals that slightly
more than 40% of children being in lower income quantiles than their father’s whether
viewed in the combined 2005 and 2013 sample or separately by year. Only 19% of
children are found to be in higher income quantiles in the combined sample, with a 24%
and 17% split between 2005 and 2013, respectively. The Table also shows an 8
percentage point decline in the proportion of children found in higher income quantile
than their father between 2005 and 2013.

Table 1: Child-father Income Quantiles and their Distribution, LFS 2005 & LFS 2013 (%)
Sons & Daughters Son Daughter Diff.
(1) (2) (3) (2–3)
Combined 2005 & 2013 sample (N=1186)
% in lower income quantile than father 42.5 36.1 52.9 -16.8***
% in the same income quantile as father 38.4 40.0 35.7 4.3
% in higher income quantile than father 19.1 23.9 11.5 12.5***
Total 100 100 100 100

2005 sample (N=397)

% in lower income quantile than father 40.1 32.7 53.1 -20.5***
% in the same income quantile as father 35.5 38.6 30.1 8.5*
% in higher income quantile than father 24.4 28.7 16.8 12.0***
Total 100 100 100 100

2013 sample (N=789)

% in lower income quantile than father 43.7 37.9 52.7 -14.9***
% in the same income quantile as father 39.8 40.8 38.3 2.5
% in higher income quantile than father 16.5 21.3 9.0 12.3***
Total 100 100 100
*** p<0.01, ** p<0.05, * p<0.1
Authors own computation based on Ethiopian LFS 2005 and 2013.

The gender differentials in the unconditional probabilities reveal that compared with
sons, daughters are less (more) likely to be in higher (less) income quantiles than their
father, differences that are found to be statistically significant. Between 2005 and 2013,
the proportion of sons in lower income quantiles than their father has increased and that
of daughters remained stable, while a similar proportionate decline has been observed for
sons and daughters who were in the same income quantile as their father’s. On the other
hand, the proportion of sons and daughters who were in the same income quantile as their
father’s has increased between 2005 and 2013, more so for daughters.

11
Table 2 reports OLS and Quantile regression results on intergenerational income
mobility, which use predicted income proxying parental ‘permanent’ income.14 The first
column reports results from OLS while the remaining columns report results from
quantile regression.15 The estimated elasticities from OLS indicate that the correlation
between father’s and children’s income stood at 43% in 2005 and declined to 37% in
2013, thus suggesting increased income mobility over the period. The elasticities from
the quantile regression reveal that the conditional mean regression masks some variations
in mobility both within and across the two periods. The combined son and daughter
sample based estimates reveal that mobility increased across the quantiles in 2005, with
the exception of the 25th quantile. In 2013, on the other hand, mobility increased only up
to and including the median quantile. The increased income mobility in 2013 the OLS
estimates revealed therefore appear to be driven by the marked increase in mobility at the
median quantile. On the other hand, there is a decline in mobility particularly at the top
income quantile. Figure 1 below depicts plots of the estimated elasticities from quantile
regression for 2005 and 2013 for the combined sample.

Figure 1: Intergenerational Income Mobility of Sons & Daughters

0.55
Estimated Elasticities

0.35

0.15

-0.05
Q(.1) Q(.25) Q(.5) Q(.75) Q(.9)

-0.25
Income Quantiles

2005 2013 Diff

14
The estimation results predicting parental income are reported in Appendix Table A2. The parental income
equation controls for the age of the father (and age squared), household size, level of education, detailed
occupational and industrial categories as well as region; and the results are very much in line with theoretical
predictions and what one would expected from a similar equation.
15
It is worth pointing out that the income measures used are all in local currency and not adjusted for inflation
even though this should not affect the observed gender differential.

12
The contrast in the gender differential in income mobility is noteworthy. Results from
OLS indicate that in 2005 sons, with an estimated income elasticity of 39%, were
significantly more mobile than daughters whose income elasticity was found to be 56%.
However, OLS estimates suggest that daughters have gained some ground in narrowing
the gap in income mobility in 2013, when their income elasticity is found to be 44% (i.e.
a 12 percentage point decline) compared with sons whose income elasticity has declined
by 3 percentage points only.

Figure 2: Intergenerational Incomne Mobility by Gender

Sons

0.75
0.65
0.55
Estimated Elasticities

0.45
0.35
0.25
0.15
0.05
-0.05
Q(.1) Q(.25) Q(.5) Q(.75) Q(.9)
-0.15
-0.25

Daughters
0.75
0.65
0.55
Estimated Elasticities

0.45
0.35
0.25
0.15
0.05
-0.05
Q(.1) Q(.25) Q(.5) Q(.75) Q(.9)
-0.15
-0.25
Income Quantiles

2005 2013 Diff

13
The results from quantile regression for sons and daughters reveal that son’s income
mobility increased linearly across the quantiles in 2005 while daughter’s mobility then
appeared to have an inverted-U pattern reaching its lowest level at the 25th quantile. In
contrast, in 2013, daughter’s mobility remained more or less stable across the quantiles
while son’s mobility shows some variations across the quantiles. Figure 2 depicts the
mobility patterns of sons and daughters in 2005 and 2013 based on the results from
quantile regression.
Overall, the estimated income elasticities suggest that: firstly, being below 50% in most
cases, they suggest that incomes are less sticky between generations in Ethiopia generally
compared with similar evidence elsewhere. Secondly, the estimated elasticities are
smaller for sons for the most part, compared with that of daughters, suggesting that there
is a gender gap in income mobility in favour of sons. Thirdly, daughters appear to have
gained some ground in income mobility terms between 2005 and 2013, albeit this being
mostly at the median and lower quantiles of the income distribution, thus with the
resulting narrowing of the gender gap in income mobility between 2005 and 2013.

Table 2: Intergenerational Income Mobility of Sons and Daughters, OLS & Quantile Regression.
OLS Q(.1) Q(.25) Q(.5) Q(.75) Q(.9)
Sons & Daughters, 2005
Father’s income 0.432*** 0.504*** 0.561*** 0.514*** 0.423*** 0.304***
(0.073) (0.183) (0.138) (0.069) (0.079) (0.102)
N 397 397 397 397 397 397

Sons, 2005
Father’s income 0.389*** 0.538** 0.490*** 0.474*** 0.403*** 0.287**
(0.094) (0.235) (0.174) (0.113) (0.098) (0.125)
N 254 254 254 254 254 254

Daughters, 2005
Father’s income 0.559*** 0.539** 0.716*** 0.674*** 0.560*** 0.576***
(0.110) (0.254) (0.228) (0.140) (0.124) (0.165)
N 143 143 143 143 143 143

Sons & Daughters, 2013 2013

Father’s income 0.369*** 0.519*** 0.435*** 0.283*** 0.333*** 0.373***
(0.040) (0.088) (0.072) (0.048) (0.049) (0.049)
N 789 789 789 789 789 789

Sons, 2013
Father’s income 0.357*** 0.577*** 0.434*** 0.277*** 0.351*** 0.310***
(0.051) (0.085) (0.095) (0.067) (0.057) (0.057)
N 478 478 478 478 478 478

14
 Daughters, 2013
Father’s income 0.436*** 0.450*** 0.472*** 0.446*** 0.452*** 0.434***
(0.066) (0.119) (0.121) (0.090) (0.086) (0.078)
N 311 311 311 311 311 311
Standard errors from 250 bootstrap replications in parentheses
The bootstrap replications are based on: 332; 225; 131; 655; 422 and 288 households, respectively, for sons &
daughters, 2005; for sons, 2005; daughters, 2005; sons and daughters, 2013; sons, 2013 and daughters 2013.
*p<0.1, ** p<0.05, *** p<0.01

4.2 Intergenerational mobility in educational and occupational status

4.2.1 Educational status

Table 3 reports unconditional probabilities from a transition matrix of the educational

attainment of children and their father. The combined 2005 and 2013 sample reveals that
63% of children have achieved a higher level of educational attainment than their father,
while only 6% have done worse, with the balance representing those attaining the same
level of education as their father. There is a statistically significant gap in
intergenerational educational attainment between sons and daughters, in favour of the
former. Thus, compared with daughters, sons are 9 percentage points more likely to attain
higher levels of education and 4 percentage points less likely to do worse than their
father. Splitting the sample by survey year provides more or less similar picture in terms
of the overall child-father attainment patterns. However, there are gender differences in
these between 2005 and 2013. Compared with daughters, sons were 7 (2) percentage
points more (less) likely to do better (worse) educationally than their father in 2005,
while in 2013 they were 11 (5) percentage points more (less) likely to do better (worse)
than their father.

Table 3: Child-father Highest Educational Levels Attained (%)

Sons & daughters Son Daughter Diff.
(1) (2) (3) (2–3)
Combined 2005 & 2013 sample (N=13252)
% in a lower level than their father 6.1 4.5 8.4 -4.0***
% in the same level as their father 30.8 28.6 33.8 -5.2***
% in higher level than their father 63.1 66.9 57.8 9.1***
Total 100 100 100

2005 sample (N=5493)

% in a lower level than their father 4.3 3.3 5.6 -2.2***
% in the same level as their father 31.8 29.8 34.6 -4.8***
% in higher level than their father 63.94 66.9 59.8 7.0***

15
 Total 100 100 100

2013 sample (N=7759)

% in a lower level than their father 7.5 5.3 10.4 -5.1***
% in the same level as their father 30.0 27.7 33.2 -5.4***
% in higher level than their father s 62.5 66.9 56.4 10.6***
Total 100 100 100
*** p<0.01, ** p<0.05, * p<0.1
Authors own computation based on Ethiopian LFS 2005 and 2013.

Conditional marginal effects from the ordered probit models estimating transition
probabilities in educational status are reported in Table 4.16 The results reveal that
compared with those with fathers without education, sons & daughters with fathers who
have some level of education are generally significantly more (less) likely to attain higher
(lower) levels of education. Thus, for example, in 2005 young adults with fathers who
attained ‘grade 11 or higher’ are 50 percentage points more likely to attain the same level
qualification as their father; and 26 percentage points less likely to have no education,
compared with the base category of parents with no education. On the other hand, young
adults with fathers with ‘grades 7-10’ level of qualification are 39 percentage points more
likely to excel their parents educationally and 35 percentage points less likely to attain
lower levels of education than their father.
The estimated marginal effects suggest that the general picture relating to the child-father
patterns of educational mobility remains the same in 2013, so children who have fathers
with some level of education are generally found to be significantly more (less) likely to
attain higher (lower) grades of education. However, the estimated marginal effects are
generally lower in magnitude in 2013. Thus, focusing on the specific examples earlier, in
2013 young adults with parents who have ‘grade 11 or higher’ level of education are only
44% percentage points more likely to attain ‘grade 11 or higher’ and only 19 percentage
points less likely to have no education. Also, young adults with parents who attained a
‘grades 7-10’ level qualification are only 22 percentage points more likely to excel their
parents educationally and only 20 percentage points less likely to attain lower levels of
education in 2013. It appears, therefore, that there has been a general decline in

16
The full list of marginal effects corresponding to each panel of Tables 4 and Table 6 are available in an
accompanying separate file of Tables.

16
educational mobility between 2005 and 2013. Moreover, the decline in educational
mobility appears to be weighted more by the decline in daughter’s attainment.

Table 4: Intergenerational Mobility in Educational Attainment of Sons & Daughters, Marginal

Effects from Ordered Probit
No education Grades1-6 Grades 7-10 Grade11 & higher
Sons & Daughters, 2005
Grades1-6 -0.178*** -0.068*** -0.000 0.246***
(0.008) (0.005) (0.003) (0.013)
Grades 7-10 -0.233*** -0.114*** -0.044*** 0.391***
(0.009) (0.008) (0.008) (0.022)
Grade11 & higher -0.260*** -0.146*** -0.093*** 0.499***
(0.008) (0.008) (0.010) (0.021)
N 5493 5493 5493 5493

Sons, 2005
Grades1-6 -0.158*** -0.087*** -0.004 0.249***
(0.010) (0.007) (0.004) (0.017)
Grades 7-10 -0.202*** -0.140*** -0.049*** 0.391***
(0.010) (0.011) (0.011) (0.028)
Grade11 & higher -0.227*** -0.186*** -0.117*** 0.530***
(0.009) (0.011) (0.015) (0.026)
N 3210 3210 3210 3210

Daughters, 2005
Grades1-6 -0.205*** -0.046*** 0.003 0.248***
(0.015) (0.005) (0.004) (0.021)
Grades 7-10 -0.276*** -0.083*** -0.040*** 0.398***
(0.017) (0.010) (0.013) (0.036)
Grade11 & higher -0.299*** -0.099*** -0.067*** 0.465***
(0.014) (0.010) (0.014) (0.032)
N 2283 2283 2283 2283

Sons & Daughters, 2013

Grades1-6 -0.112*** -0.054*** -0.005** 0.171***
(0.006) (0.004) (0.002) (0.010)
Grades 7-10 -0.133*** -0.071*** -0.019*** 0.223***
(0.008) (0.005) (0.005) (0.016)
Grade11 & higher -0.189*** -0.131*** -0.121*** 0.441***
(0.006) (0.005) (0.009) (0.016)
N 7759 7759 7759 7759

Sons, 2013
Grades1-6 -0.102*** -0.081*** -0.016*** 0.199***
(0.007) (0.006) (0.004) (0.014)
Grades 7-10 -0.117*** -0.100*** -0.035*** 0.253***
(0.008) (0.008) (0.008) (0.021)
Grade11 & higher -0.154*** -0.168*** -0.165*** 0.487***
(0.007) (0.008) (0.014) (0.021)
N 4483 4483 4483 4483

Daughters, 2013
Grades1-6 -0.119*** -0.032*** 0.004 0.146***

17
 (0.012) (0.004) (0.003) (0.016)
Grades 7-10 -0.153*** -0.046*** -0.007 0.205***
(0.014) (0.006) (0.005) (0.023)
Grade11 & higher -0.227*** -0.091*** -0.078*** 0.396***
(0.011) (0.007) (0.011) (0.023)
N 3276 3276 3276 3276
Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

4.2.2 Occupational status

Table 5 reports unconditional probabilities from a transition matrix of the occupational

status of children and their parents. The combined 2005 and 2013 sample reveals that
occupationally children are spread more or less equally among the three possible states of
children having had better, same or worse occupational status as their father. The Table
also reports a statistically significant gender gap in occupational mobility. Accordingly,
compared with daughters, sons are 4 percentage point more likely to occupy higher
occupational status than their father and 4 percentage point less likely to occupy lower
occupational status than their father, which are both found to be statistically significant.
Splitting the sample by survey year reveals that the proportions of children with the same
or better occupational status than their father has increased between 2005 and 2013, while
that of children with a lower occupational status has declined in 2013. This appears to
suggest an improvement in occupational mobility in 2013. No statistically significant
occupational gender gap was found in 2005. On the other hand, statistically significant
gender gap has been found in 2013 where, compared with daughters, sons are found to be
6 percentage points more likely to occupy higher occupational status than their father and
6 percentage points less likely to occupy lower occupational status. This suggests the
improvement in occupational mobility highlighted earlier is weighted more by
improvements in the occupational fortunes of sons.

Table 5: Child-father Occupational Status (%)

Sons & daughters Sons Daughters Diff.
(1) (2) (3) (2–3)
Combined 2005 & 2013 sample (N=13252)
% in a lower occupational group than father 32.4 30.8 34.7 -4.0***
% in the same occupational group as father 32.9 32.8 33.1 -0.22
% in higher occupational group than father 34.6 36.4 32.2 4.2***
Total 100 100 100

18
 2005 sample (N=5493)
% in a lower Occupational group than father 35.9 35.4 36.6 -1.2
% in the same Occupational group as father 32.0 31.7 32.5 -0.8
% in higher Occupational group than father 32.1 33.0 30.9 2.1*
Total 100 100 100

2013 sample (N=7759)

% in a lower Occupational group than father 30.0 27.5 33.4 -6.0***
% in the same Occupational group as father 33.6 33.7 33.5 -0.3
% in higher Occupational group than father 36.4 38.8 33.1 5.7***
Total 100 100 100
*** p<0.01, ** p<0.05, * p<0.1
Authors own computation based on Ethiopian LFS 2005 and 2013.

Conditional marginal effects from the ordered probit models estimating intergenerational mobility
in occupational status are reported in Table 6. The results reveal that compared with fathers
without any occupation; children with fathers who have higher levels of occupational status are
generally more likely to be in better occupational status. For example, compared with fathers with
no occupation, children with parents in ‘managerial and professional’ occupations are 5
percentage points more likely to have a ‘managerial and professional’ occupation; and 12
percentage points less likely to have no or elementary occupations; or, 5 percentage points less
likely to have a lower occupational status generally than their father in 2005. A similar picture
emerges in 2013, but with marginal effects that are relatively higher in magnitude than those for
2005. For example, compared with fathers with no occupation, children with fathers that have a
‘managerial and professional’ occupation were only 5 percentage points less likely to have a
lower occupational status in 2005, a lower figure than the 13 percentage points obtained for 2013.
The marginal effects also reveal some gender differential in occupational mobility in that the
estimated marginal effects are higher in magnitude for sons, thus suggesting a higher mobility for
sons than daughters. For example, focusing on the highest occupational category, sons with
fathers who have a ‘managerial and professional’ occupation are 16 percentage points more likely
to occupy the same occupational status in 2013, while the corresponding figure for daughters is
only 10 percentage points. Generally, controlling for father’s occupation, sons are more (less)
likely to have higher (lower) occupational statuses than daughters; and this is more apparent in
2013.

Table 6: Intergenerational Mobility in Occupational Status of Sons & Daughters, Marginal

Effects from Ordered Probit
None Elementary Skilled Services & Machine & Managerial &

19
 Agricultural Sales crafts Professional
Sons & daughters,
2005
Elementary 0.036 0.001** -0.002 -0.009 -0.016 -0.011*
(0.023) (0.001) (0.001) (0.006) (0.010) (0.006)
Skilled Agricultural 0.036** 0.001** -0.002*** -0.009** -0.016** -0.011**
(0.014) (0.001) (0.001) (0.004) (0.006) (0.004)
Services & Sales -0.023 -0.002 0.001 0.005 0.010 0.008
(0.018) (0.002) (0.001) (0.004) (0.008) (0.006)
Machine operator -0.057*** -0.006*** 0.002*** 0.013*** 0.027*** 0.021***
(0.018) (0.002) (0.001) (0.004) (0.009) (0.007)
Managerial & Prof. -0.108*** -0.016** 0.002* 0.022*** 0.053*** 0.048***
(0.028) (0.006) (0.001) (0.005) (0.015) (0.016)
N 5493 5493 5493 5493 5493 5493

Sons, 2005
Elementary 0.043 0.004* -0.003 -0.008 -0.021 -0.015
(0.028) (0.002) (0.002) (0.005) (0.014) (0.009)
Skilled Agricultural 0.034* 0.003* -0.003** -0.006* -0.017* -0.012*
(0.017) (0.002) (0.001) (0.003) (0.009) (0.006)
Services & Sales -0.036 -0.006 0.002* 0.006 0.019 0.015
(0.022) (0.004) (0.001) (0.003) (0.012) (0.010)
Machine operator -0.027 -0.004 0.001 0.004 0.014 0.011
(0.023) (0.004) (0.001) (0.004) (0.012) (0.010)
Managerial & Prof. -0.114*** -0.027** 0.001 0.015*** 0.063*** 0.062**
(0.033) (0.011) (0.002) (0.003) (0.019) (0.024)
N 3210 3210 3210 3210 3210 3210

Daughters, 2005
Elementary 0.018 -0.000 -0.000 -0.007 -0.006 -0.004
(0.038) (0.001) (0.001) (0.015) (0.013) (0.009)
Skilled Agricultural 0.029 -0.000 -0.001 -0.011 -0.010 -0.007
(0.023) (0.000) (0.001) (0.009) (0.008) (0.005)
Services & Sales 0.004 0.000 -0.000 -0.001 -0.001 -0.001
(0.030) (0.000) (0.001) (0.012) (0.011) (0.008)
Machine operator -0.097*** -0.006** 0.001*** 0.033*** 0.038*** 0.031***
(0.028) (0.003) (0.000) (0.009) (0.012) (0.010)
Managerial & Prof. -0.095* -0.006 0.001*** 0.032** 0.037* 0.030
(0.049) (0.006) (0.000) (0.015) (0.020) (0.019)
N 2283 2283 2283 2283 2283 2283

Sons & Daughters,

2013
Elementary 0.090*** 0.008*** -0.007*** -0.022*** -0.023*** -0.046***
(0.016) (0.001) (0.002) (0.004) (0.004) (0.008)
Skilled Agricultural 0.047*** 0.006*** -0.003*** -0.011*** -0.013*** -0.027***
(0.011) (0.001) (0.001) (0.003) (0.003) (0.006)
Services & Sales -0.000 -0.000 0.000 0.000 0.000 0.000
(0.014) (0.002) (0.000) (0.003) (0.004) (0.009)
Machine operator -0.032* -0.006 0.001* 0.007* 0.009* 0.022*
(0.018) (0.004) (0.000) (0.004) (0.005) (0.013)
Managerial & Prof. -0.137*** -0.039*** -0.009*** 0.019*** 0.039*** 0.127***
(0.015) (0.006) (0.003) (0.002) (0.004) (0.019)
N 7759 7759 7759 7759 7759 7759

Sons, 2013

20
Elementary 0.082*** 0.013*** -0.006*** -0.014*** -0.028*** -0.046***
(0.020) (0.003) (0.002) (0.004) (0.007) (0.010)
Skilled Agricultural 0.038*** 0.007*** -0.002** -0.006*** -0.013*** -0.024***
(0.014) (0.003) (0.001) (0.002) (0.005) (0.009)
Services & Sales -0.009 -0.002 0.000 0.001 0.003 0.007
(0.017) (0.004) (0.000) (0.003) (0.006) (0.012)
Machine operator -0.050** -0.013** -0.002 0.007*** 0.018** 0.040**
(0.021) (0.006) (0.002) (0.002) (0.008) (0.018)
Managerial & Prof. -0.140*** -0.052*** -0.022*** 0.009*** 0.049*** 0.155***
(0.018) (0.009) (0.006) (0.002) (0.006) (0.027)
N 4483 4483 4483 4483 4483 4483

Daughters, 2013
Elementary 0.106*** 0.002 -0.007*** -0.038*** -0.015*** -0.048***
(0.027) (0.002) (0.002) (0.010) (0.004) (0.011)
Skilled Agricultural 0.050*** 0.002** -0.003*** -0.017*** -0.007*** -0.025***
(0.019) (0.001) (0.001) (0.007) (0.003) (0.010)
Services & Sales 0.016 0.001 -0.001 -0.006 -0.002 -0.009
(0.024) (0.002) (0.001) (0.008) (0.003) (0.013)
Machine operator -0.000 -0.000 0.000 0.000 0.000 0.000
(0.032) (0.003) (0.001) (0.011) (0.005) (0.018)
Managerial & Prof. -0.128*** -0.024*** -0.000 0.035*** 0.020*** 0.097***
(0.027) (0.007) (0.002) (0.006) (0.005) (0.025)
N 3276 3276 3276 3276 3276 3276
Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

5. Summary and conclusion

The paper examined the extent of intergenerational mobility in monetary and non-
monetary economic status between young adult children and their parents in Ethiopia
using data from two of the three most recent and comprehensive national labour force
surveys conducted in 2005 and 2013.
The use of both monetary and non-monetary economic statuses measures in examining
intergenerational mobility provides a more comprehensive account of mobility than is
provided by focusing exclusively on income (monetary) based measure of mobility as
much of the literature does. Such exclusive focus may, as discussed in the paper,
understate the influence of family background on economic and social inequality. On the
other hand, focusing entirely on non-monetary measures runs the risk of misclassification
and hence of obtaining biased estimates of intergenerational mobility.
The results obtained suggest that there is generally moderate “stickiness” in income
mobility across generations in Ethiopia. Sons are found to have higher intergenerational

21
income mobility. The paper finds increased income mobility between 2005 and 2013; and
this appears to be driven largely by the marked increase in mobility at the median and
lower quantiles of the child income distribution. On the other hand, there has been a
decline in mobility particularly at the top income quantile. Comparing sons and
daughters, the latter appeared to have gained some ground in intergenerational income
mobility between 2005 and 2013, albeit this being at the median and lower quantiles of
the income distribution for the most part. Still, this seems to have contributed to a
reduction of the gender gap in intergenerational income mobility in 2013.
Similar patterns emerge in the intergenerational mobility in educational and occupational
statuses, both overall and in terms of the differential patterns between sons and daughters.
Accordingly, compared with daughters, sons are found to be more (less) likely to attain
higher (lower) levels of education than their parents generally. On the other hand, there
seems to have been a general decline in educational mobility between 2005 and 2013;
and the decline in educational mobility appears to have been weighted more by the
decline in daughter’s attainment.
Occupationally too, sons are generally more (less) likely to have a ‘better’ (‘worse’)
occupational status than their parents vis-à-vis daughters. Splitting the sample by survey
year revealed that the proportions of children with the same or better occupational status
than their father has increased between 2005 and 2013, while that of children with a
lower occupational status has declined in 2013.
There is virtually no evidence on intergenerational mobility in the context of low income
countries in general and Sub-Saharan Africa in particular. The paper thus provides some
valuable insights into intergenerational mobility in these contexts. The use of mixed
approaches involving monetary and non-monetary frameworks is likely to offer a broader
account of intergenerational mobility, which may be important given possible
measurement error in income.

22
References

Atkinson, A. (1981), On intergenerational income mobility in Britain, Journal of Post-Keynesian

Economics, 3, pp 194-218.
Atkinson, A., Maynard, A. and Trinder, C. (1983) Parents and Children: Incomes in Two
Generations. London: Heinemann.
Becker, G. (1991), A Treatise on the Family. Cambridge, Massachusetts: Harvard University
Press.
Becker, G. and Tomes, N. (1986), Human capital and the rise and fall of families, Journal of
Labor Economics, 4, pp S1-S39.
Behrman, J. and Taubman, P. (1985) Intergenerational earnings mobility in the United States:
some estimates and a test of Becker’s intergenerational endowments model, Review
of Economics and Statistics, 67, pp 144-151.
Behrman, J. and Taubman, P. (1990), The intergenerational correlation between children’s adult
earnings and their parents’ income: Results from the Michigan Panel Survey of
Income Dynamics, Review of Income and Wealth, 36, pp 115-127.
Binder, M. and Woodruff, C. (2002), Inequality and intergenerational mobility in schooling: The
case of Mexico, Economic Development and Cultural Change, 50, pp 249-268
Binder, M. and Woodruff, C. (2002), Inequality and Intergenerational Mobility in Schooling: The
Case of Mexico, Economic Development and Cultural Change, 50 (2), 249-267.
Bjorklund, A. and Chadwick, L. (2003), Intergenerational Income Mobility in Permanent and
Separated Families, Economic Letters, 80, pp 239 – 246.
Bjorklund, A. and Jantti, M. (1997), Intergenerational income mobility in Sweden compared to
the United States, American Economic Review, 87, pp 1009-1018.
Black, S., Devereux, P., Lundborg, P. and Majlesi, K (2015), Poor Little Rich Kids? The
Determinants of the Intergenerational Transmission of Wealth, NBER Working Paper
Series, Working Paper 21409.
Black, S. and Deverux, P. (2011), Recent Developments in Intergenerational Mobility, in Card,
D. and O. Ashenfelter (eds.), Handbook of Labor Economics, 4 (B), Chapter 16,
1487-1541.
Blau, F. and Kahn, L. (2000), Gender Differences in Pay, Journal of Economic Perspectives, 14
(4), pp. 75–99.
Bowles, S. (1972), Schooling and Inequality from Generation to Generation, Journal of Political
Economy, pp S219 – S251.
Bowles, S. and Gintis, H. (2002), The Inheritance of Inequality, The Journal of Economic
Perspectives, 16, pp 3-30.
Buchinsky, M. (1998), Recent advances in quantile regression models: A practical guide for
empirical research, Journal of Human Resources, 3, 88-126.
Carmichael, F. (2000), Intergenerational mobility and occupational status in Britain, Applied
Economics Letters, 7, pp 391-396.
Chadwick, L. and Solon, G. (2002), Intergenerational income mobility among daughters,
American Economic Review, 92, pp 335-344.
Corak, M. (2001), Are the kids all right? Intergenerational mobility and child well-being in
Canada, mimeo, Statistics Canada.
Corak, M. and Heisz, A. (1999), The intergenerational earnings and income mobility of Canadian
men, Journal of Human Resources, 34, pp 504-533.
Couch, K. and Dunn, T. (1997), Intergenerational correlation in labour market status: A
comparison of the United States and Germany, Journal of Human Resources, 32, pp
210-232.
Dearden, L., Machin, S. and Reed, H. (1997), Intergenerational mobility in Britain, Economic
Journal, 107, pp 47-66.

23
Dunn, C. (2004), Intergenerational transmission of lifetime earnings: New evidence from Brazil,
Department of Economics and Population Studies Centre, University of Michigan.
Eide, E. and Showalter, M. (1999), Factors affecting the transmission of earnings across
generations: A quantile regression approach. Journal of Human Resources, 34, pp
253-267.
Erikson, R. and Goldthorpe, J. (2002), Intergenerational Inequality: A Sociological Perspective,
The Journal of Economic Perspectives, 16, pp 31-44.
Ermisch, J. and Francesconi, M. (2002), Intergenerational mobility in Britain: New evidence from
the BHPS, mimeo, ISER, University of Essex.
Ferreira, S. and Veloso, F. (2011), Intergenerational Mobility of Wages in Brazil, Brazilian
Review of Econometrics, 26 (2), 181-211.
Gang, I. and Zimmermann, K. (2000), Is Child Like Parent? Educational Attainment and Ethnic
Origin, The Journal of Human Resources, 35, 550 – 569.
Gaviria, A. (2002), Intergenerational Mobility, Sibling Inequality and Borrowing Constraints,
Economics of Education Review, 21, 331 – 340.
Goldberger, A. (1989), Economic and mechanical models of intergenerational transmission,
American Economic Review, 79, pp 504-513.
Green, W. (2002), Econometric Analysis. New York: Prentice-Hall International.
Hagy, A. and Staniec, F. (2002), Immigrant Status, Race, and Institutional Choice in Higher
Education, Economics of Education Review, 21, pp 381 – 392.
Han, S. and Mulligan, C. (2001), Human capital, heterogeneity and estimated degrees of
intergenerational mobility, Economic Journal, 111, pp 207-243.
Hanushek, E. (1992), The Trade-off between Child Quantity and Quality, Journal of Political
Economy, 100, Number 1, pp 84 – 117.
Hertz, T. (2002), Intergenerational Economic Mobility of Black and White Family in the United
States, Paper presented at the Society of Labor Economist, Annual Meeting.
Hotchkiss, J. and Pitts, M. (2007), The Role of Labor Market Intermittency in Explaining Gender
Wage Differentials, The American Economic Review, 97 (2), pp. 417-421.
Johnson, P. (2002), Intergenerational Dependence in Education and Income, Applied Economic
Letters, 9, pp 159 –162.
Koenker, R. and G. Bassett (1978), Regression quantiles, Econometrica, 46, 33-50.
Kolev, A. and Robles, P. (2010), Addressing the Gender Pay Gap in Ethiopia: How Crucial is the
Quest for Educational Parity, Journal of African Economies, 19 (5), 718 – 767.
Lauer, C. (2003), Family Background, Cohort and Education: A French – German Comparison
based on a Multivariate Ordered Probit Model of Educational Attainment, Labour
Economics, 10, pp 231 – 251.
Levine, D. and Mazumder, B. (2002), Choosing the right parents: Changes in the
intergenerational transmission of inequality – between 1980 and the early 1990s, WP
2002-08, Federal Reserve Bank of Chicago.
Lien, D. and N. Balakrishnan (2005), On Regression Analysis with Data Cleaning via Trimming,
Winsorization, and Dichotomization, Communication in Statistics – Simulation and
Computation, 34 (4), 839 – 849.
Lillard, D. (2001), Earnings and income mobility: Cross-national estimates of the
intergenerational mobility in earnings, Viertejahrshefte Zur Wirtschaftsforschung, 20,
pp 51-58.
Mazumder, B. (2001), Earnings mobility in the US: A new look at intergenerational inequality,
mimeo, Federal Reserve Bank of Chicago.
Mulligan, C. (1997), Parental Priorities and Economic Inequality, Chicago: Univ. of Chicago
Press
Mulligan, C. (1999), Galton versus the human capital approach to inheritance, Journal of
Political Economy, 107, pp S184-S224.

24
Naga, R. (2002), Estimating the intergenerational correlation of incomes: An errors-in-variables
framework, Economica, 69, pp 69-91.
Nguyen, A., Haile, G. and Taylor, J (2005), Ethnic and Gender Differences in Intergenerational
Mobility: a study of 26-year-olds in the USA, Scottish Journal of Political Economy,
52(4), pp. 544-564.
Nickell, S. (1982), The determinants of occupational success in Britain, Review of Economic
Studies, 49, pp 43-53.
Osterbacka, E. (2001), Family background and economic status in Finland, Scandinavian Journal
of Economics, 103, pp 467-484.
Osterberg, T. (2000), Intergenerational income mobility in Sweden: What do tax-data show?
Review of Income and Wealth, 46, pp 421-436.
Painter, G. and Levine, D. (2000), Family Structure and Youths’ Outcomes: Which Correlations
are Causal? The Journal of Human Resources, 35, pp 524 – 549.
Peters, E. (1992), Patterns of intergenerational mobility in income and earnings, Review of
Economics and Statistics, 73, pp 456-466.
Phipps, S., Burton, P. and Lethbridge, L. (2001), In and out of the Labour Market: Long-Term
Income Consequences of Child-Related Interruptions to Women's Paid Work, The
Canadian Journal of Economics, 34 (2), 411-429.
Piraino, P. (2014), Intergenerational Earnings Mobility and Equality of opportunities in South
Africa, SALDRU Working Paper Series, No. 131, South Africa.
Solon, G. (1992), Intergenerational income mobility in the United States, American Economic
Review, 82, pp 393-408.
Solon, G. (2002), Cross-Country Differences in Intergenerational Earnings Mobility,
The Journal of Economic Perspectives, 16, pp 59-66.
Thomas, D. (1996), Education across generations in South Africa, American Economic Review,
86, pp 330-334.
UNDP (2013), Humanity Divided: Confronting Inequality in Developing Countries, United
Nations Development Programme, Bureau for Development Policy, NY.
Valero-Gil, J. and Tijierina-Guajardo, J. (2002), Effects of education on the intergenerational
transmission of labour income in Mexico, Eastern Economic Journal, 28, pp 381-
392.
Zimmerman, D. (1992), Regression toward mediocrity in economic stature, American Economic
Review, 82, pp 409-429.

25
Appendix

Table A1: Descriptive Statistics, LFS 2005 and 2013

Variable Mean Std. Min Max Mean Std. Min Max
Dev. Dev.
2005 2013
Demographic and household characteristics
Age 28.16 2.88 25 35 28.38 3.03 25 35
Female 0.42 0.49 0 1 0.42 0.49 0 1
Child’s highest grade:
don't read 0.21 0.41 0 1 0.15 0.36 0 1
primary or less 0.16 0.36 0 1 0.14 0.34 0 1
grade 7-10 0.24 0.43 0 1 0.33 0.47 0 1
grade 11 & above 0.39 0.49 0 1 0.39 0.49 0 1
Parent’s highest grade:
don't read 0.60 0.49 0 1 0.48 0.50 0 1
primary or less 0.24 0.43 0 1 0.27 0.44 0 1
grade 7-10 0.07 0.26 0 1 0.11 0.31 0 1
grade 11 & above 0.09 0.29 0 1 0.14 0.35 0 1
Child’s occupation:
none 0.35 0.48 0 1 0.30 0.46 0 1
elementary 0.19 0.39 0 1 0.18 0.38 0 1
skilled agriculture 0.08 0.27 0 1 0.13 0.34 0 1
services 0.16 0.37 0 1 0.15 0.36 0 1
craft & machinery 0.15 0.36 0 1 0.10 0.30 0 1
managerial & professional 0.06 0.25 0 1 0.14 0.34 0 1
Parent’s occupation:
none 0.40 0.49 0 1 0.39 0.49 0 1
elementary 0.07 0.26 0 1 0.10 0.30 0 1
skilled agriculture 0.23 0.42 0 1 0.23 0.42 0 1
services 0.13 0.33 0 1 0.14 0.34 0 1
craft & machinery 0.13 0.34 0 1 0.07 0.25 0 1
managerial & professional 0.05 0.21 0 1 0.08 0.26 0 1
Two parent household 0.45 0.50 0 1 0.45 0.50 0 1
No. of siblings 3.64 2.05 1 22 3.38 1.88 1 10
Region:
Tigray 0.06 0.23 0 1 0.06 0.24 0 1
Afar 0.02 0.14 0 1 0.02 0.13 0 1
Amhara 0.14 0.35 0 1 0.14 0.34 0 1
Oromia 0.17 0.38 0 1 0.16 0.36 0 1
Somalie 0.03 0.18 0 1 0.03 0.16 0 1
Benishangul-Gumuz 0.01 0.12 0 1 0.02 0.13 0 1
SNNPR 0.14 0.35 0 1 0.15 0.36 0 1
Gambela 0.00 0.06 0 1 0.02 0.14 0 1
Harari 0.03 0.16 0 1 0.03 0.16 0 1
Addis Ababa 0.36 0.48 0 1 0.35 0.48 0 1
Dire Dawa 0.03 0.16 0 1 0.03 0.18 0 1
No. of children 5,493 7,759

Earnings characteristics
Log of child’s income 5.64 0.77 3.33 7.09 7.04 0.72 5.30 8.27
Log of father’s income 5.92 0.87 3.40 7.01 7.15 0.73 5.29 8.26
Log of father’s ‘permanent’ income 5.65 0.60 3.61 6.91 6.96 0.61 4.71 8.07
No. of children 397 789

26
Table A2: Income Equation Predicting Father’s “Permanent” Income
Age (father’s) 0.055***
(0.003)
Age squared -0.001***
(0.000)
No. of siblings -0.011***
(0.003)
Two parent household 0.031***
(0.010)
Grades 1-6 0.211***
(0.019)
Grades 7-10 0.454***
(0.019)
Grade 11 or higher 0.866***
(0.020)
Elementary occupation -0.197***
(0.062)
Skilled Agricultural 0.004
(0.069)
Services & Sales 0.094
(0.062)
Machine operator 0.282***
(0.062)
Managerial & Professional 0.444***
(0.062)
Manufacturing 0.109***
(0.026)
Wholesale & retail trade 0.083***
(0.030)
Other industries 0.218***
(0.023)
Tigray 0.013
(0.020)
Afar 0.208***
(0.020)
Amhara -0.023*
(0.014)
Somalie 0.288***
(0.025)
Benishangul-Gumuz 0.069***
(0.020)
SNNPR -0.091***
(0.014)
Gambela 0.195***
(0.021)
Harari 0.051**
(0.024)
Addis Ababa 0.129***
(0.012)
Dire Dawa 0.090***
(0.026)
2013 1.178***
(0.008)
Constant 3.826***
(0.082)
R-squared 0.676
N 21248
Robust standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Base categories: Oromia region

27
Table A3: Intergenerational Mobility in Educational Attainment, Marginal Effects from Ordered
Probit, Sons and Daughters (2005)
No education Grades1-6 Grades 7-10 Grade11 & higher

Grades1-6 -0.178*** -0.068*** -0.000 0.246***

(0.008) (0.005) (0.003) (0.013)
Grades 7-10 -0.233*** -0.114*** -0.044*** 0.391***
(0.009) (0.008) (0.008) (0.022)
Grade11 & higher -0.260*** -0.146*** -0.093*** 0.499***
(0.008) (0.008) (0.010) (0.021)
Age -0.052** -0.014** 0.002* 0.064**
(0.026) (0.007) (0.001) (0.032)
Age squared 0.001** 0.000** -0.000* -0.001**
(0.000) (0.000) (0.000) (0.001)
Female 0.038*** 0.010*** -0.001*** -0.047***
(0.008) (0.002) (0.001) (0.010)
Two parent HH 0.062*** 0.017*** -0.002*** -0.076***
(0.008) (0.002) (0.001) (0.010)
No. of siblings -0.001 -0.000 0.000 0.002
(0.002) (0.001) (0.000) (0.002)
Tigray 0.004 0.001 -0.001 -0.004
(0.020) (0.005) (0.003) (0.022)
Afar 0.275*** 0.008 -0.084*** -0.199***
(0.045) (0.007) (0.017) (0.023)
Amhara 0.041** 0.009** -0.008** -0.042**
(0.016) (0.003) (0.003) (0.017)
Somalie 0.225*** 0.014*** -0.065*** -0.174***
(0.035) (0.004) (0.013) (0.021)
Benishangul-Gumuz 0.154*** 0.018*** -0.040*** -0.131***
(0.043) (0.003) (0.014) (0.030)
SNNPR 0.023 0.005 -0.004 -0.024
(0.015) (0.004) (0.003) (0.016)
Gambela -0.024 -0.007 0.003 0.028
(0.053) (0.016) (0.005) (0.064)
Harari -0.065*** -0.022** 0.004** 0.083**
(0.023) (0.009) (0.002) (0.033)
Addis Ababa -0.149*** -0.072*** -0.020*** 0.241***
(0.011) (0.005) (0.003) (0.015)
Dire Dawa 0.017 0.004 -0.003 -0.018
(0.029) (0.007) (0.005) (0.031)

N 5493 5493 5493 5493

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

28
Table A4: Intergenerational Mobility in Educational Attainment, Marginal Effects from Ordered
Probit, Sons (2005)
No education Grades1-6 Grades 7-10 Grade11 & higher

Grades1-6 -0.158*** -0.087*** -0.004 0.249***

(0.010) (0.007) (0.004) (0.017)
Grades 7-10 -0.202*** -0.140*** -0.049*** 0.391***
(0.010) (0.011) (0.011) (0.028)
Grade11 & higher -0.227*** -0.186*** -0.117*** 0.530***
(0.009) (0.011) (0.015) (0.026)
Age 0.004 0.002 -0.000 -0.006
(0.031) (0.012) (0.001) (0.043)
Age squared -0.000 -0.000 0.000 0.000
(0.001) (0.000) (0.000) (0.001)
Two parent HH 0.077*** 0.031*** -0.002 -0.106***
(0.009) (0.004) (0.001) (0.013)
No. of siblings -0.005** -0.002** 0.000 0.007**
(0.002) (0.001) (0.000) (0.003)
Tigray -0.017 -0.007 0.003 0.022
(0.021) (0.009) (0.003) (0.027)
Afar 0.245*** 0.024*** -0.081*** -0.188***
(0.060) (0.008) (0.024) (0.032)
Amhara 0.031 0.010 -0.006 -0.035
(0.020) (0.007) (0.004) (0.022)
Somalie 0.219*** 0.026*** -0.070*** -0.175***
(0.044) (0.006) (0.017) (0.027)
Benishangul-Gumuz 0.089** 0.023*** -0.023 -0.089**
(0.045) (0.008) (0.014) (0.039)
SNNPR 0.018 0.006 -0.003 -0.021
(0.017) (0.006) (0.003) (0.020)
Gambela -0.104*** -0.063* -0.006 0.173**
(0.039) (0.034) (0.016) (0.088)
Harari -0.073*** -0.038** 0.003 0.108***
(0.023) (0.015) (0.004) (0.040)
Addis Ababa -0.129*** -0.089*** -0.021*** 0.240***
(0.013) (0.008) (0.004) (0.019)
Dire Dawa 0.046 0.014 -0.010 -0.050
(0.037) (0.010) (0.009) (0.037)

N 3210 3210 3210 3210

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

29
Table A5: Intergenerational Mobility in Educational Attainment, Marginal Effects from Ordered
Probit, Daughters (2005)
No education Grades1-6 Grades 7-10 Grade11 & higher

Grades1-6 -0.205*** -0.046*** 0.003 0.248***

(0.015) (0.005) (0.004) (0.021)
Grades 7-10 -0.276*** -0.083*** -0.040*** 0.398***
(0.017) (0.010) (0.013) (0.036)
Grade11 & higher -0.299*** -0.099*** -0.067*** 0.465***
(0.014) (0.010) (0.014) (0.032)
Age -0.121*** -0.019*** 0.005** 0.135***
(0.043) (0.007) (0.003) (0.047)
Age squared 0.002*** 0.000*** -0.000** -0.002***
(0.001) (0.000) (0.000) (0.001)
Two parent HH 0.036** 0.006** -0.002* -0.040**
(0.015) (0.002) (0.001) (0.016)
No. of siblings 0.004 0.001 -0.000 -0.005
(0.003) (0.001) (0.000) (0.004)
Tigray 0.042 0.005 -0.007 -0.041
(0.041) (0.004) (0.007) (0.038)
Afar 0.313*** -0.007 -0.089*** -0.217***
(0.067) (0.011) (0.025) (0.034)
Amhara 0.054** 0.006** -0.009** -0.051**
(0.026) (0.003) (0.005) (0.025)
Somalie 0.211*** 0.005 -0.053*** -0.164***
(0.057) (0.005) (0.019) (0.036)
Benishangul-Gumuz 0.299*** -0.005 -0.084*** -0.210***
(0.084) (0.013) (0.031) (0.042)
SNNPR 0.034 0.004 -0.005 -0.033
(0.027) (0.003) (0.004) (0.027)
Gambela 0.099 0.008** -0.019 -0.089
(0.097) (0.004) (0.024) (0.076)
Harari -0.034 -0.006 0.003 0.037
(0.049) (0.009) (0.003) (0.055)
Addis Ababa -0.169*** -0.050*** -0.018*** 0.237***
(0.019) (0.005) (0.005) (0.023)
Dire Dawa -0.019 -0.003 0.002 0.020
(0.046) (0.008) (0.004) (0.049)

N 2283 2283 2283 2283

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

30
Table A6: Intergenerational Mobility in Educational Attainment, Marginal Effects from Ordered
Probit, Sons and Daughters (2013)
No education Grades1-6 Grades 7-10 Grade11 & higher

Grades1-6 -0.112*** -0.054*** -0.005** 0.171***

(0.006) (0.004) (0.002) (0.010)
Grades 7-10 -0.133*** -0.071*** -0.019*** 0.223***
(0.008) (0.005) (0.005) (0.016)
Grade11 & higher -0.189*** -0.131*** -0.121*** 0.441***
(0.006) (0.005) (0.009) (0.016)
Age -0.045*** -0.020*** -0.005** 0.070***
(0.017) (0.008) (0.002) (0.026)
Age squared 0.001*** 0.000*** 0.000** -0.001***
(0.000) (0.000) (0.000) (0.000)
Female 0.044*** 0.020*** 0.005*** -0.069***
(0.005) (0.002) (0.001) (0.008)
Two parent HH 0.019*** 0.009*** 0.002*** -0.030***
(0.006) (0.002) (0.001) (0.009)
No. of siblings 0.005*** 0.002*** 0.001*** -0.008***
(0.001) (0.001) (0.000) (0.002)
Tigray -0.020 -0.008 0.003 0.026
(0.014) (0.006) (0.002) (0.018)
Afar 0.191*** 0.033*** -0.071*** -0.153***
(0.035) (0.003) (0.017) (0.021)
Amhara 0.003 0.001 -0.000 -0.003
(0.012) (0.004) (0.002) (0.014)
Somalie 0.142*** 0.031*** -0.048*** -0.125***
(0.028) (0.004) (0.012) (0.020)
Benishangul-Gumuz 0.041 0.013* -0.009 -0.044
(0.028) (0.008) (0.008) (0.028)
SNNPR -0.031*** -0.013*** 0.003** 0.041***
(0.011) (0.004) (0.001) (0.014)
Gambela 0.016 0.006 -0.003 -0.018
(0.025) (0.008) (0.005) (0.028)
Harari -0.091*** -0.047*** -0.008 0.146***
(0.016) (0.010) (0.007) (0.031)
Addis Ababa -0.137*** -0.087*** -0.051*** 0.275***
(0.008) (0.005) (0.004) (0.013)
Dire Dawa -0.030 -0.012 0.003** 0.039
(0.019) (0.008) (0.001) (0.026)

N 7759 7759 7759 7759

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

31
Table A7: Intergenerational Mobility in Educational Attainment, Marginal Effects from Ordered
Probit, Sons (2013)
No education Grades1-6 Grades 7-10 Grade11 & higher

Grades1-6 -0.102*** -0.081*** -0.016*** 0.199***

(0.007) (0.006) (0.004) (0.014)
Grades 7-10 -0.117*** -0.100*** -0.035*** 0.253***
(0.008) (0.008) (0.008) (0.021)
Grade11 & higher -0.154*** -0.168*** -0.165*** 0.487***
(0.007) (0.008) (0.014) (0.021)
Age -0.054*** -0.038*** -0.011*** 0.103***
(0.019) (0.013) (0.004) (0.036)
Age squared 0.001*** 0.001*** 0.000** -0.002***
(0.000) (0.000) (0.000) (0.001)
Two parent HH 0.037*** 0.026*** 0.008*** -0.071***
(0.006) (0.004) (0.002) (0.011)
No. of siblings 0.002 0.001 0.000 -0.004
(0.002) (0.001) (0.000) (0.003)
Tigray -0.005 -0.003 0.000 0.008
(0.014) (0.009) (0.001) (0.023)
Afar 0.175*** 0.058*** -0.066*** -0.167***
(0.043) (0.007) (0.022) (0.028)
Amhara 0.018 0.011 -0.002 -0.027
(0.013) (0.008) (0.002) (0.018)
Somalie 0.088*** 0.041*** -0.025** -0.104***
(0.031) (0.011) (0.013) (0.029)
Benishangul-Gumuz 0.022 0.013 -0.003 -0.032
(0.028) (0.015) (0.006) (0.037)
SNNPR -0.018 -0.012 0.000 0.030
(0.011) (0.007) (0.001) (0.018)
Gambela -0.037* -0.028 -0.003 0.068
(0.020) (0.017) (0.005) (0.042)
Harari -0.056*** -0.046*** -0.011 0.114***
(0.017) (0.017) (0.009) (0.042)
Addis Ababa -0.092*** -0.090*** -0.052*** 0.234***
(0.008) (0.007) (0.005) (0.017)
Dire Dawa -0.053*** -0.042*** -0.009 0.104***
(0.016) (0.015) (0.008) (0.038)

N 4483 4483 4483 4483

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

32
Table A8: Intergenerational Mobility in Educational Attainment, Marginal Effects from Ordered
Probit, Daughters (2013)
No education Grades1-6 Grades 7-10 Grade11 & higher

Grades1-6 -0.119*** -0.032*** 0.004 0.146***

(0.012) (0.004) (0.003) (0.016)
Grades 7-10 -0.153*** -0.046*** -0.007 0.205***
(0.014) (0.006) (0.005) (0.023)
Grade11 & higher -0.227*** -0.091*** -0.078*** 0.396***
(0.011) (0.007) (0.011) (0.023)
Age -0.022 -0.006 -0.001 0.029
(0.030) (0.008) (0.001) (0.039)
Age squared 0.000 0.000 0.000 -0.001
(0.001) (0.000) (0.000) (0.001)
Two parent HH -0.010 -0.003 -0.000 0.013
(0.010) (0.003) (0.000) (0.013)
No. of siblings 0.009*** 0.002*** 0.000* -0.011***
(0.003) (0.001) (0.000) (0.003)
Tigray -0.048* -0.010 0.008* 0.050
(0.028) (0.007) (0.004) (0.031)
Afar 0.200*** 0.010** -0.073*** -0.137***
(0.057) (0.004) (0.026) (0.031)
Amhara -0.027 -0.005 0.005 0.027
(0.023) (0.005) (0.004) (0.023)
Somalie 0.215*** 0.009** -0.080*** -0.144***
(0.050) (0.004) (0.022) (0.027)
Benishangul-Gumuz 0.071 0.009* -0.021 -0.059
(0.057) (0.005) (0.019) (0.043)
SNNPR -0.054*** -0.012** 0.009** 0.057***
(0.021) (0.005) (0.004) (0.022)
Gambela 0.077 0.009** -0.023 -0.063*
(0.047) (0.004) (0.016) (0.035)
Harari -0.137*** -0.041*** 0.001 0.178***
(0.030) (0.012) (0.009) (0.047)
Addis Ababa -0.199*** -0.076*** -0.041*** 0.316***
(0.017) (0.006) (0.006) (0.020)
Dire Dawa 0.040 0.006 -0.011 -0.035
(0.041) (0.005) (0.012) (0.034)

N 3276 3276 3276 3276

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

33
Table A9: Intergenerational Mobility in Occupational Status, Marginal Effects from Ordered
Probit, Sons and Daughters (2005)
None Elementary Skilled Services & Machine Managerial
Agricultural Sales operator & &
crafts Professional

Elementary 0.036 0.001** -0.002 -0.009 -0.016 -0.011*

(0.023) (0.001) (0.001) (0.006) (0.010) (0.006)
Skilled Agricultural 0.036** 0.001** -0.002*** -0.009** -0.016** -0.011**
(0.014) (0.001) (0.001) (0.004) (0.006) (0.004)
Services & Sales -0.023 -0.002 0.001 0.005 0.010 0.008
(0.018) (0.002) (0.001) (0.004) (0.008) (0.006)
Machine operator -0.057*** -0.006*** 0.002*** 0.013*** 0.027*** 0.021***
(0.018) (0.002) (0.001) (0.004) (0.009) (0.007)
Managerial & Prof. -0.108*** -0.016** 0.002* 0.022*** 0.053*** 0.048***
(0.028) (0.006) (0.001) (0.005) (0.015) (0.016)
Age -0.068* -0.005* 0.003* 0.016* 0.031* 0.023*
(0.036) (0.003) (0.002) (0.009) (0.016) (0.012)
Age squared 0.001* 0.000 -0.000* -0.000* -0.000* -0.000*
(0.001) (0.000) (0.000) (0.000) (0.000) (0.000)
Female 0.069*** 0.005*** -0.003*** -0.017*** -0.031*** -0.023***
(0.011) (0.001) (0.001) (0.003) (0.005) (0.004)
Two parent HH -0.005 -0.000 0.000 0.001 0.002 0.002
(0.012) (0.001) (0.001) (0.003) (0.005) (0.004)
No. of siblings 0.001 0.000 -0.000 -0.000 -0.001 -0.000
(0.003) (0.000) (0.000) (0.001) (0.001) (0.001)
Tigrary 0.017 0.002 -0.001 -0.004 -0.008 -0.006
(0.024) (0.002) (0.001) (0.006) (0.011) (0.008)
Afar 0.080** 0.004*** -0.004 -0.020* -0.035** -0.025**
(0.041) (0.001) (0.003) (0.011) (0.017) (0.011)
Amhara 0.033* 0.003* -0.001* -0.008* -0.015* -0.011*
(0.017) (0.001) (0.001) (0.004) (0.008) (0.006)
Somalie 0.138*** 0.001 -0.009*** -0.036*** -0.057*** -0.037***
(0.034) (0.003) (0.003) (0.009) (0.013) (0.007)
Benishangul-Gumuz -0.048 -0.007 0.001** 0.010 0.024 0.021
(0.034) (0.006) (0.000) (0.006) (0.017) (0.016)
SNNPR 0.007 0.001 -0.000 -0.002 -0.003 -0.003
(0.017) (0.002) (0.001) (0.004) (0.008) (0.006)
Gambela 0.422*** -0.062 -0.044*** -0.118*** -0.133*** -0.066***
(0.114) (0.041) (0.017) (0.031) (0.022) (0.007)
Harari 0.011 0.001 -0.000 -0.003 -0.005 -0.004
(0.037) (0.003) (0.001) (0.009) (0.017) (0.013)
Addis Ababa 0.017 0.002 -0.001 -0.004 -0.008 -0.006
(0.016) (0.002) (0.001) (0.004) (0.008) (0.006)
Dire Dawa 0.072* 0.004*** -0.004 -0.018* -0.032** -0.022**
(0.037) (0.001) (0.002) (0.010) (0.015) (0.010)

N 5493 5493 5493 5493 5493 5493

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

34
Table A10: Intergenerational Mobility in Occupational Status, Marginal Effects from Ordered
Probit, Sons (2005)
None Elementary Skilled Services & Machine Managerial
Agricultural Sales operator & &
crafts Professional

Elementary 0.043 0.004* -0.003 -0.008 -0.021 -0.015

(0.028) (0.002) (0.002) (0.005) (0.014) (0.009)
Skilled Agricultural 0.034* 0.003* -0.003** -0.006* -0.017* -0.012*
(0.017) (0.002) (0.001) (0.003) (0.009) (0.006)
Services & Sales -0.036 -0.006 0.002* 0.006 0.019 0.015
(0.022) (0.004) (0.001) (0.003) (0.012) (0.010)
Machine operator -0.027 -0.004 0.001 0.004 0.014 0.011
(0.023) (0.004) (0.001) (0.004) (0.012) (0.010)
Managerial & Prof. -0.114*** -0.027** 0.001 0.015*** 0.063*** 0.062**
(0.033) (0.011) (0.002) (0.003) (0.019) (0.024)
Age -0.066 -0.008 0.004 0.011 0.034 0.026
(0.047) (0.006) (0.003) (0.008) (0.024) (0.018)
Age squared 0.001 0.000 -0.000 -0.000 -0.000 -0.000
(0.001) (0.000) (0.000) (0.000) (0.000) (0.000)
Two parent HH -0.003 -0.000 0.000 0.000 0.001 0.001
(0.014) (0.002) (0.001) (0.002) (0.007) (0.006)
No. of siblings 0.003 0.000 -0.000 -0.001 -0.002 -0.001
(0.003) (0.000) (0.000) (0.001) (0.002) (0.001)
Tigray -0.008 -0.001 0.000 0.001 0.004 0.003
(0.031) (0.004) (0.002) (0.005) (0.016) (0.012)
Afar 0.037 0.003 -0.003 -0.006 -0.018 -0.013
(0.049) (0.004) (0.004) (0.009) (0.024) (0.016)
Amhara 0.023 0.002 -0.002 -0.004 -0.012 -0.008
(0.023) (0.002) (0.002) (0.004) (0.012) (0.008)
Somalie 0.107*** 0.004* -0.010** -0.020** -0.050*** -0.031***
(0.041) (0.002) (0.005) (0.008) (0.018) (0.010)
Benishangul-Gumuz -0.050 -0.008 0.002* 0.008 0.026 0.022
(0.037) (0.008) (0.001) (0.005) (0.020) (0.018)
SNNPR -0.003 -0.000 0.000 0.000 0.001 0.001
(0.021) (0.003) (0.001) (0.003) (0.011) (0.008)
Gambela 0.507*** -0.100 -0.082** -0.095*** -0.161*** -0.069***
(0.180) (0.084) (0.036) (0.029) (0.029) (0.008)
Harari 0.033 0.003 -0.002 -0.006 -0.016 -0.012
(0.048) (0.004) (0.004) (0.008) (0.023) (0.016)
Addis Ababa -0.029 -0.004 0.001 0.005 0.015 0.012
(0.021) (0.003) (0.001) (0.003) (0.011) (0.008)
Dire Dawa 0.033 0.003 -0.002 -0.006 -0.016 -0.011
(0.047) (0.004) (0.004) (0.008) (0.023) (0.015)

N 3210 3210 3210 3210 3210 3210

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

35
Table A11: Intergenerational Mobility in Occupational Status, Marginal Effects from Ordered
Probit, Daughters (2005)
None Elementary Skilled Services & Machine Managerial
Agricultural Sales operator & &
crafts Professional

Elementary 0.018 -0.000 -0.000 -0.007 -0.006 -0.004

(0.038) (0.001) (0.001) (0.015) (0.013) (0.009)
Skilled Agricultural 0.029 -0.000 -0.001 -0.011 -0.010 -0.007
(0.023) (0.000) (0.001) (0.009) (0.008) (0.005)
Services & Sales 0.004 0.000 -0.000 -0.001 -0.001 -0.001
(0.030) (0.000) (0.001) (0.012) (0.011) (0.008)
Machine operator -0.097*** -0.006** 0.001*** 0.033*** 0.038*** 0.031***
(0.028) (0.003) (0.000) (0.009) (0.012) (0.010)
Managerial & Prof. -0.095* -0.006 0.001*** 0.032** 0.037* 0.030
(0.049) (0.006) (0.000) (0.015) (0.020) (0.019)
Age -0.055 -0.001 0.001 0.020 0.020 0.015
(0.057) (0.001) (0.001) (0.021) (0.021) (0.015)
Age squared 0.001 0.000 -0.000 -0.000 -0.000 -0.000
(0.001) (0.000) (0.000) (0.000) (0.000) (0.000)
Two parent HH -0.012 -0.000 0.000 0.004 0.004 0.003
(0.020) (0.000) (0.000) (0.007) (0.007) (0.005)
No. of siblings -0.001 -0.000 0.000 0.000 0.000 0.000
(0.005) (0.000) (0.000) (0.002) (0.002) (0.001)
Tigray 0.056 0.003 -0.001 -0.020 -0.022 -0.017
(0.038) (0.002) (0.001) (0.014) (0.014) (0.011)
Afar 0.150** -0.000 -0.003 -0.057* -0.052** -0.037***
(0.072) (0.006) (0.003) (0.030) (0.022) (0.013)
Amhara 0.050* 0.003 -0.001* -0.017* -0.019* -0.015*
(0.026) (0.002) (0.000) (0.009) (0.010) (0.008)
Somalie 0.173*** -0.003 -0.004* -0.067*** -0.059*** -0.041***
(0.060) (0.006) (0.002) (0.025) (0.018) (0.011)
Benishangul-Gumuz -0.070 -0.009 -0.000 0.020 0.030 0.030
(0.065) (0.012) (0.001) (0.015) (0.029) (0.032)
SNNPR 0.018 0.001 -0.000 -0.006 -0.007 -0.006
(0.029) (0.002) (0.000) (0.010) (0.012) (0.010)
Gambela 0.363*** -0.039 -0.014* -0.149*** -0.102*** -0.059***
(0.133) (0.037) (0.008) (0.057) (0.025) (0.011)
Harari -0.019 -0.002 0.000 0.006 0.008 0.007
(0.059) (0.006) (0.000) (0.018) (0.024) (0.022)
Addis Ababa 0.093*** 0.003 -0.002*** -0.034*** -0.034*** -0.026***
(0.026) (0.002) (0.001) (0.010) (0.010) (0.008)
Dire Dawa 0.127** 0.001 -0.003 -0.047** -0.045** -0.033***
(0.059) (0.004) (0.002) (0.024) (0.019) (0.012)

N 2283 2283 2283 2283 2283 2283

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

36
Table A12: Intergenerational Mobility in Occupational Status, Marginal Effects from Ordered
Probit, Sons and Daughters (2013)
None Elementary Skilled Services & Machine Managerial
Agricultural Sales operator & &
crafts Professional

Elementary 0.090*** 0.008*** -0.007*** -0.022*** -0.023*** -0.046***

(0.016) (0.001) (0.002) (0.004) (0.004) (0.008)
Skilled Agricultural 0.047*** 0.006*** -0.003*** -0.011*** -0.013*** -0.027***
(0.011) (0.001) (0.001) (0.003) (0.003) (0.006)
Services & Sales -0.000 -0.000 0.000 0.000 0.000 0.000
(0.014) (0.002) (0.000) (0.003) (0.004) (0.009)
Machine operator -0.032* -0.006 0.001* 0.007* 0.009* 0.022*
(0.018) (0.004) (0.000) (0.004) (0.005) (0.013)
Managerial & Prof. -0.137*** -0.039*** -0.009*** 0.019*** 0.039*** 0.127***
(0.015) (0.006) (0.003) (0.002) (0.004) (0.019)
Age -0.118*** -0.017*** 0.005*** 0.026*** 0.032*** 0.073***
(0.026) (0.004) (0.001) (0.006) (0.007) (0.016)
Age squared 0.002*** 0.000*** -0.000*** -0.000*** -0.001*** -0.001***
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Female 0.085*** 0.012*** -0.003*** -0.018*** -0.023*** -0.052***
(0.008) (0.001) (0.001) (0.002) (0.002) (0.005)
Two parent HH -0.018** -0.003** 0.001** 0.004** 0.005** 0.011**
(0.009) (0.001) (0.000) (0.002) (0.002) (0.006)
No. of siblings 0.011*** 0.002*** -0.000*** -0.002*** -0.003*** -0.007***
(0.002) (0.000) (0.000) (0.000) (0.001) (0.001)
Tigrary -0.047*** -0.008** 0.001*** 0.010*** 0.013*** 0.030***
(0.016) (0.003) (0.001) (0.003) (0.004) (0.011)
Afar 0.045 0.004** -0.003 -0.011 -0.012 -0.023*
(0.029) (0.002) (0.003) (0.007) (0.007) (0.014)
Amhara -0.014 -0.002 0.001 0.003 0.004 0.008
(0.013) (0.002) (0.001) (0.003) (0.003) (0.008)
Somalie 0.085*** 0.006*** -0.008** -0.021*** -0.021*** -0.041***
(0.028) (0.001) (0.003) (0.007) (0.007) (0.012)
Benishangul-Gumuz -0.016 -0.002 0.001 0.003 0.004 0.009
(0.030) (0.004) (0.001) (0.007) (0.008) (0.019)
SNNPR -0.012 -0.002 0.001 0.003 0.003 0.007
(0.013) (0.002) (0.001) (0.003) (0.004) (0.008)
Gambela 0.064* 0.005*** -0.005 -0.016* -0.016* -0.032**
(0.034) (0.002) (0.004) (0.009) (0.008) (0.015)
Harari -0.023 -0.003 0.001 0.005 0.006 0.014
(0.027) (0.004) (0.001) (0.006) (0.007) (0.017)
Addis Ababa -0.033*** -0.005*** 0.001** 0.007** 0.009** 0.021***
(0.013) (0.002) (0.001) (0.003) (0.004) (0.008)
Dire Dawa 0.060** 0.005*** -0.005* -0.015** -0.015** -0.030**
(0.029) (0.002) (0.003) (0.007) (0.007) (0.013)

N 7759 7759 7759 7759 7759 7759

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

37
Table A13: Intergenerational Mobility in Occupational Status, Marginal Effects from Ordered
Probit, Sons (2013)
None Elementary Skilled Services & Machine Managerial
Agricultural Sales operator & &
crafts Professional

Elementary 0.082*** 0.013*** -0.006*** -0.014*** -0.028*** -0.046***

(0.020) (0.003) (0.002) (0.004) (0.007) (0.010)
Skilled Agricultural 0.038*** 0.007*** -0.002** -0.006*** -0.013*** -0.024***
(0.014) (0.003) (0.001) (0.002) (0.005) (0.009)
Services & Sales -0.009 -0.002 0.000 0.001 0.003 0.007
(0.017) (0.004) (0.000) (0.003) (0.006) (0.012)
Machine operator -0.050** -0.013** -0.002 0.007*** 0.018** 0.040**
(0.021) (0.006) (0.002) (0.002) (0.008) (0.018)
Managerial & Prof. -0.140*** -0.052*** -0.022*** 0.009*** 0.049*** 0.155***
(0.018) (0.009) (0.006) (0.002) (0.006) (0.027)
Age -0.131*** -0.027*** 0.002 0.019*** 0.046*** 0.092***
(0.033) (0.007) (0.001) (0.005) (0.011) (0.023)
Age squared 0.002*** 0.000*** -0.000 -0.000*** -0.001*** -0.001***
(0.001) (0.000) (0.000) (0.000) (0.000) (0.000)
Two parent HH 0.001 0.000 -0.000 -0.000 -0.000 -0.000
(0.011) (0.002) (0.000) (0.002) (0.004) (0.008)
No. of siblings 0.009*** 0.002*** -0.000 -0.001*** -0.003*** -0.006***
(0.003) (0.001) (0.000) (0.000) (0.001) (0.002)
Tigray -0.047** -0.010** 0.000 0.007** 0.017** 0.033**
(0.019) (0.005) (0.001) (0.003) (0.007) (0.014)
Afar 0.034 0.005 -0.002 -0.006 -0.012 -0.019
(0.034) (0.004) (0.003) (0.006) (0.011) (0.018)
Amhara -0.010 -0.002 0.000 0.002 0.003 0.006
(0.016) (0.003) (0.001) (0.003) (0.006) (0.010)
Somalie 0.064* 0.008** -0.005 -0.011* -0.022* -0.034**
(0.035) (0.003) (0.004) (0.007) (0.011) (0.017)
Benishangul-Gumuz -0.013 -0.002 0.000 0.002 0.004 0.008
(0.038) (0.008) (0.001) (0.006) (0.013) (0.025)
SNNPR -0.018 -0.004 0.001 0.003 0.006 0.012
(0.016) (0.003) (0.001) (0.003) (0.006) (0.011)
Gambela 0.043 0.006 -0.003 -0.007 -0.014 -0.024
(0.047) (0.006) (0.005) (0.008) (0.016) (0.024)
Harari 0.020 0.003 -0.001 -0.003 -0.007 -0.012
(0.036) (0.005) (0.003) (0.006) (0.012) (0.020)
Addis Ababa -0.051*** -0.011*** 0.000 0.007*** 0.018*** 0.036***
(0.016) (0.004) (0.001) (0.002) (0.006) (0.011)
Dire Dawa 0.045 0.006 -0.003 -0.008 -0.015 -0.025
(0.037) (0.004) (0.004) (0.007) (0.012) (0.019)

N 4483 4483 4483 4483 4483 4483

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

38
Table A14: Intergenerational Mobility in Occupational Status, Marginal Effects from Ordered
Probit, Daughters (2013)
None Elementary Skilled Services & Machine Managerial
Agricultural Sales operator & &
crafts Professional

Elementary 0.106*** 0.002 -0.007*** -0.038*** -0.015*** -0.048***

(0.027) (0.002) (0.002) (0.010) (0.004) (0.011)
Skilled Agricultural 0.050*** 0.002** -0.003*** -0.017*** -0.007*** -0.025***
(0.019) (0.001) (0.001) (0.007) (0.003) (0.010)
Services & Sales 0.016 0.001 -0.001 -0.006 -0.002 -0.009
(0.024) (0.002) (0.001) (0.008) (0.003) (0.013)
Machine operator -0.000 -0.000 0.000 0.000 0.000 0.000
(0.032) (0.003) (0.001) (0.011) (0.005) (0.018)
Managerial & Prof. -0.128*** -0.024*** -0.000 0.035*** 0.020*** 0.097***
(0.027) (0.007) (0.002) (0.006) (0.005) (0.025)
Age -0.091** -0.006** 0.004** 0.030** 0.013** 0.049**
(0.043) (0.003) (0.002) (0.014) (0.006) (0.023)
Age squared 0.002** 0.000** -0.000** -0.001** -0.000** -0.001**
(0.001) (0.000) (0.000) (0.000) (0.000) (0.000)
Two parent HH -0.043*** -0.003*** 0.002*** 0.014*** 0.006*** 0.023***
(0.015) (0.001) (0.001) (0.005) (0.002) (0.008)
No. of siblings 0.013*** 0.001*** -0.001*** -0.004*** -0.002*** -0.007***
(0.004) (0.000) (0.000) (0.001) (0.001) (0.002)
Tigray -0.044 -0.004 0.002* 0.014 0.007 0.026
(0.029) (0.003) (0.001) (0.009) (0.004) (0.018)
Afar 0.060 0.002 -0.004 -0.021 -0.008 -0.028
(0.052) (0.001) (0.004) (0.019) (0.007) (0.023)
Amhara -0.018 -0.001 0.001 0.006 0.003 0.010
(0.022) (0.002) (0.001) (0.007) (0.003) (0.012)
Somalie 0.113** -0.000 -0.008** -0.040** -0.015** -0.049***
(0.046) (0.003) (0.004) (0.017) (0.006) (0.017)
Benishangul-Gumuz -0.020 -0.002 0.001 0.007 0.003 0.011
(0.048) (0.004) (0.002) (0.016) (0.007) (0.028)
SNNPR 0.001 0.000 -0.000 -0.000 -0.000 -0.000
(0.023) (0.002) (0.001) (0.008) (0.003) (0.013)
Gambela 0.091* 0.001 -0.006 -0.032* -0.012* -0.041**
(0.049) (0.002) (0.004) (0.018) (0.006) (0.019)
Harari -0.071* -0.008 0.002** 0.022* 0.011* 0.044
(0.042) (0.007) (0.001) (0.012) (0.006) (0.029)
Addis Ababa 0.001 0.000 -0.000 -0.000 -0.000 -0.000
(0.022) (0.001) (0.001) (0.008) (0.003) (0.012)
Dire Dawa 0.085* 0.001 -0.006 -0.030* -0.012* -0.039**
(0.045) (0.002) (0.004) (0.016) (0.006) (0.018)

N 3276 3276 3276 3276 3276 3276

Robust standard errors in parentheses, with household clusters
*** p<0.01, ** p<0.05, * p<0.1

39