ArticlePDF Available

Abstract

Parental entrepreneurship is a strong, probably the strongest, determinant of own entrepreneurship. We explore the origins of this intergenerational association in entrepreneurship. In particular, we identify the separate effects of pre- and post-birth factors (nature and nurture), by using a unique dataset of Swedish adoptees. Its unique characteristic is that it not only includes data on occupational status for the adoptees and their adoptive parents, but also for their biological parents. Moreover, we use comparable data on entrepreneurship for a large, representative sample of the Swedish population. Based on the latter sample, and consistent with previous findings, we show that parental entrepreneurship increases the probability of children's entrepreneurship by about 60%. We further show that for adoptees, both biological and adoptive parents make significant contributions. These effects, however, are quite different in size. The effect of post-bir th factors (adoptive parents) is approximately twice as large as the effect of pre-birth factors (biological parents). The sum of these two effects for adopted children is almost identical to the intergenerational transmission of entrepreneurship for own-birth children. We explore several candidate explanations for this important post-birth effect and present suggestive evidence in favor of role modeling.
NORC at the University of Chicago
The University of Chicago
Why Do Entrepreneurial Parents Have Entrepreneurial Children?
Author(s): Matthew J. Lindquist, Joeri Sol, and Mirjam Van Praag
Source:
Journal of Labor Economics,
Vol. 33, No. 2 (April 2015), pp. 269-296
Published by: The University of Chicago Press on behalf of the Society of Labor Economists and
the NORC at the University of Chicago
Stable URL: http://www.jstor.org/stable/10.1086/678493 .
Accessed: 04/03/2015 05:02
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
The University of Chicago Press, Society of Labor Economists, NORC at the University of Chicago, The
University of Chicago are collaborating with JSTOR to digitize, preserve and extend access to Journal of
Labor Economics.
http://www.jstor.org
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
Why Do Entrepreneurial Parents
Have Entrepreneurial Children?
Matthew J. Lindquist, Swedish Institute for Social Research,
Stockholm University
Joeri Sol, Amsterdam School of Economics, University of Amsterdam
Mirjam Van Praag, Amsterdam School of Economics,
University of Amsterdam
We explore the origins of the intergenerational association in en-
trepreneurship using Swedish adoption data that allow us to quan-
tify the relative importance of prebirth and postbirth factors. We
find that parental entrepreneurship increases the probability of chil-
dren’s entrepreneurship by about 60%. For adoptees, both biolog-
ical and adoptive parents make significant contributions to this as-
sociation. These contributions, however, are quite different in size.
Postbirth factors account for twice as much as prebirth factors in
our decomposition of the intergenerational association in entrepre-
neurship. We investigate several candidate explanations for this large
postbirth factor and present suggestive evidence in favor of role mod-
eling.
I. Introduction
Why do some people become entrepreneurs but not others? The entre-
preneurship literature assertsa number of factors that influence this choice.
We thank David Cesarini, Nicos Nicolaou, Erik Plug, and seminar partici-
pants at ABEE ðAmsterdamÞ, ISEG ðTechnical University of LisbonÞ, IZA ðBonnÞ,
Linnaeus University, Norwegian Business School, IFN ðStockholmÞ, SOLE 2013,
and the Uddeville Conference on Innovation and Industrial Economics ðUniversity
[Journal of Labor Economics, 2015, vol. 33, no. 2]
© 2015 by The University of Chicago. All rights reserved. 0734-306X/2015/3302-0003$10.00
Submitted January 22, 2013; Accepted October 21, 2013; Electronically published January 15, 2015
269
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
The most prominent among these factors is parental entrepreneurship. Hav-
ing an entrepreneur for a parent increases the probability that a child ends
up as an entrepreneur by 30%200% ðDunn and Holtz-Eakin 2000; Arum
and Mueller 2004; Sørensen 2007; Colombier and Masclet 2008; Andersson
and Hammarstedt 2010, 2011Þ.
1
While this stylized fact is widely accepted, there is no consensus con-
cerning the origins of this intergenerational transfer of entrepreneurship.
Thus far, none of the studies that explore various environmental explana-
tions control for the transfer of genes from parent to child. This may bias
their results given that recent twin studies find a large genetic component
in the choice to become an entrepreneur ðNicolaou et al. 2008; Zhang
et al. 2009; Nicolaou and Shane 2010, 2011Þ.
2
On the other hand, twin
studies do not address environmental determinants of entrepreneurship.
We contribute to this literature by decomposing the intergenerational
transmission of entrepreneurship into prebirth factors ðgenes, prenatal and
perinatal environmentÞand postbirth factors, using Swedish adoption data
that include information on entrepreneurship for all four parents of adopted
children.
3
This allows us togauge the relative importance of nature and nur-
ture in the reproduction of entrepreneurship from one generation to the
next. We also run comparable exercises for a large, representative sample
of nonadoptees.
We find that having an entrepreneur for a parent increases the proba-
bility that own-birth children become entrepreneurs by 60%. The size of
this association is consistent with earlier studies.
4
Our decomposition
exercise with adopted children reveals that both biological and adoptive
parents make significant contributions. However, the impact of postbirth
factors ði.e., the influence of adoptive parentsÞis approximately twice as
large as the impact of prebirth factors ði.e., the influence of biological par-
of FaroÞfor their useful comments. Matthew Lindquist gratefully acknowledges
financial support from the Swedish Council for Working Life and Social Research
ðFASÞ. Contact the author at matthew.lindquist@sofi.su.se. Information concern-
ing access to the data used in this article is available as supplementary material
online.
1
Andersson and Hammarstedt ð2010Þand Laspita et al. ð2012Þreport correla-
tions across three generations.
2
Genome-wide studies have thus far been unable to find specific genetic mark-
ers affecting entrepreneurship ðsee Koellinger et al. 2010; Van der Loos et al. 2013Þ.
3
The four-parent adoption methodology we use was pioneered by Bjo
¨rklund,
Lindahl, and Plug ð2006Þ, who studied the intergenerational transmission of in-
come and education. Other recent papers using the Bjo
¨rklund et al. ð2006Þmeth-
odology and similar data include Hjalmarsson and Lindquist ð2013Þ, who study
intergenerational correlations in crime, and Cesarini, Johannesson, and Oskarsson
ð2014Þ, who study voting behavior.
4
For Sweden, see Andersson and Hammarstedt ð2010, 2011Þ.
270 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
entsÞ. This difference is significant and is robust to several definitions of
“entrepreneur.”
We then go on to examine several candidate explanations of this large
postbirth effect. After controlling for the transfer of genes, we find little
evidence in support of inheritance of the family business, access to cheap
capital, and parent-offspring similarities in choice of industry as expla-
nations of our estimated postbirth effect. But we do find large same-sex
associations in entrepreneurship, which we argue is indirect evidence in
favor of role modeling.
Our results suggest that there is scope to sway people in the direction
of entrepreneurship, either by public policies or within the education sys-
tem, as postbirth factors play an important role in determining this occu-
pational choice. This may be desirable because of the positive effects of
entrepreneurship on job creation and innovation ðsee, e.g., Birch 1979;
Acs 1999; and Van Praag and Versloot 2007Þ. In addition, our findings
suggest that a further exploration of the effects and determinants of role
models for entrepreneurship may be fruitful; Bosma et al. ð2012Þtake a first
step in this direction.
In the following section, we present the four-parent adoption meth-
odology used to estimate prebirth and postbirth effects. In Section III, we
describe the Swedish adoption procedure and the data. Our baseline re-
sults are presented in Section IV, which is followed by a sensitivity anal-
ysis in Section V. Section VI examines the plausibility of several nurture
explanations for the postbirth transmission of entrepreneurship. Sec-
tion VII concludes.
II. Empirical Methodology
We begin by regressing our measure of entrepreneurship for a non-
adopted child ðEbc
iÞborn into family ion our measure of entrepreneurship
for the parents ðEbp
iÞ, in order to obtain an estimate of the intergenera-
tional association in entrepreneurship, b1:
Ebc
i5b01b1Ebp
i1vbc
i:ð1Þ
Previous research has found strong intergenerational associations in en-
trepreneurship. Thus, we expect b1to be significantly positive.
The key question addressed in this paper is why we find an intergen-
erational association in entrepreneurship. Potential mechanisms can be
placed into two broad categories that we label prebirth and postbirth
factors. The Bjo
¨rklund et al. ð2006Þmethodology makes use of adoption
data with information on all four parents of adopted children in order to
estimate these prebirth and postbirth factors directly. Thus, we can assess
the relative importance of prebirth and postbirth factors for generating the
observed intergenerational association in entrepreneurship, b1.
Why Do Entrepreneurial Parents Have Entrepreneurial Children? 271
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
Using our adoption data, we estimate the following linear regression
model:
Eac
j5a01a1Ebp
i1a2Eap
j1vac
j:ð2Þ
We regress our measure of entrepreneurship for a child ðEac
jÞborn into
family ibut then adopted by family jon the entrepreneurship status of
both sets of parents. Under certain assumptions ðspelled out and tested in
Sec. VÞ, the coefficient on biological parents’ entrepreneurship ðEbp
iÞ,a1,is
a consistent estimate of prebirth factors, and the coefficient on adoptive
parents’ entrepreneurship ðEap
jÞ,a2, is a consistent estimate of postbirth fac-
tors. An additional set of assumptions is required to generalize our esti-
mates of the relative importance of prebirth and postbirth effects beyond
our sample of adopted children ðthese are also spelled out and tested in
Sec. VÞ.
We estimate equations ð1Þand ð2Þusing ordinary least squares ðOLSÞ.
The terms vbc
iand vac
jin equations ð1Þand ð2Þare the OLS regression error
terms. Since entrepreneurial activities are bound to vary at different stages
of the lifecycle, and since our data are partially censored by age ðmore on
this in Sec. IIIÞ, we include year of birth dummies for children and each
parent in our regressions. We also include a county of residence dummy
for the child at a young age ðin 1965Þand a gender dummy.
III. Institutions and Data
A. Adoptions in Sweden
The children we study were born during the 1940s, 1950s, and 1960s. At
this time, adoption of small children was done anonymously and private
adoptions were not allowed.
5
The formal nature of the adoption process
implies that the biological mother is typically identified. Moreover, social
workers attempted to identify the biological father. The identities of the
biological parents ðwhen knownÞand adopting parents were recorded in
the court decision and kept in the census records.
There were very few explicit legal requirements concerning who was
eligible to adopt. One had to be at least 25 years old and free of tuber-
culosis or sexually transmitted diseases. Informally, the social authorities
used the following rules and recommendations. The adopting family must
have adequate housing and a steady income, should be married, and the
adopting mother should be able to stay at home while the child was small.
In practice, as we shall see in the next section, adoptive parents tend to be
somewhat positively selected in terms of education and income, but not as
much as they are today.
Children were not placed into their new families at random. Whenever
possible, the social authorities wanted to match children based on their
5
Adoption by relatives was allowed but was very rare ðNordlo
¨f 2001Þ.
272 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
biological parents’ intellectual capabilities, talents, and physical appear-
ance. Their hope was that parents would “recognize” themselves in their
adopted child, and vice versa. However, after conditioning on observable
characteristics ðage, marital status, income, and educationÞ, the evaluation
literature ðreviewed in Bohman ½1970Þ finds no evidence that the social
authorities were able to predict which parents would provide the most
stable homes and the needed emotional environment.
There were four initial placement possibilities for newborn children: a
special nursery, a home for unwed mothers, temporary foster care, or di-
rectly with the adopting family. Children with visible handicaps, severe
health problems, or whose parents suffered from severe cases of mental
illness, alcoholism, or criminality were not always put up for adoption.
This means that those children who were put up for adoption were a pos-
itively selected group from a somewhat negatively selected pool of chil-
dren. These two countervailing effects make adopted children look quite
“average.” The sample of adoptees studied by Bohman ð1970Þ, for exam-
ple, had the same average birth weight and health outcomes ðat ages 10
11Þas their nonadopted peers in school.
B. Data
Our data were assembled as follows. Statistics Sweden began by draw-
ing a 25% random sample from Sweden’s Multigenerational Register,
which includes all persons born from 1932 onward who have lived in
Sweden at any time since 1961. We call these our index persons. We also
asked Statistics Sweden to identify all individuals adopted by at least one
parent in Sweden. Adoptive as well as biological mothers and fathers
and siblings of each adopted individual as well as each index person were
matched onto the sample. This gives us a sample of more than 7.5 million
individuals.
We can identify 85% of the biological mothers and 43% of the biolog-
ical fathers of adoptees born in Sweden.
6
Knowing the identities of the
biological parents of adopted children is quite unique to these Swedish
data and is the key to our empirical strategy.
7
Register data concerning
entrepreneurship, income, education, industry, place of residence, year of
birth and death, and year of immigration and/or emigration were then
matched to this sample using the unique identification number that each
Swedish resident is assigned.
We use this data set to create two samples; an adoptive sample and a
nonadoptive sample. Table 1 lists the sample restrictions that we impose
and the corresponding impacts on sample sizes. Our raw data set contains
6
In Sec. V, we address the issue of missing fathers.
7
The only other ðnon-SwedishÞdata set that we are aware of that has infor-
mation on all four parents is a new Taiwanese data set. Tsou, Liu, and Hammit
ð2012Þuse these data to study the intergenerational transmission of education in
Taiwan.
Why Do Entrepreneurial Parents Have Entrepreneurial Children? 273
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
7,408,044 nonadopted individuals and 143,490 individuals adopted by at
least one parent. Restricting our adoption sample to those adopted by
both parents reduces it to 91,447 individuals. We restrict our nonadoption
sample to the 2,448,405 index persons ði.e., those in the original 25% ran-
dom sampleÞ.
For both the adoption and nonadoption samples, we impose the fol-
lowing additional restrictions. Since we only have information on the bi-
ological parents of adopted children who were born in Sweden, we elim-
inate all children who were born abroad. We also omit all children with at
least one parent born before 1920. These individuals are most likely too
old to have a chance to show up in our data on entrepreneurship, which
start in 1985. Children who died or emigrated from Sweden before 1985
are dropped from the sample, as they cannot show up in our data on en-
trepreneurship. Likewise, we omit any child who had at least one parent
ðbiological or adoptiveÞdie or emigrate from Sweden before 1985. We
also eliminate individuals born in 1970 or later. We choose this year as our
cutoff because ðiÞthe birth control pill was approved in 1965 and ðiiÞle-
gal abortions were gradually introduced in Sweden from 1965 to 1975. As
a result of these medical and legal changes, the number of Sweden-born
adoptees fell dramatically, and the biological parents of adopted-away
children became more negatively selected over time. Finally, we drop all
children for which we cannot identify both biological parents. Our base-
Table 1
Sample Restrictions
Restriction Nonadoptees Adoptees
Changes in
Nonadoptees
Changes in
Adoptees
ð1ÞAll individuals adopted by at
least one parent 143,490
ð2ÞKeep only those adopted by both
parents ðand both are identifiedÞ91,447 252,043
ð3ÞAll nonadopted individuals 7,408,044
ð4ÞAll index nonadopted individuals 2,448,405 24,959,639
ð5ÞExclude children born abroad 1,987,817 46,807 2460,588 244,640
ð6ÞExclude children with one or
more parents born before 1920 1,524,512 20,720 2463,305 226,087
ð7ÞExclude all children who died
or emigrated from Sweden
before 1985 1,491,342 20,540 233,170 2180
ð8ÞExclude all children with one or
more parents who died or emi-
grated from Sweden before 1985 1,437,623 17,639 253,791 22,901
ð9ÞExclude children born 1970
or later 449,750 10,000 2987,873 27,639
ð10ÞExclude children with
biological mothers missing 422,389 8,513 227,361 21,487
ð11ÞKeep those for whom both
biological parents are identified 412,183 3,941 210,206 24,572
274 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
line sample includes 412,183 nonadopted children and 3,941 adopted chil-
dren. Among the adoptees, we have 2,149 sons and 1,792 daughters.
Consistent with the Swedish tax authorities, we define individuals as
entrepreneurs when they derive the majority of their taxable labor in-
come from a business that they own in full or in part. For the years 1985
2008, we have information on ðsole and sharedÞbusiness ownership for
unincorporated enterprises. For the years 19932008,
8
we also know if a
person received the majority of his/her taxable labor income from an in-
corporated enterprise owned in part or in full by himself/herself ðand pos-
sibly employing personnelÞ. An incorporated business in our data is a pri-
vately owned ði.e., nonlistedÞlimited liability stock company.
We use these two pieces of information ðbusiness owners of an unin-
corporated or incorporated enterpriseÞin order to categorize people as en-
trepreneurs.
9
Our baseline measure of “entrepreneurship,” Entrepreneur,
is an extensive margin variable equal to one if the individual ever catego-
rizes as entrepreneur and zero otherwise. We will use four stricter defini-
tions of entrepreneurship as well, and these will be defined in Section IV.
We use tax register data for the years 19682007 for each person in our
sample to create a measure of income, Income, based on the broadest def-
inition of income available to us. It is equal to the log of an individual’s
average pretax net total factor income, measured in real terms, over all
available years, where the very few zeros are treated as missing. For two of
the stricter definitions of entrepreneurship, we create an auxiliary income
measure Entrepreneurial Income, only based on those years an individual
is labeled an Entrepreneur. Education is measured in seven levels, rep-
resented as dummies in the regression equations and compressed into
Years of Schooling for the descriptive statistics.
10
We also have information
concerning the Industry in which people work, comprised of 42 two-digit
SNI industry codes.
Descriptive statistics for the baseline sample are shown in table 2.
Panel A shows means and standard deviations, whereas panel B shows
10
Most of this information has been taken from Sweden’s national education
register for the year 1990. If education was missing in this primary source, then
secondary sources were searched. This was done in the following order: the na-
tional education registers for 1993, 1996, and 1999, and, finally, the 1970 Censuses.
8
We do not have information before 1993 on those working in their own in-
corporated enterprise. This implies that we are underestimating the true extent of
entrepreneurship for the years 198592. For the years 19932001, roughly 2%of
the sample is in this position. This might be approximately true for the years 1985
92 as well.
9
Note that farmers are included in Statistics Sweden’s definition of business
owners, since farms are typically run as companies ðeither incorporated or unin-
corporatedÞ. In 2004, Statistics Sweden changed their routines for collecting in-
formation on business ownership as well as its definition. Since then, it includes
business owners who report zero profits or even losses.
Why Do Entrepreneurial Parents Have Entrepreneurial Children? 275
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
Table 2
Descriptive Statistics
Panel A: Variables with Means and Standard Deviations
Own-Birth Children Adopted Children
Mean SD Mean SD
Daughters:
Entrepreneur .14 .35 .14 .35
Income 11.70 .46 11.64 .48
Years of schooling 12.40 2.23 12.22 2.00
Age in 1985 24.9 6.44 22.3 4.28
Sons:
Entrepreneur .25 .43 .23 .42
Income 12.02 .49 11.90 .51
Years of schooling 11.96 2.33 11.67 1.99
Age in 1985 24.9 6.45 22.5 4.44
Birth Parents
Own-Birth Children Adopted Children
Mean SD Mean SD
Mothers:
Entrepreneur .15 .36 .10 .30
Income 11.57 .57 11.59 .51
Years of schooling 9.52 2.72 9.37 2.46
Age in 1985 50.7 7.38 45.5 6.73
Fathers:
Entrepreneur .26 .44 .20 .40
Income 12.21 .45 11.96 .48
Years of schooling 9.69 2.98 9.17 2.52
Age in 1985 53.5 7.38 48.6 7.24
Adoptive Parents
Mean SD
Mothers:
Entrepreneur
Income 11.54 .56
Years of schooling 9.67 2.83
Age in 1985 53.6 5.97
Fathers:
Entrepreneur .27 .45
Income 12.33 .43
Years of schooling 10.20 3.10
Age in 1985 55.9 5.77
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
correlations of interest. The incidence of entrepreneurship is quite similar
for own-birth and adopted children. The share of Entrepreneur is 14% for
both adopted and own-birth daughters and 25% and 23% for own-birth
and adopted sons, respectively. Adopted children have slightly lower lev-
els of income and education than own-birth children. They are roughly
2.5 years younger, too, which highlights the importance of controlling for
year of birth in the empirical specifications.
Adoptive parents and parents with own-birth children have nearly iden-
tical values for Entrepreneur, but the values are lower for birth parents
who put a child up for adoption. The share of Entrepreneurs among birth
mothers who adopt away their child is 10%, while for adoptive mothers
and mothers with own-birth children the share is 15%. The share of En-
trepreneurs among birth fathers who adopt away their child is 20%, while
for adoptive fathers and fathers with own-birth children the shares are
27% and 26%, respectively. Note also that the adoptive parents are several
years older ðon averageÞthan the parents with own-birth children,who are,
in turn, several years older than the birth parents who adopt away their
child. Adoptive fathers have a full year of schooling more than birth fathers
of adopted children, and their incomes are also higher.
Panel B of table 2 shows that the likelihood of being an entrepreneur is
higher for children with parents who were entrepreneurs. For own-birth
children the correlation between their own entrepreneur status and their
parents’ status is between 11% and 12% and similar for both parents.
For adopted children, the correlation between parental and own entre-
preneurship is larger for adoptive parents than for the biological parents
and, again, not much different for fathers and mothers.
Table 2 (Continued)
Panel B: Raw Correlations
Own-Birth Children
ð1Þð2Þð3Þ
ð1ÞRespondent entrepreneur 1.000
ð2ÞBiological father entrepreneur .120** 1.00
ð3ÞBiological mother entrepreneur .112** .381** 1.00
Adopted Children
ð4Þð5Þð6Þð7Þð8Þ
ð4ÞRespondent entrepreneur 1.000
ð5ÞBiological father entrepreneur .045** 1.000
ð6ÞBiological mother entrepreneur .025 .065** 1.000
ð7ÞAdoptive father entrepreneur .090** .002 .009 1.000
ð8ÞAdoptive mother entrepreneur .079** .005 2.027 .416** 1.000
NOTE.Our data on self-employment start in 1985.
** Significant at the 5%level.
Why Do Entrepreneurial Parents Have Entrepreneurial Children? 277
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
Panel B of table 2 further shows that the entrepreneurship status of fa-
thers and mothers is highly correlated, around 40%. This may be due to
assortative mating or of setting up firms jointly. For the biological par-
ents of adopted children, this correlation is much lower, but significant
ð6.5%Þ, consistent with the fact that these parents have often been to-
gether for shorter periods of time. Finally, it is also notable that the en-
trepreneur status of biological and adoptive parents is unrelated. Children
of entrepreneurial parents are not more or less likely to be placed with
adoptive parents who are also entrepreneurs.
IV. Results: The Relative Importance of Nature and Nurture
Table 3 presents the results of the baseline regressions for the sample of
own-birth children in the top panel and for the sample of adopted children
in the middle panel. The dependent variable in all equations is a dummy
that is one for “children” who are or have been entrepreneurs and zero
otherwise ðEntrepreneurÞ. The explanatory variable of interest in column 1
is a dummy ðdefined likewiseÞthat indicates whether either parent is or
has been an entrepreneur. In column 2, only the fathers’ entrepreneurial
experience is considered, whereas column 3 does so for the mothers. Col-
umn 4 includes all parents in one regression. Besides the coefficients that
indicate the effect of these variables on the child’s likelihood of entrepre-
neurship in terms of percentage points, the table also shows the effect in
percentages ðin parenthesesÞ. All regressions include year of birth dum-
mies for children and parents and gender and county of residence ðin or
around 1965Þdummies for the child ðcoefficients are not reportedÞ. The
bottom panel provides F-tests on the equality of prebirth and postbirth
factors and a comparison of the estimates between the two samples.
First, consider the intergenerational transfer of entrepreneurship for
own-birth children. Taken as a whole, the likelihood to experience a spell
of entrepreneurship significantly increases by having an entrepreneur for a
parent. Parental entrepreneurship raises the probability that the offspring
will experience entrepreneurship by 11.6 percentage points ðor 61%; see
col. 1Þ. Entrepreneurial experience of fathers and mothers is equally im-
portant ðsee cols. 2 and 3Þ. The effects of parental entrepreneurship when
fathers and mothers are included separately are similar in magnitude to
column 1, but they decrease when both are included simultaneously ðcol. 4Þ.
The drop in the coefficients indicates that the entrepreneurship likelihood
of the partners is correlated positively ðsee panel B of table 2Þ. Still, either
parent’s experience with entrepreneurship significantly increases the prob-
ability that a child ends up as an entrepreneur by roughly 45%. Thus, the
child is almost twice as likely to become an entrepreneur when both par-
ents have experienced entrepreneurship.
Next, in the middle panel, we show the results for adopted children.
The entrepreneurial experience of both biological parents and adoptive
parents significantly raise the probability that an adoptee will be observed
278 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
Table 3
Baseline Results, Ever Been an Entrepreneur
ð1Þð2Þð3Þð4Þ
Own-birth children:
Entrepreneur biological parent .116***
ð.001Þ
½61%
Entrepreneur biological father .115 .088
ð.002Þð.002Þ
½61%½46%
Entrepreneur biological mother .128 .089
ð.002Þð.002
½67%½47%
Adopted children:
Entrepreneur biological parent .037**
ð.015Þ
½19%
Entrepreneur biological father .042** .043**
ð.017Þð.017Þ
½22%½23%
Entrepreneur biological mother .034 .030
ð.023Þð.023Þ
½18%½16%
Entrepreneur adoptive parent .084
ð.014Þ
½44%
Entrepreneur adoptive father .087 .069
ð.015Þð.016Þ
½46%½36%
Entrepreneur adoptive mother .093 .065
ð.019Þð.021Þ
½49%½34%
Sum of biological and adoptive parent
coefficients .121
ð.021Þ
.080.161
Sum of biological and adoptive father
coefficients .129 .112
ð.022Þð.023Þ
.085.173 .066.158
Sum of biological and adoptive mother
coefficients .128 .095
ð.030Þð.032Þ
.069.186 .033.156
F-test of differential effects of biological
and adoptive parents 4.97** 4.25** 4.10** 3.13*
Year of birth dummies child Yes Yes Yes Yes
Year of birth dummies parents Yes Yes Yes Yes
County of residence in 1965 Yes Yes Yes Yes
No. of own-birth observations 412,183 412,183 412,183 412,183
No. of adoptive observations 3,941 3,941 3,941 3,941
NOTE.Estimates are from the OLS regressions that include a gender dummy for the child. Robust
standard errors are in parentheses. The numbers in brackets ðshown as %Þconvert the estimated coeffi-
cients from percentage points to percentages relative to the prevalence of entrepreneurship in the relevant
sample of children ð19%for both samples; see table 2Þ.
* Significant at the 10%level.
** Significant at the 5%level.
*** Significant at the 1%level.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
as an entrepreneur. Biological parents’ entrepreneurship increases the off-
spring’s tendency to become entrepreneurs by approximately 20%. The
estimates are consistent across regressions, and they are significant at the
5% level except for the biological mothers’ experience.
11
The impact of
biological fathers and mothers is almost identical in size, which makes it
unlikely that the in utero environment influences this occupational choice.
The estimates for the effect of adoptive parents’ entrepreneurship indi-
cate that postbirth factors contribute about twice as much to the inter-
generational transfer of entrepreneurship as prebirth factors do. A history
with entrepreneurship among adoptive parent raises the likelihood that an
adoptee will be observed as an entrepreneur by 45%.
12
The F-tests in the
bottom panel show that the effect of the adoptive parent is significantly
larger than the effect of the biological parent ðfor col. 1, the p-value is .026Þ.
Finally, the bottom panel of table 3 presents the sum of biological and
adoptive parent coefficients and the confidence intervals of these sums.
The sum of the estimated prebirth and postbirth factors for adoptees is
remarkably similar to the intergenerational association for own-birth chil-
dren. That is, we find no evidence that being adopted has an impact on the
intergenerational transfer of entrepreneurship.
13
In summary, table 3 gives us the following insights. First, there is a
significant parent-child transmission of entrepreneurship. The likelihood
that someone is observed as an entrepreneur increases by about 45%65%
if the parent is or was an entrepreneur. This finding is consistent with
earlier studies ðe.g., Sørensen ½2007found an association of similar size in
DenmarkÞ. Second, this transmission is the same for entrepreneurial fa-
thers and mothers. Third, when considering adopted children, we see that
both biological and adoptive parents have a significant contribution. Ap-
proximately one-third of the intergenerational association in entrepre-
11
It may well be that the difference in significance is due to the lower occur-
rence of entrepreneurship among mothers than fathers ðsee panel A of table 2Þ.In
Sec. V.B, we show indeed that the effect of biological mothers turns significant
once we extend our sample size to those adoptees for whom the information on the
biological fathers is missing.
12
Again, we see a drop in the coefficients when both adoptive fathers and
mothers are included. This drop is not observed for the biological parents of
adoptees: for them the entrepreneurship outcome of the father and the mother is
much less strongly correlated ðsee panel B of table 2Þ.
13
As a robustness check, we test whether being adopted has an impact on the
transfer of entrepreneurship in the pooled sample of own-birth children and
adoptees. We regress entrepreneurship of children on entrepreneurship of their
prebirth and postbirth parents ðwhich only differ for adopteesÞand a cross-term of
entrepreneurship by either postbirth parent with an indicator for being adopted.
The coefficient of this cross-term should be zero when the total effect of parental
entrepreneurship is the same for adoptees and own-birth children. The results
mimic those in table 3, col. 1, and the estimate for the cross-term is indeed
insignificant and the size ð0.005Þis identical to the difference between the estimates
in the top and bottom panel of col. 1 in table 3.
280 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
neurship is accounted for by prebirth factors, whereas two-thirds of the
association can be attributed to postbirth factors. The difference between
the estimated size of prebirth and postbirth factors is significant. Fourth,
being adopted has no detectable impact on the intergenerational transfer
of entrepreneurship.
The definition of entrepreneurship used in table 3 implies that obser-
vations are counted as entrepreneurs as soon as they have owned a busi-
ness no matter how long, serious, or successful their spell of entrepre-
neurship was. As a check, we reestimated column 1 of table 3 using four
stricter definitions of entrepreneurship. The first definition is equal to one
whenever someone has been the owner of an incorporated firm and zero
otherwise. It leaves us with 40% of the entrepreneurs in the base sample.
The second definition restricts the original definition in terms of duration
and equals one if someone has been an entrepreneur for at least 3 years
and is zero otherwise, thereby cutting off 30% from the original sample of
entrepreneurs. Third, we cut off the same percentage, that is, the bottom
30%, but in this case by using the entrepreneurial income distribution.
The fourth definition combines the second and third; it is one for en-
trepreneurs who have been in business for at least 3 years and earned in-
comes belonging to the upper 70% of the income distribution. This is the
case for 52% of the entrepreneurs. Both when applying the stricter defi-
nition of entrepreneurship to the children only and when applying it sym-
metrically to both the child and their parents, we find estimates resonating
the results in table 3.
14
Thus, our main result holds regardless of the defi-
nition of “who is an entrepreneur,” which is a much debated matter in the
entrepreneurship literature ðe.g., Parker 2009Þ.
V. Internal and External Validity
A. Internal Validity
We run sensitivity analyses for two possible problems: ðiÞnonrandom
assignment of adoptees to their adoptive parents;
15
ðiiÞage at adoption.
16
14
One notable exception are the results for incorporated, where the parental
transmission is more than three times as large when the stricter definition of
entrepreneurship is also applied to parents. This seems to manifest itself through
postbirth factors only.
15
Please note that, even if the assignment of adoptees was fully random, our
decomposition exercise may not identify causal effects, as the parent’s entrepre-
neurial status may be correlated with unobserved characteristics.
16
We have also tested for nonlinear effects, i.e., gene-environment interaction,
that could influence our estimates of prebirth and postbirth factors. We reesti-
mated the baseline regressions for adoptees while including an interaction term
that is equal to one when both the biological and the adoptive parent are entre-
preneurs. The coefficients belonging to the interacted terms are all zero, and the
estimated main effects are similar to the baseline results, implying no significant
nonlinear effects.
Why Do Entrepreneurial Parents Have Entrepreneurial Children? 281
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
In case nonrandom placement of children to adoptive families occurs,
our estimates of the prebirth and postbirth factors might become biased,
as we may pick up postbirth effects in our prebirth estimates, and vice
versa. Fortunately, we can test the sensitivity of our results to the violation
of this assumption ðalbeit only based on observable characteristicsÞ. Non-
random placement with respect to entrepreneurship status does not seem
to be an issue, as the biological and adoptive parents’ entrepreneurship
status are uncorrelated ðsee panel B pf table 2Þ, even after controlling for
other observable characteristics.
17
We further check the validity of this
suggestion by assessing to what extent our baseline results ðrow 1 of ta-
ble 4Þfor the adoptive parents change upon the exclusion of the biological
parents’ entrepreneurship status ðrow 2 of table 4Þ. We perform the same
test on the estimates of the coefficients of the biological parents’ entre-
preneurship status by excluding the adoptive parents’ entrepreneurship
status controls from the regression ðrow 3 of table 4Þ. The resulting co-
efficients are very similar to the corresponding estimates of the baseline
equation.
Finally, from Bjorklund et al. ð2006Þ, we know that the correlation be-
tween adoptive and biological parents is not zero when considering edu-
cation level or income. Therefore, we check whether our baseline esti-
mates for adopted children change upon the inclusion of controls for the
education and income levels of the biological parents ðrow 4Þand the
adoptive parents ðrow 5Þ, respectively. The results hardly change. All in
all, our sensitivity analysis shows no evidence of nonrandom placement in
terms of entrepreneurship nor does nonrandom placement in terms of
education and income affect our results.
If late placements affect outcomes in a manner that is different from
direct placements in the adopting family, then we may be underestimating
the postbirth effects of these children’s adopting parents or overestimat-
ing the prebirth effects of parents who adopt away. A nontrivial share of
our sample of adopted children may have experienced postdelivery place-
ments after more than 12 months. Using nationwide data drawn from the
same sources as our own data, Bjo
¨rklund et al. ð2006Þreport that 80%
of adoptees born in the 1960s were living with their new families before
age 1. Using the census data from 1960, 1965, and 1970, our analysis shows
17
We estimate regressions of adoptive parent entrepreneurship status on the
biological parents’ entrepreneurship status in three specifications: ðiÞno controls,
ðiiÞthe same birth year and county controls as in table 3, and ðiiiÞextra controls for
education levels ðdummiesÞand incomes of the biological and adoptive parents.
The resulting coefficients are zero, suggesting that nonrandom placement with
respect to entrepreneurship status is not an issue.
282 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
Table 4
Sensitivity Analysis
Parent Father Mother
Biological Adoptive Biological Adoptive Biological Adoptive
Adopted children:
ð1ÞBaseline results ðfrom table 3; n53,941Þ.037**
ð.015Þ
.084***
ð.014Þ
.042**
ð.017Þ
.087***
ð.015Þ
.034
ð.023Þ
.093***
ð.019Þ
Test for nonrandom assignment:
ð2ÞExclude biological parent entrepreneurship and characteristics .084***
ð.014Þ
.089***
ð.015Þ
.091***
ð.019Þ
ð3ÞExclude adoptive parent entrepreneurship and characteristics .036**
ð.015Þ
.041**
ð.016Þ
.030
ð.023Þ
ð4ÞInclude biological parent education and income
a
.035**
ð.015Þ
.084***
ð.014Þ
.040**
ð.017Þ
.088***
ð.015Þ
.033
ð.023Þ
.093***
ð.019Þ
ð5ÞInclude adoptive parent education and income
a
.037**
ð.015Þ
.083***
ð.014Þ
.041**
ð.016Þ
.089***
ð.015Þ
.034
ð.023Þ
.093***
ð.019Þ
Missing biological fathers:
ð6ÞInclude all with identified biological mothers ðn58,513Þ.037**
ð.015Þ
.095***
ð.013Þ
Own-birth children:
ð7ÞBaseline ðcols. 1 and 3 of table 3; n5412,183Þ.116***
ð.001Þ
.088***
ð.002Þ
.089***
ð.002Þ
Comparable samples:
ð8ÞPositively selected parents; characteristics match those of
adoptive parents ðn53,939Þ
.102***
ð.014Þ
.086***
ð.016Þ
.078***
ð.020Þ
ð9ÞNegatively selected parents; characteristics match those of
biological parents with adopted away children ðn53,742Þ
.116***
ð.016Þ
.112***
ð.018Þ
.078***
ð.025Þ
NOTE.Robust standard errors are in parentheses.
a
Education levels are included by using dummies for the number of years of education, where missing observations are included in a separate dummy variable. Missing values for
parental income levels are set equal to their means and a dummy for missing income is included.
* Significant at the 10%level.
** Significant at the 5%level.
*** Significant at the 1%level.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
a similar percentage of 75%80%.
18
This implies that our large estimate
of the postbirth effect may be slightly underestimated.
B. External Validity
We investigate three potential problems to assess whether it is reason-
able to generalize the results from our sample of adoptees to the popu-
lation at large: ðiÞmany adoptees have unknown fathers; ðiiÞprebirth
characteristics of adopted children may differ from those of own-birth
children; ðiiiÞadoptive parents may be different from birth parents of non-
adopted children.
19
Our estimates may be biased due to restricting the sample of adoptees
to those for whom data are available for both of the biological parents.
This restriction is substantial, for more than half of the adoptees the
information on the biological fathers is missing. For the extended set of
adoptees, we reestimate the effects of the entrepreneurship status of bi-
ological and adoptive mothers on the adoptees’ outcome. The estimated
coefficients hardly change, and the effect of the biological mother be-
comes significant due to the larger sample size ðsee row 6 of table 4Þ.
The parents of adopted children are different from the parents of own-
birth children on average, as was indicated in table 2. In case these dif-
ferences affect our estimates, the comparison between adopted and non-
adopted children becomes problematic. To investigate whether our base-
line results for adoptees are a reasonable comparison to nonadoptees, and
thus externally valid, we reestimate the baseline equations for nonadopt-
ees using two different, more comparable samples. The first sample ad-
dresses the issue that adoptive parents tend to be positively selected, that
is, adopted children may face advantageous postbirth environments. It
consists of own-birth children and their parents, where we require that
the parents have similar observables to those of adoptive parents. Like-
wise, the second new sample consists of own-birth children and their par-
ents, where the parents are required to have similar observables to the
biological parents of adopted children. This sample addresses the issue that
adoptees may be endowed with less advantageous prebirth characteristics,
18
The actual placement guidelines in 1945 suggested that a child should be
adopted at age 1 after a 6-month trial period in the home. In 1968, the guidelines
suggested that the baby should be placed with its new family at age 36 months
and be adopted 34 months later. Note also that some adoptees may be mis-
classified as “late” adoptees because they were first placed in a foster home and
later adopted by their foster parents ðBohman 1970Þ. Also, children were most
likely not registered at their biological mother’s address until the official adoption
process was complete.
19
External validity also hinges on the assumption that adoptive parents treat
their children no different than own-birth parents do, nor should adoptees respond
systematically differently from own-birth children to the received parenting. Un-
fortunately we cannot test this assumption with our data.
284 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
since biological parents tend to be negatively selected. Both samples are
created using a propensity score matching method.
20
Rows 8 and 9 of table 4 present the results of reestimating our baseline
intergenerational association of entrepreneurship for these new samples of
own-birth children and parents. The results are very similar to the base-
line results for nonadopted children ðrow 7 of table 4Þ. They all fall within
the ranges of the summed effects of biological and adoptive parents for
adoptees that were indicated in the bottom panel of table 3.
21
As far as can
be judged from propensity matching on observable characteristics, our
baseline estimates are not sensitive to the fact that adopted children and
their adoptive parents are different from own-birth children and their par-
ents: these differences do not translate into meaningful changes in the es-
timated intergenerational association in entrepreneurship.
VI. An Exploration of Potential Postbirth Mechanisms
A. Four Candidate Mechanisms
We have learned that postbirth factors account for more of the inter-
generational transfer of entrepreneurship than prebirth factors do. But
what exactly is this entrepreneurial “nurture” effect? In this section, we
explore the plausibility of several candidate explanations that exist in the
literature ðsee Fairlie and Robb ½2007and Parker ½2009for a summaryÞ.
22
First, children may eventually inherit the family business. Previous stud-
ies find that inheritance is generally not large enough to explain much of
the intergenerational transfer of entrepreneurship ðParker 2009Þ. US evi-
dence produces estimates that 5.6% of the businesses are acquired by in-
heritance ð1.6%Þor gifts ð4%Þ. Canadian and Danish evidence produces
similar numbers of 5.5% ðAldrich, Renzulli, and Langton 1998Þand 8%
ðSørensen 2007Þ, respectively.
20
We employed a nearest-neighbor matching method without replacement. In
case of a tie, we included both neighbors. The propensity score was estimated
using a Probit model with adopted ðyes 51, no 50Þas the dependent variable.
Regressors included the child’s birth year, gender, mother’s age at child’s birth,
father’s age at child’s birth, mother’s income, father’s income, mother’s education,
father’s education, mother’s entrepreneurship status, and father’s entrepreneurship
status. When estimating the propensity score for our first sample of “positively”
selected parents, we matched biological parents with own-birth children to adop-
tive parents. When estimating the score for our second sample of “negatively”
selected parents, we match ed biological parents with own-birth children to the
biological parents of adopted-away children.
21
The coefficient in row 10 for the negatively selected group of fathers of own-
birth children changes, however, quite a bit ð2 percentage pointsÞ.
22
Examining candidate explanations for the biological association is not pos-
sible with our data.
Why Do Entrepreneurial Parents Have Entrepreneurial Children? 285
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
In this paper, we can only measure an upper bound of the probability
that a business was obtained through inheritance by using the indirect
method proposed by Sørensen ð2007Þ: entrepreneurs may have inherited
their business if their first start as an entrepreneur occurs at the same time
as their parents exit and if both firms are active in the same industry. In
our data, only 2.2% of all entrepreneurs who have entrepreneurial parents
satisfy this condition. Disregarding the exact timing, 10.8% of all en-
trepreneurs with entrepreneurial parents have their first stint as an en-
trepreneur in the same industry as their parents’ last stint. We view this
as an upper bound on the percentage of children who inherit the family
business, since it includes all persons ending up in the same industry as
their parents for whatever reason.
23
To assess the extent to which the inheritance mechanism may account
for our large postbirth estimates, we reestimated our baseline experiment
excluding these possible cases of inheritance. The results are very similar
ðavailable upon requestÞ, leaving the large postbirth effect unexplained.
The second mechanism is that children of entrepreneurs may have ac-
cess to cheaper capital. Dunn and Holtz-Eakin ð2000Þshow that the assets
of parents are correlated positively with transitions to entrepreneurship.
However, they point out that the underlying mechanism is parental en-
trepreneurship and human capital rather than wealth itself. In Sørensen
ð2007Þ, parental wealth does not explain the transfer of entrepreneurship.
In addition, other studies show that few business owners borrow capital
from family, at maximum 8% ðAldrich et al. 1998; Fairlie and Robb 2007Þ.
We address the cheap capital hypothesis by reestimating the adoptive
parents coefficients after adding controls for parental income and wealth.
Parental wealth is proxied by a dummy variable equal to one if the sum of
the parents’ pretax total factor incomes is in the top decile of the income
distribution and zero otherwise ðsee Hurst and Lusardi 2004Þ. These var-
iables have no detectable impact on our estimate of the postbirth associa-
tion. We also test the cheap capital hypothesis by creating a new dichot-
omous dependent variable that is equal to one if the child had started his
or her first company in a capital intensive industry ðwhich presumably
requires a larger initial capital investmentÞand zero if they did not. We
then examine whether or not adoptive parents’ wealth and/or entrepre-
23
However, the number of family companies who successfully transfer own-
ership from one generation to the next is probably much closer to 10.8%than
to 2.2%. In a survey of small businesses in Sweden whose principle owner was
aged 50 or older ði.e., likely to be thinking about retiringÞ, 8.7%of these owner-
respondents said that they had already transferred some share of their company
to their children ðNUTEK 2004Þ. When asked how they themselves had come to
own their own business, 2.4%answered that they had inherited their business,
0.2%had received it as a gift, and 6.9%had received ownership by purchasing
shares along with inheritance and/or a gift ðNUTEK 2004Þ.
286 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
neurial status can predict these presumably capital intensive startups
they do not.
The third mechanism is that entrepreneurial parents may provide their
children with general business human capital. If it is easier for children of
entrepreneurs to obtain “general business” or managerial human capital,
entrepreneurship might be a more promising career path and thus cause
a correlation across generations. Fairlie and Robb ð2007Þfind that having
self-employed parents increases profits and sales, and lowers closure, but
only when the entrepreneur has work experience in the family business.
Roberts ð1991Þand Sørensen ð2007Þfind no evidence that the children of
self-employed perform better as entrepreneurs. This explanation, how-
ever, cannot be tested in the realm of our study.
The fourth mechanism is that parents may pass on occupation- and/or
industry-specific skills or tastes to their children ðLaband and Lentz 1983;
Dunn and Holtz-Eakin 2000; Sørensen 2007; Corak and Piraino 2011Þ.
We address the industry-specific human capital explanation by reestimat-
ing the adoptive parent coefficients ðour postbirth effectsÞafter including
a dummy that is one whenever the child has ever worked in the exact same
industry as his/her parents in whatever labor market position.
24
This
slightly lowers the nurturing part of the intergenerational transmission of
entrepreneurship ðfathers by 8% and mothers by 4%Þ. Hence, the transfer
of industry-specific skills or tastes does not appear to explain much of the
adoptive parent-child correlation.
In summary, none of the candidate explanations considered above ðin-
heritance, cheap capital, or transfer of industry-specific skills and/or tastesÞ
appear to explain a significant share of the intergenerational association in
entrepreneurship. Next we turn to parental role modeling.
B. Parents as Role Models
The fifth explanation put forth in Parker’s ð2009Þoverview is that en-
trepreneurial parents may transmit the taste for entrepreneurship through
role modeling. This may be as subtle as increasing the child’s awareness of
entrepreneurship as a career option ðCarroll and Mosakowski 1987Þor
shaping the child’s values, such as a taste for autonomy. Sørensen ð2007Þ
shows a sizable transmission of entrepreneurship for children who have
only been exposed to parental entrepreneurship before the age of 16.
We address role modeling by allowing the effects of entrepreneurial fa-
thers and mothers to vary between daughters and sons. A stronger same-
sex transmission of entrepreneurship can be seen as an indication of the
presence of role modeling. Role model identification theory implies that
24
Please note that we thereby exclude the possibility that occupational-specific
skills or tastes are transferred to children working in different but related indus-
tries.
Why Do Entrepreneurial Parents Have Entrepreneurial Children? 287
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
role models are more often of the same gender ðRuef, Aldrich, and Carter
2003Þ. More generally, homophily is prevalent in many relationships; that
is, individuals have a tendency to bond easier with similar others ðMc-
Pherson, Smith-Lovin, and Cook 2001Þ. Thus, if the nurture part of the
parent-child association is particularly strong for mothers and their daugh-
ters on the one hand and for fathers and their sons on the other hand,
we could view this as evidence of role modeling ðAndersson and Ham-
marstedt 2011Þ. However, an alternative interpretation of a stronger same-
sex transmission is that it results from different behavior of parents toward
same-sex children. For example, Thomas ð1994Þfinds that mothers invest
more in the education of their daughters and fathers channel more re-
sources toward their sons.
In table 5, we show the results from various tests related to this role
modeling hypothesis. The top panel shows the results from estimating the
transmission of the entrepreneurship status of adoptive parents to their
children, for daughters and sons separately, where each column reflects
a separate regression equation. The bottom panel shows the results for
the parents of own-birth children. These are interesting as well for testing
our role modeling hypothesis, since the “nature” part of this relationship
is not likely to be different for boys and girls. Furthermore, the sample
of own-birth children is much larger and is therefore helpful to generate
more precise estimates of the differential nurturing effects of fathers and
mothers on sons and daughters.
Column 1 of table 5 shows what we call the basic result for the role
modeling exercise. The only controls included are the county dummies for
the children and the birth year dummies of the children and their rearing
parents. The left-hand side of the upper panel shows that the transmission
of entrepreneurship from adoptive parents to girls goes exclusively via the
mother. The right-hand side shows that the strongest effect for boys is
caused by the father. The single-sided F-tests show that the differences
between the positive coefficients pertaining to the father’s and mother’s
entrepreneurship status dummies are significant ðat the 10% levelÞin the
expected direction, for both girls and boys.
The bottom half of the first column shows comparable results for the
larger sample of own-birth children. In this case, parents affect the en-
trepreneurship status of the children through a combination of prebirth
and postbirth factors. The difference, however, is clear: fathers affect sons
more strongly, whereas for daughters the effect of the mother is signifi-
cantly larger.
25
The summed percentage effects of fathers and mothers
25
This finding supports the results of Andersson and Hammarstedt ð2011Þ, who
also find that intergenerational entrepreneurship is affected by both parents, but
more strongly by the parent of the same sex.
288 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
on the likelihood of entrepreneurship are strikingly similar for boys and
girls. Note that table 2 showed us that entrepreneurship is almost twice
as likely among boys, so in order to compare the effects of either parent in
percent increase one needs to double the coefficients for girls.
Column 2 includes controls for labor market characteristics of the nur-
turing parents as well as the variables that were used to check the other
postbirth mechanisms in the section above. We still observe a stronger
transfer for parents and children of the same sex. This suggests that we
find a direct “entrepreneurship role modeling effect” on top of an indirect
effect through occupational choice or education that is transmitted from
fathers ðmothersÞto sons ðdaughtersÞ. We also test whether the results of
this specification change upon excluding from the sample those who have
possibly inherited their business from their parents. The results ðwhich are
available on requestÞare comparable. Column 3 includes the entrepre-
neurship status and birth year dummies of the set of birth parents who
adopt away and mimics the baseline role modeling result.
Finally, in column 4, we try to distinguish between our role modeling
interpretation of the stronger same-sex transmission and the alternative
explanation that parents invest more in their children of the same sex.
We do so by including the ðmean-centeredÞnumber of sisters ðbrothersÞ
and an interaction term with the entrepreneurship status of the mother
ðfatherÞin the estimation for girls ðboysÞ: if the same-sex parent-child
transmission of entrepreneurship is driven by differential parenting
efforts, one would expect a negative interaction effect. That is, the effect
of having an entrepreneur for a father is stronger for sons who do not
have to share their father’s attention with a brother, whereas this should
have no impact on the transmission under our role-modeling explana-
tion.
Our findings from columns 13 remain largely unchanged when we al-
low the same-sex parent-child transmission of entrepreneurship to vary by
the number of same-sex siblings. In our sample of adoptees, we do not find
any support for the interpretation of stronger same-sex parenting efforts;
the interaction term in the girls equation is insignificant, while for boys it is
significant but positive. In the larger sample of own-birth children, there is
indication that the same-sex parent-child transmission of entrepreneurship
is diminishing with the number of same-sex siblings. However, the mothers
still affect daughters more strongly, whereas for sons the effect of the
father is larger. The same-sex parent-child transmission remains stronger
as long as girls do not have more than two sisters and boys do not have
more than two brothers ðwhich holds for 91% of our sampleÞ. All in all,
table 5 shows that the entrepreneurship transmission effect of the nur-
turing parent of the same sex is consistently larger than for the other nur-
turing parent.
Why Do Entrepreneurial Parents Have Entrepreneurial Children? 289
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
Table 5
Test of Gender Role Modeling
ð1Þð2Þð3Þð4Þ
Girls Boys Girls Boys Girls Boys Girls Boys
Adopted children:
Nurturing parents effect:
Father .006 .126*** .006 .118*** .000 .125*** .003 .126***
ð.021Þð.024Þð.020Þð.024Þð.021Þð.025Þð.021Þð.029Þ
Mother .065** .055* .062** .049 .069** .062** .064** .056*
ð.027Þð.030Þð.027Þð.030Þð.027Þð.031Þð.027Þð.030Þ
No. of sisters/no. of brothers 2.010* 2.006
ð.006Þð.007Þ
No. of sisters mother/no. of brothers father 2.025 .029*
ð.018Þð.016Þ
F-test of difference ðone-sidedÞ2.28* 2.42* 2.05* 2.38** 2.94** 1.78* 3.48** .72
No. of adoptive observations 1,792 2,149 1,792 2,149 1,792 2,149 1,792 2,149
Own-birth children:
Nurturing parents effect:
Father .042*** .131*** .038*** .117*** .042*** .132***
ð.002Þð.002Þð.002Þð.002Þð.002Þð.002Þ
290
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
ð1Þð2Þð3Þð4Þ
Mother .068*** .108*** .063*** .092*** .068*** .108***
ð.003Þð.003Þð.003Þð.003Þð.003Þð.003Þ
No. of sisters/no. of brothers 2.001 .002***
ð.001Þð.001Þ
No. of sisters mother/no. of brothers father 2.006*** 2.011***
ð.002Þð.002Þ
F-test of difference ðone-sidedÞ46.28*** 25.73*** 40.54*** 31.53*** 49.14*** 50.45***
No. of own-birth observations 200,964 211,219 200,964 211,219 200,964 211,219
Controls included:
Education, income, wealth, same industry No Yes No No
Entrepreneur status and birth year of nonnurturing parents No No Yes No
NOTE.The OLS regressions include the entrepreneurship status of both nurturing parents. Standard errors in parentheses are robust. All regressions include birth year
dummies and county of residence dummies in 1965 for the children and birth year dummies for the nurturing parents. Column 2 includes education dummies, income levels, and
wealth dummies for both nurturing parents. It also includes a dummy that indicates whether the parent has ever worked in the same industry as the child, either as an entrepreneur
or as a wage employee. Column 3 includes, in addition to the variables in col. 1, the entrepreneurship status and birth year dummies of the biological parents of adopted children.
Column 4 includes the ðmean-centeredÞnumber of sisters ðbrothersÞin the girls ðboysÞregression and an interaction of this variable with the mother’s ðfather’sÞentrepreneurship
status.
* Significant at the 10%level.
** Significant at the 5%level.
*** Significant at the 1%level.
291
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
VII. Conclusion
Questions concerning the origins of entrepreneurship have attracted a
lot of attention and for good reasons. Entrepreneurship has important
economic and social benefits, which could potentially justify both private
and public spending on entrepreneurial training, business school educa-
tion, and public programs aimed at promoting entrepreneurship ðKlapper,
Laeven, and Rajan 2006; Storey 2006Þ. The true value of these policies,
however, cannot be judged correctly without more information concern-
ing the origins and malleability of entrepreneurship.
A well-established stylized fact emerging from entrepreneurship eco-
nomics concerning the origins of entrepreneurship is the high correlation
between the entrepreneurial choices of parents and their children. In our
study, we find that children of entrepreneurs are 60% more likely to be-
come entrepreneurs than others using a large and representative sample of
the Swedish population. The key question our paper addresses is to what
extent this strong intergenerational transmission of entrepreneurship is
driven by prebirth or postbirth factors. In essence, we want to know
whether entrepreneurs are born or bred. The larger the contribution of
nurture vis-a`-vis nature, the larger the potential benefit of programs aimed
at fostering entrepreneurship.
We identify prebirth and postbirth factors by analyzing employment
histories of Swedish adoptees together with their biological and adoptive
parents. Our decomposition exercise reveals a strong correlation of entre-
preneurship between both types of parents and their children. Our main
results shed light on the relative importance of prebirth and postbirth
factors. We find that the importance of adoptive parents is twice as large as
the influence of biological parents.
The evidence of biological underpinnings of entrepreneurship is in line
with recent twin studies ðNicolaou et al. 2008; Zhang et al. 2009; Nicolaou
and Shane 2010Þ. However, a direct comparison of our results with these
studies is not possible, as we address the question of nature versus nur-
ture from a different angle. Twin studies decompose the total variation
in entrepreneurship choices into genes and shared and nonshared envi-
ronment, whereas we decompose the intergenerational association. Thus,
twin studies allow for a larger set of environmental influences, and they
allow prebirth factors that promote entrepreneurship to be passed on from
nonentrepreneurial parents as well. Twin studies find that 40% of the
variance in entrepreneurship choices is explained by genes, and with the
exception Zhang et al. ð2009Þ, they find no influence of the twins’ shared
environment. This last finding stands in stark contrast to our own findings.
In particular, the zero effect of the twins’ shared environment is difficult to
reconcile with our findings, since shared environment should include most
of our important postbirth factors, such as role modeling. Since both
adoption studies and twin studies have their own methodological chal-
292 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
lenges, it is important to analyze this important question using both
methods. We are the first to do so using the adoption method. We view that
as the first contribution of our study.
Our second contribution is that we explore a set of likely explanations
for the nurture effect. Previous studies on intergenerational entrepreneur-
ship have identified and measured five environmental mechanisms: ðiÞthe
inheritance of a family business, ðiiÞaccess to cheaper capital, ðiiiÞless
costly acquisition of general business human capital, ðivÞthe transfer of
industry- or firm-specific human capital, and ðvÞa kind of “catch all” ex-
planation that includes preferences and parental role modeling. In contrast
to previous studies, we can explore most of these different mechanisms
while at the same time controlling for genes. We find indirect evidence
in favor of parental role modeling based on an analysis of gender-specific
parent-child transmissions of entrepreneurship. This finding deepens our
understanding of the role that parents play in fostering entrepreneur-
ship. Moreover, this finding raises a number of intriguing questions re-
lating role modeling to entrepreneurship that could be placed on the re-
search agenda.
Recent empirical evidence indicates that networks and peer groups ðe.g.,
Djankov et al. 2006; Stuart and Ding 2006; Nanda and Sørensen 2010Þ,as
well as regional inheritances and clusters ðReynolds, Storey, and West-
head 1994; Lafuente, Vailliant, and Rialp 2007Þ, influence entrepreneur-
ship decisions. This literature suggests that role modeling may be driv-
ing these effects. Given our result about the fruitful role of parental role
models, further research into the nature and effect of other role models,
for instance, entrepreneurs in the classroom, in regions, in peer groups, or
in networks, might point out the nonspecificity of parental role models
and the possible substitution possibilities in the wider social networks
of people.
References
Acs, Zoltan J. 1999. Are small firms important? Their role and impact.
Boston: Kluwer Academic.
Aldrich, Howard E., Linda A. Renzuli, and Nancy Langton. 1998. Passing
on privilege: Resources provided by self-employed parents to their self-
employed children. Research in Social Stratification and Mobility 16:
291317.
Andersson, Lina, and Mats Hammarstedt. 2010. Intergenerational trans-
missions in immigrant self-employment: Evidence from three genera-
tions. Small Business Economics 34:26176.
———. 2011. Transmission of self-employment across immigrant gen-
erations: The importance of ethnic background and gender. Review of
Economics of the Household 34:26176.
Why Do Entrepreneurial Parents Have Entrepreneurial Children? 293
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
Arum, Richard, and Walter Mueller, eds. 2004. The reemergence of self-
employment: A comparative study of self-employment dynamics and
social inequality, Princeton, NJ: Princeton University Press.
Birch, David G. W. 1979. The job creation process. Center for the Study of
Neighborhood and Regional Change. Cambridge, MA: MIT Press.
Bjo
¨rklund, Anders, Mikael Lindahl, and Erik Plug. 2006. The origins of
intergenerational associations: Lessons from Swedish adoption data.
Quarterly Journal of Economics 121:9991028.
Bohman, Michael. 1970. Adopted children and their families: A follow-up
study of adopted children, their background, environment and adjust-
ment. Stockholm: Proprius Fo
¨rlag.
Bosma, Niels, Jolanda Hessels, Veronique Schutjens Mirjam van Praag,
and Ingrid Verheul. 2012. Entrepreneurship and role models. Journal of
Economic Psychology 33:41024.
Carroll, Glenn R., and Elaine Mosakowski. 1987. The career dynamics of
self-employment. Administrative Science Quarterly 32:57089.
Cesarini, David, Magnus Johannesson, and Sven Oskarsson. 2014. Pre-
birth factors, post-birth factors and voting: Evidence from Swedish
adoption data. American Political Science Review 108:7187.
Colombier, Nathalie, and David Masclet. 2008. Intergenerational corre-
lation in self-employment: Some further evidence from French ECHP
data. Small Business Economics 30:42337.
Corak, Miles, and Patrizio Piraino. 2011. The intergenerational trans-
mission of employers. Journal of Labor Economics 29:3768.
Djankov, Simeon, Yingyi Qian, Ge´rard Roland, and Ekatarina Zhur-
avskaya. 2006. Who are China’s entrepreneurs? American Economic Re-
view 96:34852.
Dunn, Thomas, and Douglas Holtz-Eakin. 2000. Financial capital, human
capital, and the transition to self-employment: Evidence from inter-
generational links. Journal of Labor Economics 18:282305.
Fairlie, Robert W., and Alicia M. Robb. 2007. Families, human capital, and
small business: Evidence from the Characteristics of Business Owners
Survey. Industrial and Labor Relations Review 60:22545.
Hjalmarsson, Randi, and Matthew J. Lindquist. 2013. The origins of
intergenerational associations in crime: Lessons from Swedish adoption
data. Labour Economics 20:6881.
Hurst, Erik, and Annamaria Lusardi. 2004. Liquidity constraints, house-
hold wealth, and entrepreneurship. Journal of Political Economy 112:
31947.
Klapper, Leora, Luc Laeven, and Raghuram Rajan. 2006. Entry regulation
as a barrier to entrepreneurship. Journal of Financial Economics 82:591
629.
Koellinger, Philipp D., Matthijs J. H. M. van der Loos, Patrick J. F.
Groenen, A. Roy Thurik, Fernando Rivadeneira, Frank J. A. van Rooij,
Andre´ G. Uitterlinden, and Albert Hofman. 2010. Genome-wide
294 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
association studies in economics and entrepreneurship research: prom-
ises and limitations. Small Business Economics 35:118.
Laband, David N., and Bernard F. Lentz. 1983. Like father, like son:
Toward an economic theory of occupational following. Southern Eco-
nomic Journal 50:47493.
Lafuente, Esteban, Yancy Vailliant, and Josep Rialp. 2007. Regional dif-
ferences in the influence of role models: Comparing the entrepreneurial
process of rural Catalonia. Regional Studies 41:77996.
Laspita, Stavroula, Nicola Breugst, Stephan Heblich, and Holger Pat-
zelt. 2012. Intergenerational transmission of entrepreneurial intentions.
Journal of Business Venturing 27:41435.
McPherson, Miller, Lynn Smith-Lovin, and James M. Cook. 2001. Birds
of a feather: Homophily in social networks. Annual Review of Sociology
27:41544.
Nanda, Ramana, and Jesper B. Sørensen. 2010. Workplace peers and en-
trepreneurship. Management Science 56:111626.
Nicolaou, Nicos, and Scott Shane. 2010. Entrepreneurship and occupa-
tional choice: Genetic and environmental influences. Journal of Eco-
nomic Behavior and Organization 76:314.
———. 2011. The genetics of entrepreneurship. In Handbook of re-
search on innovation and entrepreneurship, ed. David B. Audretsch,
Oliver Falck, Stephan Heblich, and Adam Lederer. Cheltenham: Ed-
ward Elgar.
Nicolaou, Nicos, Scott Shane, Lynn Cherkas, Janice Hunkin, and Tim D.
Spector. 2008. Is the tendency to engage in entrepreneurship genetic?
Management Science 54:16779.
Nordlo
¨f, Barbro. 2001. Svenska adoptioner i Stockholm, 19181973. FoU-
rapport 2001:8. Socialtja
¨nstfo
¨rvaltningen.
NUTEK. 2004. Generationsskiften i fo
¨retagen problemanalys. B 2004:2.
Verket fo
¨rna
¨ringslivsutveckling. Stockholm.
Parker, Simon C. 2009. The economics of entrepreneurship. Cambridge:
Cambridge University Press.
Reynolds, Paul D., David J. Storey, and Paul Westhead. 1994. Cross-
national comparisons of the variation in new firm formation rates.
Regional Studies 28:44356.
Roberts, Edward B. 1991. Entrepreneurs in high technology: Lessons from
MIT and beyond. New York: Oxford University Press.
Ruef, Martin, Howard E. Aldrich, and Nancy M. Carter. 2003. The
structure of founding teams: Homophily, strong ties, and isolation
among U.S. entrepreneurs. American Sociological Review 68:195222.
Sørensen, Jesper B. 2007. Closure and exposure: Mechanisms in the in-
tergenerational transmission of self-employment. Research in the Soci-
ology of Organizations 25:83124.
Storey, David J. 2006. Evaluating SME policies and programmes: Tech-
nical and political dimensions. In The Oxford handbook of entrepre-
Why Do Entrepreneurial Parents Have Entrepreneurial Children? 295
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
neurship, ed. Mark Casson, Bernard Yeung, Anuradha Basu, and Nigel
Wadeson. Oxford: Oxford University Press.
Stuart, Toby E., and Waverly W. Ding. 2006. When do scientists become
entrepreneurs? The social structural antecedents of commercial activity
in the academic life sciences. American Journal of Sociology 112:97144.
Thomas, Duncan. 1994. Like father, like son; like mother, like daughter:
Parental resources and child height. Journal of Human Resources 29:
95088.
Tsou, Meng-Wen, Jin-Tan Liu, and James K. Hammit. 2012. The inter-
generational transmission of education: Evidence from Taiwanese adop-
tions. Economics Letters 115:13436.
Van der Loos, Matthijs J. H. M., Cornelius A. Rietveld, Niina Eklund,
Philipp D. Koellinger, Fernando Rivadeneira, et al. 2013. The molecular
genetic architecture of self-employment. PLoS ONE 8: e60542. doi:10.
1371/journal.pone.0060542.
Van Praag, C. Mirjam, and Peter H. Versloot. 2007. What is the value of
entrepreneurship? A review of recent research. Small Business Eco-
nomics 29:35182.
Zhang, Zhen, Michael J. Zyphur, Jayanth Narayanan, Richard D. Arvey,
Sankaip Chaturvedi, Bruce J. Avolio, Paul Lichtenstein, and Gerry
Larsson. 2009. The genetic basis of entrepreneurship: Effects of gender
and personality. Organizational Behavior and Human Decision Pro-
cesses 110:93107.
296 Lindquist et al.
This content downloaded from 145.18.180.95 on Wed, 4 Mar 2015 05:02:35 AM
All use subject to JSTOR Terms and Conditions
... Pupils from an entrepreneurial family background, meaning one or more members of their immediate family are entrepreneurs, are poised to improve their EI more than the pupils without such a background. Lindquist et al. (2012) demonstrated that nurture is more important than nature. That is, children adopted into an entrepreneurial family are more likely to become entrepreneurs themselves than the biological children of entrepreneurs adopted by non-entrepreneur parents. ...
... Another factor that has been shown to play a crucial role in EC development is entrepreneurial family background. Lindquist et al. (2012), for instance, have demonstrated that entrepreneurial parents most probably pass on a particular set of EC that are speci c for developing and running a business to their children. Moreover, it does not matter whether these children are biological children or adopted, which might mean that role modeling is more important than genetics. ...
... Also, if the parents of the pupils taking part in the EEP are active in entrepreneurship, then the EEP will impact the development of ESE more than if the parents are not. Lindquist et al. (2012) demonstrated that parental role modeling increases the probability of children becoming entrepreneurs. Such children are thus more likely to believe in their abilities to perform the entrepreneurial act successfully. ...
... How do sociocultural imprints affect the decision to become an entrepreneur? One line of inquiry has highlighted the role of individual-or organization-level social imprints such as parental role modeling (Sørensen 2007, Lindquist et al. 2015, peer effects (Nanda and Sørensen 2010, Kacperczyk 2013, Qin and Estrin 2015, and mentors (Roach and Sauermann 2015, Azoulay et al. 2017, Rocha and Van Praag 2020. This literature stresses the importance of socially transmitted individual-level differences in attitudes and preferences regarding entrepreneurship. ...
... We corroborate these findings using instrumental variable regressions and an alternative measure of entrepreneurial culture derived from stated preferences. Our results are robust to a number of alternative explanations often put forward in the (immigrant) entrepreneurship literature (Fairlie and Lofstrom 2015), such as the potential role of unobserved entrepreneurship-specific human capital, financial resources, labor market discrimination, family support, and direct parent-child linkages, including parental selfemployment status (Sørensen 2007, Lindquist et al. 2015. We also show that the effect of entrepreneurial culture is stronger-i.e., positively moderated-if immigrants stem from cultures that practice more intensive parenting styles, a result that directly speaks to the intergenerational transmission of cultural dispositions. ...
... It could be that the observed effect for entrepreneurial culture does not reflect an effect of culture per se but is driven by direct linkages between parents and children or noncultural characteristics of parents. Children of parents who are entrepreneurs are more likely to become entrepreneurs themselves because self-employed parents transmit entrepreneurial dispositions but also, social, financial, and (entrepreneurial) human capital to their offspring (Sørensen 2007, Laspita et al. 2012, Lindquist et al. 2015, Wyrwich 2015, Vladasel et al. 2021. Although we have no information on parental occupations for the individuals in our U.S. sample, we can construct a proxy for parental selfemployment that reflects the past ancestry by state-level self-employment rates of first-generation immigrants based on the 1960-1990 U.S. Decennial Census files. ...
Article
Full-text available
Does national culture influence entrepreneurship? Given that entrepreneurship and the economic, formal institutional, and cultural characteristics of nations are deeply intertwined and co-vary, it is difficult to isolate the effect of culture on entrepreneurship. In this study, we examine the self-employment choices of second-generation immigrants who were born, educated, and currently live in one country, but were raised by parents stemming from another country. We argue that entrepreneurship is influenced by durable, portable, and intergenerationally transmitted cultural imprints such that second-generation immigrants are more likely to become entrepreneurs if their parents originate from countries characterized by a strong entrepreneurial culture. Our multilevel analysis of two independent samples—65,323 second-generation immigrants of 52 different ancestries who were born, were raised, and live in the United States and 4,165 second-generation immigrants of 31 ancestries in Europe—shows that entrepreneurial culture is positively associated with the likelihood that individuals are entrepreneurs. Our results are robust to alternative non-cultural explanations, such as differences in resource holdings, labor market discrimination, and direct parent-child linkages. Overall, our study highlights the durability, portability, and intergenerational transmission of entrepreneurial culture as well as the profound impact of national culture on entrepreneurship. Supplemental Material: The online appendix is available at https://doi.org/10.1287/orsc.2022.1645 .
... Parental entrepreneurship is a strong predictor of the offspring's entrepreneurial choice (Dunn & Holtz-Eakin, 2000). Lindquist et al. (2015) show that biological children of entrepreneurial parents are 60% more likely to be entrepreneurs as well. This can be due to genetic inheritance (e.g., Nicolaou et al., 2008) or social learning through role modeling, as Lindquist et al. (2015) and Chlosta et al. (2012) suggest. ...
... Lindquist et al. (2015) show that biological children of entrepreneurial parents are 60% more likely to be entrepreneurs as well. This can be due to genetic inheritance (e.g., Nicolaou et al., 2008) or social learning through role modeling, as Lindquist et al. (2015) and Chlosta et al. (2012) suggest. 9 As the family environment during childhood is known to strongly influence the development of cognitive and noncognitive skills (Heckman, 2006), it also plausibly affects cognitive reflection. ...
Article
Intuition is a central element of entrepreneurial decision-making. We conceptually replicate a published study by using new representative data from 1961 adults and the widely used Cognitive Reflection Test, which assesses the ability to avoid intuitive decisions and to switch to an analytical process. We extend the analysis by exploring occupational sorting versus environmental influence as mechanisms, the role of overconfidence, and heterogeneity. Our results confirm that entrepreneurs do not resist intuitive (but potentially wrong) decisions as much as hired managers do. Our extensions suggest that this difference is not fully explained by occupational sorting, but partially by overconfidence.
... Gender and ethnicity play important role in selection of role model consciously and unconsciously. Lindquist et al. (2015) finds that the father is the most influential role model for male offspring and the mother is the most important role model for female offspring. However, some contradictory researches are also found (Austin & Nauta, 2016;Wohlford et al., 2004). ...
... Family values and resources can encourage entrepreneurship. Research indicates that having a family entrepreneur increases the likelihood of entrepreneurship (Dunn and Holtz-Eakin, 2000;Sørensen, 2007;Arum and Mueller, 2009;Lindquist, 2015;Youssef et al., 2021). Relational support which include sharing resources and assistancecan influence young people's career choices. ...
Article
Full-text available
Purpose: The study aims to identify the factors contributing to the formation of entrepreneurial intentions among recent university graduates. The original value of this study lies in providing support and a better understanding of the entrepreneurial decision-making process of Vietnamese fresh graduates. Research design, data and methodology: The research methodology employs Exploratory Factor Analysis (EFA) with a sample size of 385 university graduates in Vietnam. Results: The results indicate that three key factors, including relational support, educational support, and attitude towards business, positively impact entrepreneurial intentions. Conclusions: The managerial implications are provided to promote entrepreneurial intention among recent university graduates.
... Gender and ethnicity play important role in selection of role model consciously and unconsciously. Lindquist et al. (2015) finds that the father is the most influential role model for male offspring and the mother is the most important role model for female offspring. However, some contradictory researches are also found (Austin & Nauta, 2016;Wohlford et al., 2004). ...
Article
Full-text available
Purpose: The aim of this paper is to measure the impact of role models developed through interaction between entrepreneurs and students on self-efficacy and the entrepreneurial intention on Aviation Management Students Design/methodology/approach: Experimental research is conducted by the researchers on 80 students of a management institute offering a unique program of Aviation Management. Role model identification, attitude toward the role model, entrepreneurial self-efficacy and entrepreneurial intention were measured through structural equation modelling techniques. Findings : Exposure to real role models from Aviation Industry who are the first-generation successful entrepreneurs through face-to-face interaction created a positive impact on self-efficacy and behavioural intentions when the students identify themselves with the role models. Successful role models reinforce role model identification, favourable attitudes for entrepreneurship and in turn, augment self-efficacy and entrepreneurial intention. Practical implications: The study brought out that related role models enhance the self-efficacy and entrepreneurial motivation. Hence, this should be part of the curriculum for entrepreneurial education as it would result in intension and finally behaviour to be an entrepreneur. Also, the role model get influence with the students and they become mentors as mentioned by the role models during a discussion with the students. Originality/value Unknown real-life role models in related fields serve as role models by interaction and enhance the impact on the entrepreneurship studies and new entrepreneurs can be developed by arranging interaction between entrepreneurs from a related field.
... The recent managerial literature on entrepreneurship also stresses the role of the socioeconomic environment on starting a new firm. First, family background (Burke et al., 2008;Lindquist et al., 2015) and education (Caliendo et al., 2021;Djankov et al., 2010) are identified as important factor to explain firm formation. The geographical environment also matters: local industrial restructuring, agglomeration economies, and availability of resources at the local level (Armington & Acs, 2002;Wenting & Frenken, 2011) impact the decision to start a business. ...
Article
We study what profiles of individuals were the most likely to give up their entrepreneurial project at the beginning of the Covid‐19 crisis. To do so, we run an experiment during the first lockdown in the United Kingdom. Our results show that the first months of the crisis have a sharp screening effect: 63% of prospective entrepreneurs postpone or cancel their project in May 2020. Taste for risk or competition does not characterize those who stick to their project. Instead, low opportunity costs to continue and concern for one's own interest instead of cooperation are common among entrepreneurs who persist.
Chapter
The entrepreneur is regarded as a driver of economic development encouraging researchers to delve into the causes that lead some people to choose the entrepreneurial activity. The response has followed two approaches: biological and sociological. The former privileges non-cognitive skills while the sociological favors factors are associated with learning and contextual conditions. Thus, the question arises: Is behavior determined by non-cognitive skills or can it be 'shaped' throughout an individual's lifetime? Using an exploratory approach supported by a literature review and contacts with the Portuguese (context under analysis) educational system, the authors aim to understand which skills are critical, the way they are acquired and developed, and the role the educational system plays in their development, concluding that entrepreneurial activity implies cognitive and non-cognitive skills which the system must deliver provided the teachers' training in advance.
Article
Full-text available
Entrepreneurship has evolved into a valuable tool for facilitating job creation in response to Nigeria’s youth bulge and declining job opportunities in the formal sector. This study assessed the impact of entrepreneurship on youths’ livelihood, focusing on the National Directorate of Employment (NDE) program in Kano state. The study employed both qualitative and quantitative methods. A total of 266 respondents, including 148 young entrepreneurs and 118 non-entrepreneurs, were sampled using a multi-stage sampling technique, with Focused Group Discussions (FGDs) used to collect qualitative data from 25 youths. Data were analyzed using descriptive statistics and an Endogenous Switching Treatment Effect Regression (ESTER) model. The descriptive analysis showed that engagement in entrepreneurship was determined by age, having consistent income source, credit access, number of entrepreneurs in the household, access to entrepreneurship training, household size, and asset ownership. Even though entrepreneurs earned higher monthly income than non-entrepreneurs, their income was just slightly above the Nigerian minimum wage. Furthermore, asset ownership, access to credit, group membership, and access to training, all of which are strong predictors of entrepreneurship, were higher among entrepreneurs compared to non-entrepreneurs. The findings also revealed that entrepreneurs had better livelihood outcomes, as measured by income and self-assessed living condition, than non-entrepreneurs. Even though these outcomes could have resulted from other externalities, the empirical analysis helped to address such endogeneity, thereby attributing the outcome estimates solely to entrepreneurship. These results, therefore, show the relevance of entrepreneurship in alleviating poverty and generating better livelihood outcomes for young Nigerians.
Article
Full-text available
What role peers play in individuals’ decisions to become entrepreneurs and to what extent peer effects play a role in influencing behaviours at the various stages of business venturing are important questions for scholars and policymakers. This systematic review takes stock of the recent additions to the literature around the phenomenon of peer influence in entrepreneurship. The review identified 2894 documents which were then narrowed down through three consecutive filtering stages. We thematically analysed the final sample of 27 empirical studies that shed light on how individual peers influence the process and outcomes of these individuals’ entrepreneurial intentions and behaviour, allowing for critical analysis. We propose a conceptual schema of social influence that occurs in interactions among entrepreneurial individuals within business venturing and across the three stages of pre‐formation, formation and growth. Our framework reconciles the conceptual classification around discovering, evaluating and exploiting entrepreneurial opportunities with the mechanisms of social influence affecting entrepreneurial behaviours. Grounded in the findings of the literature review, this framework synthesizes peer influence in entrepreneurship with the tripartite distinction of the behavioural motives recognized in contemporary theories of social influence. We suggest promising directions for further research on how interactions with peers might affect individuals’ entrepreneurial behaviours.
Article
Full-text available
This article analyzes a rich Swedish data set with information on the electoral turnout of a large sample of adoptees, their siblings, their adoptive parents, and their biological parents. We use a simple regression framework to decompose the parent-child resemblance in voting into pre-birth factors, measured by biological parents’ voting, and post-birth factors, measured by adoptive parents’ voting. Adoptees are more likely to vote if their biological parents were voters and if they were assigned to families in which the adoptive parents vote. We find evidence of interactions between the pre- and post-birth factors: the effect of the post-birth environment on turnout is greater amongst adoptees whose biological mothers are nonvoters. We also show that the relationships between parental characteristics, such as education, and child turnout, persist even in the absence of a genetic link between parent and child. The regression-based framework we utilize provides a basis for the integration of behavior-genetic research into mainstream political science.
Book
Are Small Firms Important? Their Role and Impact proposes and supports the claim that small firms make two indispensable contributions to the economy. First, they are an integral part of the renewal process that pervades market economies. New and small firms play a crucial role in experimentation and innovation that leads to technological change, productivity and economic growth. Second, small firms are the essential mechanism by which millions enter the economic and social mainstream of American society. The public policy implications for sustained economic growth and social well-being is the continued high-level creation of new and small firms by all segments of society. It should be the role of government policy to facilitate that process by eliminating entry barriers, lowering transaction costs, and minimizing regulation.
Article
This article is primarily focused on further developing the theme of the political economy way of evaluating the impact of Small and Medium-sized Enterprise (SME) policies. It reaches five key conclusions. First, that evaluation needs to become more central to the policy-making process. Evaluation should not be undertaken solely as a historic accounting exercise to determine whether public money has been spent wisely, although that role is of value. Instead of being, 'at the end of the line', evaluation should be used to inform current policy, so that current objectives and targets may be modified in the light of evidence of policy effectiveness. Hence considerations of how policy is evaluated should therefore be incorporated into policy formulation when new ideas are being developed. They could even influence the choices made by governments about how best to engage with SMEs. Specifically, evaluation has to be incorporated as a key element in policy development.
Article
We draw on cross-cultural theory and the Global Leadership and Organizational Behavior Effectiveness project to develop a model for the transmission of entrepreneurial intentions within families in different cultures. Using data on more than 40,000 individuals from 15 countries, we show that beyond the transmission of entrepreneurial intentions from parents to children, grandparents – either directly or “indirectly†via the parents – impact the offspring's intentions. Moreover, we find that parents' and grandparents' influences partly substitute for one another. The strength of these effects varies across cultures. Our results provide a more detailed picture of the intergenerational transmission of entrepreneurial intentions.
Article
This paper examines the causal effect of parental schooling on children’s schooling using a large sample of adoptees from Taiwan. Using birth-parents’ education to help control for selective placement of children with adoptive parents, we find that adoptees raised with more highly educated parents have higher educational attainment, measured by years of schooling and probability of university graduation. We also find evidence that adoptive father’s schooling is more important for sons’ and adoptive mother’s schooling is more important for daughters’ educational attainment. These results support the notion that family environment (nurture) is important in determining children’s educational outcomes, independent of genetic endowment.
Article
This paper studies the intergenerational transmissions of self-employment abilities among immigrants in Sweden. The results show that second-generation immigrants are over-represented in self-employment compared to natives. Male immigrants from countries neighbouring to Sweden and natives alike seem to use both mothers and fathers as role models in their self-employment decision, but the father is the stronger role model among male immigrants from more geographically distant regions. Female immigrants use both their father and their mother as role models in their self-employment decision. Furthermore, male immigrants and male natives tend to become self-employed in the same business sector as their fathers; female immigrants and female natives with self-employed parents are over-represented in self-employment but not necessarily in the same business sector as their parents.
Article
This is a book about the formation, development, and success or failure of new high technology companies, focusing on those that grew under the auspices of entrepreneurs from Massachusetts Institute of Technology (MIT) in Boston at the end of World War Two. Trained in high-technology in MIT's labs and academic departments or in the local industrial marvel that became known as the "Route 128 phenomenon", these entrepreneurs took their technical and innate skills with them to found their own new companies. The book is based on extensive empirical research on these firms conducted over a period of twenty-five years and much previously written work on the subject, and is the culmination of such earlier work and synthesized findings. It centers on people, technology, money, and markets, and its main goal is to provide insights that may eventually contribute to fulfilling other entrepreneurs' dreams and other communities' hopes. The book chapters comprise three connected sections - treating birth, transition and growth, and success or failure.