ArticlePDF Available

Failure Is an Option: Institutional Change, Entrepreneurial Risk, and New Firm Growth


Abstract and Figures

Does an institutional change that lowers failure barriers improve new firm growth?We take advantage of a quasi-natural experiment in Japan that drastically reduced the stringency of bankruptcy regulations to examine this question. Using longitudinal data over a 10-year period, we find that bankruptcy reform increases the rates of bankruptcy and founding-and, more importantly, the likelihood of high-growth ventures-by disproportionately encouraging elite individuals (i.e., those with superior human and social capital) to start firms. In turn, these firms are more likely than others to achieve high growth. Broadly put, we contribute to research at the nexus of institutional theory and entrepreneurship by emphasizing the connectedness of barriers to failure, venture growth, and elite entrepreneurs. We also highlight how institutional change that eases bankruptcy change can foster a regenerative cycle of failure, founding, and growth by attracting more capable entrepreneurs. Overall, we conclude that lowering barriers to failure via lenient bankruptcy laws encourages more capable-and not just more-entrepreneurs to start firms.
Content may be subject to copyright.
Articles in Advance,pp. 1–20 ISSN 1047-7039 (print), ISSN 1526-5455 (online)
Failure Is an Option: Institutional Change, Entrepreneurial Risk,
and New Firm Growth
Robert N. Eberhart,aCharles E. Eesley,bKathleen M. Eisenhardtb
aDepartment of Management, Leavey Business School, Santa Clara University, Santa Clara, California 95053; bDepartment of Management
Science and Engineering, Stanford University, Stanford, California 94305
Contact: (RNE); (CEE); (KME)
Received: July 3, 2014
Revised: February 26, 2016; October 21, 2016
Accepted: December 4, 2016
Published Online in Articles in Advance:
Copyright: ©2017 INFORMS
Abstract. Does an institutional change that lowers failure barriers improve new firm
growth? We take advantage of a quasi-natural experiment in Japan that drastically reduced
the stringency of bankruptcy regulations to examine this question. Using longitudinal data
over a 10-year period, we find that bankruptcy reform increases the rates of bankruptcy
and founding—and, more importantly, the likelihood of high-growth ventures—by dis-
proportionately encouraging elite individuals (i.e., those with superior human and social
capital) to start firms. In turn, these firms are more likely than others to achieve high
growth. Broadly put, we contribute to research at the nexus of institutional theory and en-
trepreneurship by emphasizing the connectedness of barriers to failure, venture growth,
and elite entrepreneurs. We also highlight how institutional change that eases bankruptcy
change can foster a regenerative cycle of failure, founding, and growth by attracting more
capable entrepreneurs. Overall, we conclude that lowering barriers to failure via lenient
bankruptcy laws encourages more capable—and not just more—entrepreneurs to start firms.
Supplemental Material:
The online appendix is available at
institutional theory
regulatory change
institutional change and entrepreneurship
Since the foundational work of Meyer and Rowan
(1977) and DiMaggio and Powell (1983), research link-
ing institutional theory to entrepreneurship has rec-
ognized the crucial role of legitimacy, uncertainty,
and risk for new firm outcomes (Tolbert et al. 2011).
This research shows that the regulatory, normative,
and cultural/cognitive institutional environments can
shape the legitimacy of new firms as well as affect
the uncertainty and related risk that they face (Scott
2008). For example, institutional arrangements such as
business certification can aid new firms by helping
them to overcome the concerns of resource providers
who are uncertain about whether to provide assis-
tance (Sine et al. 2005). Certification can provide the
legitimacy that new firms need to convince poten-
tial stakeholders to participate in them (Lounsbury
and Glynn 2001, Suchman 1995). Similarly, the insti-
tutional environment can affect the uncertainty and
related risk of new firms, and these, in turn, can affect
entrepreneurs’ behaviors and the outcomes of their
firms (Hiatt and Sine 2014). In sum, research on insti-
tutional theory emphasizes how the institutional envi-
ronment can influence the legitimacy of new firms,
shape the uncertainty and risks that they face, and
affect entrepreneurial behaviors and, ultimately, firm
One institution that has received increasing scholarly
attention is bankruptcy regulation. Bankruptcy regula-
tion is crucial because bankruptcy is a salient outcome
for all firms, particularly new firms. This regulatory
institution shapes the ease of exit for a failing firm,
the financial cost of that exit, and the degree to which
the legitimacy of the executives leading the failed firm
is diminished by bankruptcy (Wiesenfeld et al. 2008).
Beyond exit, bankruptcy regulation can also influence
entry. That is, bankruptcy regulation affects the uncer-
tainty and risk of starting new firms, and in turn, it
affects the choice of potential entrepreneurs to launch
firms. To this point, empirical evidence links more
stringent bankruptcy laws with lower entrepreneur-
ship rates (Armour and Cumming 2008, Lee et al.
2007). For example, nations with stricter bankruptcy
laws (e.g., little protection of personal assets, lengthy
and costly bankruptcy process) have lower rates of firm
formation (Peng et al. 2009). Yet while the effects of
easier exit via bankruptcy on firm founding are clear,
what remains unclear is whether ease of exit also influ-
ences the behaviors of entrepreneurs once they have
founded their firms. Overall, we have an incomplete
understanding of how the institution of bankruptcy
regulation influences the post-founding behaviors of
entrepreneurs, the riskiness of those behaviors, and
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
2Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS
subsequent firm outcomes, notably growth. This is
our focus.
Institutional theory is a useful lens for examin-
ing these issues. Prior research argues that lenient
bankruptcy laws may have a negative effect on new
firm growth. The argument is that such laws are likely
to restrict the financial resources of new firms because
either resource providers perceive too much risk or
they worry that more-skilled individuals will avoid
entrepreneurship since lenient bankruptcy laws may
attract too many competitors (Berkowitz and White
2004, La Porta and López-de-Silanes 2001). A second
argument is that lenient bankruptcy laws may also
attract marginal, overly optimistic individuals to entre-
preneurship. These individuals may have excessively
optimistic assessments of their managerial ability, mar-
ket challenges, and likelihood of a successful firm, all
of which are likely to contribute to high failure rates
among new firms (Hogarth and Karelaia 2012, Lowe
and Ziedonis 2006, Simon et al. 1999). Recent work sug-
gests that lenient bankruptcy laws may restrict capital
and thus reduce rates of patenting (Cerqueiro et al.
2016). Thus, this work argues that lenient bankruptcy
laws may limit new firm growth.
By contrast, other research emphasizes that easier
exit through bankruptcy may increase venture growth
by stimulating “creative destruction.” That is, failing
firms liberate resources such that new firms can use
those resources to address potentially better oppor-
tunities. For example, research finds that firm exits
triggered by shifting regulations and other causes can
release human capital and other resources for produc-
tive reuse (Hoetker and Agarwal 2007). Bankruptcies
may enable innovation because they generate knowl-
edge spillovers from failed firms that provide grist for
innovation by new firms (Knott and Posen 2005). Thus,
this work suggests that lenient bankruptcy laws may
increase new firm growth.
One approach to resolving this puzzle is to use
institutional theory to consider more precisely the im-
plications of an institutional change that eases the
stringency of bankruptcy laws. This change is likely
to influence the legitimacy, uncertainty, and risk con-
cerns that heterogeneous types of individuals per-
ceive and face. That is, this approach may be fruit-
ful because different types of individuals are likely
to have distinctive legitimacy concerns, as well as to
perceive and act differently in the face of uncertainty
and risk. Moreover, such institutional change may dis-
proportionately affect elite individuals because they
pay greater attention to legitimacy issues than oth-
ers do (Eesley 2016). Building on this work, we argue
that easier exit via lenient bankruptcy law may attract
more capable entrepreneurs who might otherwise per-
ceive better outlets for their talents when bankruptcy
laws are strict. Thus, if bankruptcy leniency in partic-
ular influences more capable individuals to start firms
because of lowered financial and legitimacy costs in
the face of bankruptcy, an institutional change toward
more lenient bankruptcy laws is likely to result in
increased new firm growth. We address this possibility
by asking, do institutional changes that that lower failure
barriers increase venture growth by attracting more capable
We examine our research question in Japan. We
exploit a quasi-natural experiment in which the
Japanese government enacted a substantial institu-
tional change to the bankruptcy laws that made them
more lenient. Prior to this reform, Japan’s bankruptcy
laws were among the most stringent in the world (Imai
and Kawagoe 2000, Lee et al. 2011, Schaede 2008).
We examine longitudinal data on 24,624 Japanese
firms over a 10-year period to determine the effects of
this bankruptcy reform on bankruptcy rates, found-
ing rates, and new firm growth. A key feature of our
study is its examination of whether this reform par-
ticularly affected those whom we term “elite” individ-
uals (i.e., those with high human and social capital)
who research identifies as especially likely to found
high-growth ventures (Baum and Bird 2010, Eesley and
Roberts 2012, Eisenhardt and Schoonhoven 1990). We
supplement these analyses with qualitative interview
data with Japanese entrepreneurs, investors, govern-
ment officials, and a venture attorney well versed in
Japanese bankruptcy law.
We provide contributions at the nexus of institu-
tional theory and entrepreneurship research. First, we
unpack the effects of a significant institutional change,
lowering failure barriers via bankruptcy reform. Prior
research finds that easier failure via lenient bankruptcy
laws increases firm founding (Armour and Cumming
2008, Lee et al. 2011). We add to institutional theory
by showing that this institutional change also leads
to higher-growth ventures. Thus, lowering the barriers
to failure is a significant institutional lever for stimu-
lating the formation of superior firms, not just more
firms. Shifts in legitimacy, uncertainty, and perception
of risk underlie this result. Second, we add to institu-
tional theory by providing new insights into how insti-
tutional change affects different types of individuals. Prior
research at the nexus of institutional theory and entre-
preneurship finds that lowering barriers to entry and
barriers to growth differentially affects different types
of individuals (Eesley 2016, Sine et al. 2005). We extend
this work by indicating that lowering barriers to failure
(i.e., making exit easier) also affects different types of
individuals in distinctive ways. Specifically, this insti-
tutional change increases entry by elite entrepreneurs,
who are then more likely to launch high-growth ven-
tures. Third, we contribute to institutional theory by
bringing venture growth to center stage. We therefore
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS 3
better align with the aims of entrepreneurs and policy
makers by presenting the view that institutions can sig-
nificantly shape the performance of new firms, not just
their founding and death. Finally, we provide insight
into how an institutional change can affect the funda-
mental interrelationships among failure, founding, and
growth that are central to flourishing economies—i.e.,
the dynamics of creative destruction. We conclude that
making failure easier (via lenient bankruptcy regula-
tion) increases the likelihood that more capable indi-
viduals will become entrepreneurs and build more
successful firms.
Theoretical Background
Our research question focuses on understanding how
institutional changes that lower failure barriers affect
venture growth. We specifically expect that low-
ering failure barriers via bankruptcy reform will
alter the legitimacy, uncertainty, and risk faced by
entrepreneurs and thus shape their perceptions of risk,
affect their behavior, and influence the growth of their
ventures. In support, research on institutional com-
plementarity indicates that behaviors are affected by
both direct institutional arrangements such as entry
barriers and institutions such as bankruptcy regulation
that govern the trajectory of firms (Aoki 1994). Several
research streams at the nexus of institutional theory
and entrepreneurship offer additional insights.
A first stream examines how institutional changes
such as regulatory reforms alter barriers to entry, and
so affect the rate of venture founding. For example,
a cross-industry study of Massachusetts Institute of
Technology alumni found that public policy changes
that streamlined the business environment were an
important determinant of increased entrepreneurship
rates (Hsu et al. 2007). Similarly, a related study of solar
power firms showed that simplifying regulations moti-
vated increased founding rates because it reinforced
favorable social perceptions of entrepreneurship (Meek
et al. 2010). Other institutional changes lower barriers
to entry by opening previously restricted markets. For
instance, when the U.S. government passed the Public
Utility Regulatory Policy Act of 1978, it became legal
for independent energy firms to sell electricity to the
grid. This, in turn, spurred the founding of energy
ventures by providing legitimacy to the new sector
that attracted investor (Sine and David 2003). Overall,
this research supports the conclusion that institutional
changes including regulatory reform can alter barriers
to entry and, thus, rates of venture founding.
A second stream finds that institutional changes
such as regulatory shifts may differentially affect par-
ticular types of individuals (Armanios et al. 2016,
Eesley 2016). That is, the effects of institutional changes
may be moderated by the characteristics of individu-
als such as their values, risk aversion, and cognitive
frames (Pacheco et al. 2014, Scott 2008). A study of the
“green” building supply industry, for instance, found
that the effects of a regulatory change on stimulating
firm founding were more influential for members of
particular social groups than for employees of incum-
bent firms (York and Lenox 2014). In another example,
the situational factors facing particular entrepreneurs
affected the influence of new regulations on uncer-
tainty (Sine et al. 2005). Regulatory changes that
prompted entry into new types of power generation
were more encouraging to “greentech” entrepreneurs,
who faced greater market and technical uncertainty,
than to “browntech” entrepreneurs, who faced less
uncertainty. In sum, the implication is that institutional
changes may differentially affect particular types of
A third stream examines the interplay between
founding and failures. Here, the argument is that fail-
ures free up resources to be redeployed into potentially
more productive uses. In other words, firm failures
may increase founding rates by opening new market
opportunities and recycling resources. For example,
new firms in the disk drive industry frequently reused
the assets of failed firms, such as their technological
innovations, to their advantage (Hoetker and Agar-
wal 2007). In an example of institutional change, the
enactment of prohibition laws in the United States led
to the demise of breweries, as expected (Hiatt et al.
2009). However, this institutional change also triggered
the entry of new soft drink firms, as entrepreneurs
exploited the market opportunity for new beverages
created by the exit of alcoholic beverage producers.
Some entrepreneurs also “recycled” brewery resources
into their new manufacturing facilities, thus easing
the resource requirements of their new firms. Over-
all, failures can stimulate firm founding by opening
new opportunities, enabling knowledge spillovers, and
making additional resources available.
A final stream focuses on the institutional con-
text that is shaped by bankruptcy laws—laws that
govern the protections, sanctions, costs, and liabili-
ties of failing firms and individuals. In some coun-
tries, bankruptcy laws are lenient, such as Chap-
ter 11 bankruptcy in the United States (Altman 1993).
By contrast, bankruptcy laws in countries such as
Sweden, Chile, and Thailand are stringent (La Porta
and López-de-Silanes 2001). Nations with stringent
bankruptcy laws (e.g., little personal asset protection
for entrepreneurs, long and complex bankruptcy pro-
cess) have lower rates of venture formation (Armour
and Cumming 2008, Peng et al. 2009). Similarly, United
States offer more asset protection in bankruptcy have
higher rates of entrepreneurship (Fan and White 2003,
Georgellis and Wall 2002). Underlying these results is
the insight that the stringency of bankruptcy laws can
influence the risks that entrepreneurs perceive and so
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
4Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS
can alter their willingness to start firms. Overall, more
lenient bankruptcy laws are associated with higher
rates of venture formation.
Taken together, these studies indicate that institu-
tional changes can influence barriers to entry, en-
courage venture formation, and differentially affect
particular types of individuals. In addition, failures
can stimulate founding, thus creating a regenerative
economy. Finally, institutional changes that make exit
via bankruptcy easier are likely to affect the risks
that potential entrepreneurs perceive and so influence
whether they start firms. A next step in advancing insti-
tutional theory is to unpack how (if at all) a major
institutional change, enacting lenient bankruptcy law,
disproportionately influences the behaviors of elite
individuals (i.e., those with high human and social
capital) and their firms’ outcomes.
Our core argument is that an institutional change that
lowers barriers to failure via more lenient bankruptcy
regulation (what we term “bankruptcy reform”) will
affect the entrepreneurial behaviors of elite individ-
uals and the outcomes of their firms more than
those of other individuals. As noted earlier, we define
elite individuals as those people with greater human
capital, such as intelligence, education, and work expe-
rience, and greater social capital, such as connections
to resource providers and high-status others. Since
such people are more likely to form high-performing
ventures, they are an important—perhaps the most
important—target of entrepreneurship policy initia-
tives that aim to enhance economic expansion (Burton
and Beckman 2007, Chatterji 2009, Gulati and Higgins
2003). While we could study other sources of hetero-
geneity across individuals, whether an individual is
elite is highly relevant for the efficacy of institutional
changes aimed at economic growth through entrepre-
neurship. Finally, as we argue next, bankruptcy reform
particularly influences the entrepreneurial behaviors
of elites.
Entering Bankruptcy
We begin by considering the direct effects of an in-
stitutional change that lowers failure barriers via
bankruptcy reform. These include both financial and
legitimacy effects. This enables us to unpack mecha-
nisms that underlie the effects of bankruptcy reform
and to examine whether this reform influences entry
into bankruptcy disproportionately among elites. To
begin, an individual’s decision to enter bankruptcy can
be thought of as the choice between the cost and ben-
efits of operating a marginally performing firm versus
the cost and benefits to the best alternative employ-
ment minus bankruptcy costs. First, bankruptcy costs
include financial losses such as the loss of personal
assets (e.g., personal bank account, home, vehicles) and
the expenses due to a long and complex bankruptcy
process (e.g., court costs, legal fees) (Fan and White
2003, Lee et al. 2011). Second, bankruptcy costs also
include personal legitimacy losses such as the negative
effects of a diminished personal reputation as a result
of managing a failed firm. For example, research finds
that declaring bankruptcy causes audiences to stigma-
tize the bankrupt firm (Sutton and Callahan 1987). This
stigma may delegitimize both the firm and its man-
agers by inflicting reputational damage. Overall, the
argument is that as bankruptcy costs fall, individuals
are more likely to choose bankruptcy for a marginal or
failing firm.
Second, bankruptcy costs are likely to be partic-
ularly high for elites. For example, the loss of per-
sonal legitimacy may limit future opportunities, and
so bankruptcy has a particularly high cost for elites.
Wiesenfeld et al. (2008) found that elite managers of
bankrupt firms suffer significant devaluation of their
reputations by social, legal, and economic arbiters who
seek to make sense of a company’s failure. Deval-
ued reputation diminishes the personal legitimacy of
mangers, and thus it may limit access to future oppor-
tunities and damage their social ties. This is particu-
larly consequential for elites because they are likely to
have more attractive opportunities when their reputa-
tions are undamaged than nonelites. These opportu-
nities include both employment in established firms
and opportunities to start other firms. Elite individ-
uals are also more likely to seek skilled positions
higher in an organization’s hierarchy, where compe-
tition is substantial. Compared with nonelites, being
stuck in a long, delegitimizing bankruptcy process is
thus particularly costly to elites. In sum, elites face
substantial bankruptcy costs (both financial and legit-
imacy) that can significantly affect their decision to
enter bankruptcy prior to reform.
As bankruptcy costs fall with bankruptcy reform,
individuals reach a threshold of bankruptcy costs
versus the costs to maintain a firm below which
they choose bankruptcy (Gimeno et al. 1997). While
bankruptcy reform is likely to increase entry into
bankruptcy for all individuals, the reform particularly
affects elite individuals for several reasons. First, the
bankruptcy costs of elites drop more significantly after
reform. Financial costs fall because lenient bankruptcy
regulations typically offer greater protection of per-
sonal assets and shorten the length of the bankruptcy
process, allowing elites to take advantage of this
enhanced asset protection and to move more quickly
into alternative income sources. Legitimacy costs are
also attenuated after reform. Bankruptcy reform sig-
nals altered normative beliefs about the negatives of
business failure and makes them more favorable. Fur-
thermore, since reputational damage is diminished as
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS 5
bankruptcy becomes more widespread and thus more
acceptable, this decrease particularly affects elite indi-
viduals since they have greater social capital and rep-
utation at risk. A key implication of lower bankruptcy
costs is then that the decision to enter bankruptcy now
depends more on the comparison between the finan-
cial and legitimacy costs of operating a marginally per-
forming firm and the financial and legitimacy benefits
of the best employment alternative. Thus, bankruptcy
reform changes the decision calculus of elites by
decreasing the relevance of bankruptcy costs, includ-
ing lost legitimacy, and increasing the relevance of
comparison with other employment alternatives.
By contrast, in the case of nonelites, their bankruptcy
costs prior to reform are relatively less. For example,
they are likely to have fewer personal assets to lose
than elites because they may have had less lucrative
employment opportunities and so have less opportu-
nity to accumulate assets. Moreover, nonelites are also
likely to have a greater portion of their assets invested
in their firms rather than in other asset classes such
as stocks, land, and bonds. Since assets invested in
a bankrupt firm are likely to be lost, some benefits
offered by lenient bankruptcy law are unrealized for
nonelites. Nonelites are also likely to have fewer legit-
imacy costs related to bankruptcy. Nonelites are more
likely to have opportunities in lower-skilled occupa-
tions or at lower organizational levels where their track
record of managing a business failure is less conse-
quential and where less personal legitimacy is needed
to access such opportunities.
The nonelites’ bankruptcy costs (both financial and
legitimacy) are relatively lower, so their decision to
enter bankruptcy depends more on the comparison
of operating a marginal firm versus entering the best
alternative employment than it does for elites. The
key point is that bankruptcy reform does not sub-
stantially change this comparison for nonelites, and
thus reform has less influence on their decision to
enter bankruptcy. Although bankruptcy reform may
motivate nonelite individuals to enter bankruptcy, its
effects are greater on the choices of elites because
elites’ decrease in bankruptcy costs is likely to be sub-
stantially higher. Overall, we expect that bankruptcy
reform will encourage individuals—particularly elite
individuals—to enter bankruptcy.
Hypothesis 1 (H1). A lenient bankruptcy reform will
increase the propensity of elite individuals to enter their firms
into bankruptcy as compared with nonelite individuals.
Founding Firms
We now turn to the indirect effects of an institutional
change that reduces barriers to failure by arguing that
bankruptcy reform increases the likelihood of found-
ing, particularly by elite individuals. In contrast to
bankruptcy, the decision to start a firm depends on
whether the returns to entrepreneurship (i.e., starting a
firm) are higher than the returns to wage employment
alternatives. It also depends on whether the founders
have the legitimacy needed to convince stakeholders
such as potential employees and resource holders to
contribute to the new firm. Both financial returns and
legitimacy are affected by uncertain future bankruptcy
costs. In general, as bankruptcy costs (financial and
legitimacy) fall, individuals are more likely to found
While bankruptcy reform increases the likelihood
of founding a firm for all individuals (Peng et al.
2008), the reform particularly affects elite individuals
for several reasons. First, bankruptcy reform changes
the decision calculus of elite individuals by substan-
tially lowering their bankruptcy costs (H1). As these
costs (financial and legitimacy) fall, they become less
relevant to the decision to found a firm. Instead, the
comparison between the returns to entrepreneurship
and wage employment becomes more relevant, and
founding a firm becomes more attractive. Bankruptcy
reform is also likely to reduce the damage to managers’
reputations in the event and so maintain their personal
legitimacy. This facilitates their ability to gather the
needed resources to start a firm and in turn increases
their likely returns to entrepreneurship.
Second, since bankruptcy reform increases firm fail-
ures (H1), it may recirculate resources back into the
economy and so make at least some of these resources
available to form new firms (Hiatt et al. 2009, Hoetker
and Agarwal 2007). This may increase venture growth
by providing more resources to tackle more ambitious
opportunities with greater growth potential (Ahuja
and Katila 2004, Miller and Shamsie 1996). While these
resource advantages are potentially available to every-
one, they are particularly likely to go to elites.1Elites
have better social connections to resource providers
than do nonelites (Hallen and Eisenhardt 2012). Elites
are also likely to have greater legitimacy as a result of
their higher status such that resource providers again
prefer them to nonelites. Elites also are more likely
to have the experience that resource providers prefer
(Eisenhardt and Schoonhoven 1990, Hsu 2007, Stuart
et al. 1999). Prior research supports this view. For
instance, a study of Internet security firms showed that
founders with higher human and social capital (i.e.,
more elite), including more ties to prominent investors,
were more likely to receive venture captial funding
(Hallen 2008). By contrast, nonelites have slower access
to these resources.
Third, complimentary effects of bankruptcy reform
also exist. For instance, falling bankruptcy costs
expand the number of viable entrepreneurial oppor-
tunities, making some previously unacceptable
entrepreneurial opportunities acceptable. That is,
falling bankruptcy costs lower the hurdle that the
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
6Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS
returns to entrepreneurship for any given opportunity
must overcome. Thus, we expect elite individuals to
have a greater number of viable entrepreneurial oppor-
tunities and so to be more likely to start firms after
the reform than before. In addition, we expect that
elite individuals will be more motivated to search for
entrepreneurial opportunities after reform. They will
recognize that entrepreneurship is now relatively more
attractive than it was prior to reform. Thus, elites will
be more likely to search for entrepreneurial opportu-
nities that offer superior returns to wage employment,
more likely find a suitable opportunity given more
opportunities, and thus more likely to start a firm.
By contrast, and as argued above (H1), the bank-
ruptcy costs of nonelites are relatively lower than those
of elites. A nonelite’s decision to found a firm does
not change much before and after reform. The num-
ber of viable entrepreneurial opportunities remains
relatively more similar. This similarity, in turn, is
unlikely to stimulate greater interest in searching for
entrepreneurial opportunities since the viable set is
relatively unchanged. In addition, nonelites have less
personal legitimacy to lose and are thus relatively less
affected by such losses. We expect, therefore, that the
falling costs of bankruptcy will have less effect on the
decision of nonelite individuals to start firms.
Another reason that bankruptcy reform particu-
larly influences the decision of elite individuals to
start firms relates to the psychology of decision mak-
ing under uncertainty. Before reform, elite individu-
als who contemplate starting a firm face the risk of
high bankruptcy costs, including the certain loss of
the returns to employment. This situation likely leads
elite individuals to perceive substantial risk, frame
their decision to start a firm in terms of loss, and
choose employment to avoid that loss (i.e., loss aver-
sion), even if the returns to employment are lower
than the expected value of the returns to entrepre-
neurship. In other words, when elites (or anyone for
that matter) perceive the risk of substantial loss, they
try to avoid that loss (Kahneman and Tversky 1979,
Thaler et al. 1997).
By contrast, after reform, elite individuals no longer
face high and likely bankruptcy costs. They are more
likely to perceive diminished risk; frame their decision
to start a firm in terms of the gains to entrepreneur-
ship, not losses; and so be more likely to start a firm
(Kahneman and Tversky 1979). Thus, elites are more
likely to found firms after bankruptcy reform because
they are more likely to alter their perception of risk
from loss to gain and become more willing to enter
a low-probability entrepreneurship “lottery.” Research
supports this argument. For example, in a study of
new U.S. energy firms, potential entrepreneurs who
faced a greater likelihood of loss as a result of technical
uncertainty were more affected by government poli-
cies that reduced uncertainty than were other poten-
tial entrepreneurs (Sine et al. 2005). Our interviews of
Japanese entrepreneurs support this argument as well.
The elite chief executive officer (CEO) of a fast-growing
Japanese social networking firm told us,
We might have stayed as employees of our old e-com-
merce firm were it not for what the change in bankruptcy
legislation told us. We decided that if the government
were reducing the penalties of bankruptcy...maybe
what happened to bankrupt firms in the past wouldn’t
happen to us. It really helped us decide to start our
Another elite Japanese founder described it to us in
the following way:
The risk of a bad outcome was very real to us and
needed to be considered carefully. Our decision to go
ahead in 2004 depended on our knowledge that a
bankruptcy, should it be needed, was becoming a more
rational and less costly procedure.
Overall, we expect that bankruptcy reform will
encourage individuals to start firms, especially elites.
This hypothesis enables a test of the differential effects
of bankruptcy reform on the founding rates for elite
versus nonelite individuals.
Hypothesis 2 (H2). A lenient bankruptcy reform increases
the propensity of elite individuals to found a firm as com-
pared with nonelite individuals.
New Firm Growth
We next examine how an institutional change that
reduces barriers to failure via bankruptcy reform influ-
ences venture growth. To begin, research indicates that
elite individuals are likely to found more successful
firms than are nonelites. Their higher human and social
capital provide them with the capabilities that make
them more likely to found firms that can grow to a
larger scale than the firms of others (Eisenhardt and
Schoonhoven 1990, Maurer and Ebers 2006). There
are several reasons for this. One is that individuals
with higher human capital have higher cognitive abil-
ities that allow them to build more successful ven-
tures (Baum and Bird 2010). These individuals also
have access to superior education that further improves
their human capital and likely success (Hsu et al. 2007).
Furthermore, through their education and talent, they
are often better able to gain business experience at
higher managerial levels, which further enhances their
human capital (Beckman and Burton 2008). As a result,
individuals with higher human capital are better able
to choose attractive opportunities and effectively man-
age new firms. For example, firms founded by more
educated, more experienced, and older entrepreneurs
are more likely to enjoy higher growth (Eesley and
Roberts 2012).
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS 7
Similarly, individuals with superior social capital
are also likely to found high-performing firms (Adler
and Kwon 2002). Individuals acquire social capital
via industry experience, professional ties, educational
experience, and social memberships. Individuals with
higher social capital are likely to have more access to
valuable opportunities (Eisenhardt and Schoonhoven
1990). They may learn about opportunities that are
unknown or unavailable to others through their social
networks, and they may have access to better resources,
including complementary assets that are crucial for
growth (Aldrich and Zimmer 1986, Stuart and Soren-
son 2003). Ties to prominent others may also certify
their quality to potential resource providers and sig-
nal that the venture is likely to succeed (Hallen and
Eisenhardt 2012). Thus, the venture is likely to gain
more and better resources that, in turn, improve the
likelihood of superior growth (Maurer and Ebers 2006).
In sum, elite individuals are particularly likely to form
high-growth firms.
Thus far, we have argued that elite individuals are
likely to start firms that grow faster than those founded
by nonelites. We now turn to the implications of
bankruptcy reform. Prior to bankruptcy reform, elite
individuals who become entrepreneurs are likely to
perceive high risk and so focus on avoiding losses (H2).
Thus, they will choose only extraordinary opportu-
nities that fit their skills and social connections, and
they manage them conservatively in order to avoid
bankruptcy. While this approach avoids failure, it
likely puts a ceiling on growth by avoiding riskier (but
perhaps higher payoff) opportunities and more ambi-
tious strategies.
By contrast, after bankruptcy reform, elite individ-
uals face sharply lower bankruptcy costs. As argued
in H2, elite individuals are likely to face less risk and
may shift their perception of risk from loss to gain.
In addition, they are likely to become more willing
to engage in risky behavior, such as choosing riskier
but potentially higher payoff opportunities, and to
manage those opportunities aggressively in ways that
more fully exploit their skills and social connections
(Kahneman and Tversky 1979, Thaler et al. 1997). Also
as argued in H2, elite individuals are likely to have dis-
proportionately better access to the increased resources
that bankruptcy reform may make available (Hallen
2008, Hsu 2007). This again disproportionately enables
elite individuals to pursue bigger and more ambitious
Furthermore, as we argued in H2, when elites suffer
delegitimization, they lose access to opportunities and
suffer diminished value of their social ties (Wiesenfeld
et al. 2008). Following this reasoning, changing the
stigma of failure through bankruptcy reform enables
individuals to preserve their legitimacy and so con-
tinue to have access to superior resources and oppor-
tunities. Overall, bankruptcy reform likely encour-
ages elite individuals to perceive less risk; enables
them to maintain their legitimacy; and allows them to
address more ambitious opportunities, obtain superior
resources, and exploit their skills and social connec-
tions more fully. This, in turn, is likely to create higher-
growth firms than it would have before reform.
By contrast, bankruptcy reform has less influence
on nonelites. Their situation after reform is more sim-
ilar to their circumstances before reform than is the
case for elite individuals. Even if nonelites choose
riskier and more ambitious opportunities, the growth
of their firms is restrained by the limitations of their
lesser human and social capital (Baum and Bird 2010,
Eisenhardt and Schoonhoven 1990). Thus, nonelite-
led firms are affected less by bankruptcy reform. The
growth of their firms is more constrained by their
relatively more limited skills, legitimacy, access to
resources, and opportunities as compared with elite
In sum, bankruptcy reform has two effects on ven-
ture growth. First, it attracts more elites, who are more
likely to create higher-growth firms than are others,
to start firms. Second, it encourages elite individuals
to address bigger and more ambitious opportunities
with more aggressive strategies that exploit their supe-
rior human and social capital more fully. By contrast,
nonelite individuals face a relatively unchanged situa-
tion before and after reform. Moreover, while they may
also address bigger opportunities more aggressively,
they are still constrained by the limitations of their
human and social capital. These arguments suggest the
following hypothesis.
Hypothesis 3 (H3). A lenient bankruptcy reform increases
the propensity of elite individuals to found higher-growth
firms as compared with nonelite individuals.
We analyzed bankruptcy, venture formation, and
growth in Japan over a 10-year span from 1998 through
2007. This period spans the comprehensive Japanese
bankruptcy reform during 2003 and provides a quasi-
natural experimental setting in which to test our
hypotheses. Japan is also an attractive setting because
Japanese corporate law is based on the laws in the
United States and other advanced economies, increas-
ing the generalizability of our study (Vogel 2006).2
Our sample consists of firms in the COSMOS2 database
from Teikoku Databank, Ltd. (TDB). TDB is a com-
mercial credit rating firm and one of two firms in
Japan providing credit ratings to corporate clients.
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
8Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS
Since Japanese firms rely on this database to evalu-
ate supplier and customer credit, it is a comprehensive
and accurate source for capturing firms with commer-
cial activity. TDB data are frequently used by scholars
(Miyamoto and Rexha 2004, Schaede 2008, Singleton
and Globerman 2002).
We use the 2009 COSMOS2 database that includes
all incorporated Japanese firms that TDB investigated
for creditworthiness. As noted above, it is a com-
prehensive source for data on commercial Japanese
firms. Each firm record includes data such as found-
ing date, location, number of employees, sales rank,
key customers, and demographic data on founders
and directors. The records also include financial data
spanning the most recent three years. We selected all
firms from COSMOS2 that were incorporated from
1998 through 2007 that were (a) not owned by another
company, (b) organized as a corporation, and (c) oper-
ating as indicated by employees or revenues. We start
our observations in 1998 following the Japanese gov-
ernment’s mid-1990s implementation of regulatory
changes that affected banks and the governance of
large corporations. We end our observations at the start
of the global financial crisis at the end of 2007, provid-
ing a 10-year window.3We augment this sample with
all firms from the TDB Bankruptcy database for 1998–
2007. Each record in this bankruptcy database is similar
to the COSMOS2 database and includes bankruptcy-
filing information, such as receiver banks and debts
owed. The combined sample consists of 24,642 firms.
While our sample is taken from a comprehensive and
accurate source (as stated above), there are missing
data. The number depends on the analysis and is low
(less than 10% of the sample). Our data also do not
include very early firms that fail before incorpora-
tion. To verify that eliminating firms with incomplete
data from our analyses does not materially affect our
results, we compare key characteristics of included and
excluded firms such as current revenue, number of
employees, capital, CEO age, and industry composi-
tion. We find no significant differences (p>0.05).
Our sample strategy, by construction, omits firms
that exit for nonbankruptcy reasons such as acquisi-
tion. Bias is unlikely because these events are rare,
accounting for less than 0.01% of our sample. They
are rare because regulations during our study required
a two-thirds shareholder vote and prohibited trian-
gular mergers (Arikawa and Miyajima 2007). Finally,
we enrich our understanding via interviews. We inter-
viewed founders and CEOs of Japanese ventures, a
Tokyo venture attorney well versed in bankruptcy
law, an official of the Hiranuma Plan (a program
to encourage Japanese technology start-ups), and a
director-general of the Ministry of Economy, Trade and
Dependent Variables
For H1, our dependent variable captures the bank-
ruptcy event. We organize our data into a panel of
annual observations. For each firm-year, the dichoto-
mous variable, bankruptcy, is coded 1 in the year that a
firm declares bankruptcy; otherwise, it is 0. For H2, our
dependent variable captures the founding of a venture.
We organize our data into a panel of annual obser-
vations. For each firm-year, the dichotomous variable,
found, is coded 1 in the founding year; otherwise, it is 0.
For H3, we assess venture growth using the com-
pound annual sales growth rate since founding, growth,
as the dependent variable. We use sales growth as
our dependent variable for several reasons. First, sales
growth is a performance measure that is relevant
across industries. Indeed, revenue is an antecedent out-
come to other key financial measures, such as prof-
itability and cash flow, and is ultimately necessary for
survival. Second, sales growth is particularly relevant
here because it drives economic expansion and job
creation—aims of many policy initiatives for new firms
(Ahlstrom 2010). Third, sales growth is more accurately
measured than is profitability because of challenges in
measuring the denominator of the profitability ratio,
such as assets. Fourth, sales growth is particularly ger-
mane to venture performance because it captures how
quickly a firm gains traction with customers. Survival,
by contrast, is a coarse performance measure that cap-
tures only a firm’s existence, not the quality of its per-
formance. For example, it cannot distinguish between
“living dead” and thriving firms. It is also more appro-
priate than employee growth because of wide differ-
ences in numbers of employees across industries. Not
surprisingly, sales growth is a frequently used mea-
sure of venture performance (Baum and Bird 2010,
Hmieleski and Baron 2009). Finally, growth fits the
logic of our arguments supporting H3. Consistent with
past research, we use a standard compound annual
growth rate calculation that takes the nth root of the
total percentage growth rate, in which nis the num-
ber of years in the period considered using a beginning
value of ¥1. We compute this measure for 2007 or, in
the case of a bankrupt firm, for the bankruptcy year.
Independent Variables
Elite Founders. A key predictor variable is whether
the founding CEO is an “elite” individual. Combin-
ing the TDB COSMOS2 and TDB Bankruptcy data, we
measure whether an individual founder is elite by her
graduation from one of the top 10 Japanese universi-
ties. We define a binary variable, elite alum, as 1 if the
founding CEO graduated from one of these universi-
ties and 0 otherwise. These top 10 universities are the 7
former imperial universities (Tokyo, Kyoto, Hokkaido,
Tohoku, Nagoya, Osaka, and Kyushu), Tokyo Institute
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS 9
of Technology (top engineering university), and 2 lead-
ing private universities (Keio University and Waseda
University). These 10 are consistent with the top
Japanese universities named in well-known rankings
(Times Higher Education 2011).
Reform Year. We use 2003 as the focal date of bank-
ruptcy reform because the bankruptcy law reform was
fully effective in that year. Prior to bankruptcy reform,
executives at bankrupt firms were dismissed from their
positions, exposed their personal assets to forfeiture,
endured a lengthy and costly bankruptcy process, and
suffered severe damage to their personal reputations
(Xu 2007).
Bankruptcy reform efforts began in 2000 as part of a
broad effort to invigorate Japan’s economy by closing
moribund firms and encouraging new firms to start
(Helwege and Packer 2003). First, the Minji Saisei Ho
(Civil Rehabilitation Law) became effective in 2000 and
replaced the older Wagi Ho (Composition Law) that
governed bankrupt firms. The purpose of this reform
was to permit incumbent managers to hold off cred-
itors while they reorganized their firms—similar to
and inspired by the U.S. Chapter 11 bankruptcy law.
In addition, extensive personal liability protection and
personal asset protection such as for homes, cash, and
unpaid salary were included in the Minji Saisei Ho
(Hirose 2009, Yanagida 2014).
Notwithstanding these initial efforts, these 2000
reforms were ineffective. The principal obstacle was
that the Japanese courts infrequently allowed bankrupt
firms to take advantage of these reforms and instead
forced firms to liquidate under a different law—the
Hassan Ho (Bankruptcy Law)—thereby losing the new
protections and provisions afforded by the 2000 law.
The result was that the bankruptcy reforms of 2000
had little substantive influence on bankruptcy leniency.
Rather, bankruptcy remained harsh (Foote 2011).
In 2003, three key events made bankruptcy reform
actually effective in that year. First, the Minji Sai-
sei Ho was amended to encourage the courts to
approve bankruptcy plans (and take away their power
to block bankruptcy), and to provide streamlined
bankruptcy procedures for small firms. Second, a new
law, the Kaisha Kousei Ho (Corporate Reorganiza-
tion Law), extended these bankruptcy reforms to all
firms. These amended and new regulations eased, sim-
plified and accelerated the bankruptcy process, and,
most importantly, limited judicial interference. They
thus enabled the realization of the protection fea-
tures of the 2000 law. Third, the Industrial Revital-
ization Corporation of Japan was established in April
2003 to ease the financial burdens on executives going
through bankruptcy. Overall, the 2003 reforms made
the Japanese bankruptcy process more lenient by both
enabling the earlier 2000 reforms to take effect (notably
by blocking the power of the courts) and by adding fur-
ther protections and simplifications to the bankruptcy
process. The streamlined procedures such as automatic
stays of creditor action reduced the time to gain a res-
olution in bankruptcy. This saved the direct costs of
lawyers and court fees as well as the costs of manage-
rial distraction and lost legitimacy (Mori 2006, Ohara
2015).4Finally, later legislation consolidated these laws
but did not add new provisions. In sum, since the ini-
tial bankruptcy reforms of 2000 only became relevant
with the 2003 reforms, we construct a dichotomous
variable, post reform, that is set to 0 for years 1998 to
2002 and 1 for each year after 2002.
Model Specifications and Econometric Issues
Piecewise Hazard Method. For H1and H2, we test the
effects of the 2003 bankruptcy reform on the likeli-
hood of bankruptcy and founding, respectively. We
employ piecewise hazard models for several reasons.
First, hazard models readily account for right censor-
ing, which is important since some firms may enter
bankruptcy after our study (Tuma et al. 1979). Sec-
ond, the piecewise specification allows us to model
the time-varying hazard of bankruptcy and found-
ing explicitly in terms of firm age and observation
time elapsed (for H1and H2, respectively), enabling
us to make inferences about differences in founding
and bankruptcy “hazards” before and after reform. A
final advantage is that piecewise hazard models do not
impose strong parametric assumptions with regard to
the hazard rate (Blossfeld and Rohwer 2002). The piece-
wise hazard method has been used in numerous stud-
ies, including those examining the factors that predict
bankruptcy and those predicting which context effects
influence whether individuals found firms (Bharath
and Shumway 2008, Dobrev and Barnett 2005, Stuart
and Sorenson 2003).
For H1, firms begin being “at risk” of a bankruptcy at
the time of founding and the transition to bankruptcy
is a “failure” indicated by the bankruptcy variable. Since
the likelihood of bankruptcy can change with the age of
the firm, we divide firm age into three time segments:
zero to three, four to six, and seven to nine years. We
then discern the effect of bankruptcy reform on the
bankruptcy of elite-founded firms by the interaction of
elite alum with the post reform variable. Our interest is in
the coefficient of this interaction. Our results are robust
to different time segments as well.
For H2, entrepreneurs start to be at risk of a founding
at the beginning of our observation period, and the
transition to founding is a failure indicated by the found
variable. Since the likelihood of founding can change
in the time before and after reform, we break our time
segments at the 2003 reform year, and we include the
elite alum interaction with the post reform variable to
discern the effect of reform on the founding of elite
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
10 Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS
firms. As before, our interest is in the coefficient of this
Panel Logit Method. As discussed earlier, hazard mod-
els are advantageous because they account for cen-
soring and time-varying covariates with statistical effi-
ciency. However, biased results may occur because haz-
ard ratios can depend on the length of time firms
are observed. In addition, selection errors can occur
if, for example, bankruptcy reform motivated found-
ing in different industries in which firms are more or
less likely to fail (Hernán 2010). To account for these
possibilities, we use a second statistical approach—
panel logit models—as a robustness check. This tech-
nique is appropriate because bankruptcy and found-
ing are binary choices, and this technique relaxes
the requirement that independent variables are nor-
mally distributed. This is useful because influences
on bankruptcy and founding can be discontinuous
(Hoetker 2007, Wooldridge 2010).
A key issue is that selection effects may drive the
results because firms in different industries may have
differing bankruptcy risks. To address this, we incor-
porate a coarsened exact matching (CEM) approach.
CEM is a method to preprocess data to reduce the
imbalance between “treatment” and “control” groups
(Blackwell et al. 2009). To do this, we match individual
firms in the treatment and control groups to provide
better covariate balance across ventures. These groups
should be as comparable as possible for a reliable esti-
mation of the effect of the reform. We use the coars-
ened exact matching implementation in Stata to deter-
mine the boundaries of matching. The matching vari-
ables should be relatively unrelated to the reform or
to bankruptcy (i.e., likelihood of getting treatment) but
related to firm growth (i.e., outcome) (Brookhart et al.
2006, Iacus et al. 2012). Accordingly, we matched on
several of a firm’s initial attributes—industry, region,
and capital (measures defined below).
To estimate effects, we used random effects and the
generalized linear model (GLM) regression method,
which accounts for both heteroskedastic errors and
autocorrelation that may arise because we measure
each firm across multiple years (Liang et al. 1992). To
facilitate causal inference further, we employ indus-
try, region, and year fixed effects to control for time-
invariant industry selection choices as well as year-
to-year and region changes that might correlate with
unobserved influences. As discussed in the Results sec-
tion, our findings here are highly consistent with those
of the hazard models.
Difference-in-Differences Estimation. To analyze H3,
we use a difference-in-differences (DID) estimation
method with fixed effects for industry, region, and
time.5A concern is that the unmeasured change that
is often present in the environment is rarely isolated.
For instance, demographic changes are important, as
are foreign trade stability, local business methods, and
the political situation. These can affect venture growth
(Sine and David 2003). DID estimation allows for con-
trol of these and other unobserved effects by consider-
ing unmeasured factors as affecting the treatment and
control groups equally (Campbell 1969, Forman et al.
2009). A key assumption is that the selected control
group is affected by the general conditions but unaf-
fected by the bankruptcy reform. To prove this, we
compare the selected control group to a treated sam-
ple that experiences the effects of bankruptcy reform.
This allows an identification of the focal reform. We
use consulting firms (e.g., engineering, design, archi-
tecture) as our control group because these firms have
minimal tangible resources and little need for debt or
outside capital, thereby reducing concerns about per-
sonal liability and bankruptcy. Yet these firms are also
likely affected by general economic conditions, as they
are for-profit enterprises.
We calculate the differences in outcomes before and
after reform for both groups and compare them. To
accomplish this, we estimate coefficients using GLM
regressions. The DID coefficients of interest are those
for the interaction term, post reform ×consulting that we
then compare to the post reform ×elite alum interaction
We also conducted DID estimations with an alter-
nate control group, primary industries (farming, min-
ing, etc.), as a robustness check. These firms are part
of the overall economy. Yet since they are often fam-
ily owned for generations and have relatively stable
earnings, they are less likely to enter bankruptcy or be
affected by bankruptcy reform. We repeated our DID
analyses with primary industry firms as an alternative
control group, primary, and obtained highly consistent
Macroenvironment Controls. We include several con-
trol variables and annual fixed effects to control for the
general Japanese environment that could affect ven-
ture bankruptcy, founding, and growth. First, for our
H1and H2panel logit analyses and for H3, we con-
trol for background events in any particular year by
including annual fixed effects. In addition, to control
for general monotonic background trends, we include
a variable, years_count, which is equal to the number
of years elapsed since 1998—the starting year of our
Second, we control for the influence of the macroe-
conomic environment because general economic con-
ditions may affect whether individuals start firms and
may affect performance metrics such as sales growth.
In all our analyses, we include an annual time-varying
covariate, GDP growth.GDP growth is calculated as
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS 11
the three-year change in per-capita GDP in constant
2009 yen using data from the Statistics Japan database
(Statistics Bureau of Japan 2011). We take the natural
log to account for skew.
Finally, the level of competition can also affect bank-
ruptcy, founding, and growth. We control for competi-
tive effects with the variable, comp. density, which is the
number of industry firms in the same two-digit Stan-
dard Industrial Classification (SIC) code as the focal
firm (Geroski et al. 2010).
Local Environment Controls. We include a related
control for local social embeddedness, colocated, that is
a binary variable with a value of 1 if the firm’s founder
was born in the same prefecture as the focal firm is
located. Prior literature finds that local social embed-
dedness can affect firm growth (Powell et al. 1996,
Uzzi 1999).
Firm-Level Controls. We control for firm characteris-
tics. Prior literature indicates that firm size can affect
bankruptcy propensity (Thornhill and Amit 2003). To
account for this, in H1, we control for number of
employees and revenue, as measured by the log of the
employee count and the log of revenue recorded for
2008 or in the year of bankruptcy, respectively.
We control for initial industry, industry, in all anal-
yses with industry fixed effects. Our industry classifi-
cation follows past literature using SIC codes for each
firm obtained from the TDB data. Since SIC codes
can be inconsistently assigned and are too numerous
for industry-level analysis, we follow prior research
by using the first two-digit SIC code and then group-
ing these into logical bins (Folta and O’Brien 2003).
Thus, we classify sample firms into one of several
industry categories: primary,manufacturing, construc-
tion, technology, consulting, sales, service, and finance
(nonbank). (Coding is available from the authors.)
Matching Variables. For all three hypotheses, we em-
ploy a coarse exact matching strategy to control for
initial endowment and regional advantages. First, we
match firms to the region they were founded. We
do this because the effects of local communities on
firm objectives and actions to explain outcomes have
been explored in prior literature (Granovetter 1985,
Mizruchi and Fein 1999). Furthermore, elite founders
may cluster in regions near their alma maters or fol-
low opportunity to metropolitan regions manifesting
localized effects that we need to control. Accordingly,
we match firms on the prefecture, region, where their
headquarters are located based on the TDB database.
We also match firms based on their initial endow-
ment of capital. Research makes it clear that differ-
ing levels of initial resource endowments are likely to
affect firm growth and the likelihood of bankruptcy
(Cooper et al. 1994). We use the log of starting capi-
tal, initial_capital (log), as a matching attribute to con-
trol initial endowment effects. This is given in constant
2009 yen.
We use CEM because exact matching drops many
observations owing to curse-of-dimensionality issues.
The idea of CEM is to temporarily group the variables
under study into meaningful bins and perform an exact
match on those bins while retaining the uncoarsened
values (Blackwell et al. 2009). This method bounds
the maximum imbalance between groups through the
ex ante selection of sensible bins. For our case, because
initial_capital (log) is a continuous variable without nat-
ural cut points, we bin initial capital into evenly spaced
bins using the Scott algorithm in Stata. For regional
coarsening, we separate prefectural locations into the
eight standard Japanese regions (chiho) to which they
We first present descriptive statistics for firm found-
ing in Table 1, and pairwise correlations in Table 2.
Table 1shows that firms founded before the reform
grew at an average annual cumulative rate of 0.72%,
while ventures founded after the reform (when elites
were a greater proportion of founders) grew at faster
rate—i.e., an average annual cumulative rate of 1.11%.
This difference is statistically significant (p<0.001).
Elite founders comprise 3.5% of all founders before
the reform, but that percentage nearly doubled to 6.8%
after the reform (p<0.001).6These univariate results
support the notion that bankruptcy reform increased
both the proportion of elite founders and venture
The correlation matrix (see Table 2) indicates low
correlations except between comp. density and con-
struction firms. We ran separate regressions for these
variables, and our results were unaffected. We also
analyzed the variance inflation factors (VIFs) for all
independent variables, and all were less than 5.0, indi-
cating that these variables are unrelated and multi-
collinearity is not a concern.
Turning to our multivariate analyses, we first exam-
ine whether the focal bankruptcy reform increased
bankruptcy likelihood, especially among elite-founded
firms (H1). Table 3shows the results using a piecewise
hazard analysis. The coefficients are hazard ratios, and
as such, values greater than 1 indicate an increased
hazard of bankruptcy. Model 3-1 includes controls
only. As expected, firms with more revenue and
employees are less likely to declare bankruptcy. In
addition, as expected, greater rates of new competitive
entry are associated with greater bankruptcy.
Model 3-2 adds the reform variables. The interaction
variable of interest, post reform ×elite alum, is added in
models 3-3 and 3-4. (Model 3-4 is without founding
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
12 Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS
Table 1. Descriptive Statistics—Foundings
Pre reform Post reform t-Statistics
Variable Mean NStd. error Mean NStd. error t
elite alum 0.035 785 0.001 0.068 410 0.003 13.041∗∗∗
Industry proportions
primary 0.009 200 0.001 0.005 30 0.001 2.002
manufacturing 0.100 2,222 0.002 0.120 598 0.004 4.566∗∗∗
construction 0.248 5,516 0.003 0.174 1,036 0.005 11.671∗∗∗
technology 0.044 971 0.001 0.011 497 0.001 16.521∗∗∗
consulting 0.023 512 0.003 0.031 234 0.002 1.361
sales 0.323 5,341 0.005 0.337 2,139 0.006 2.381∗∗
service 0.146 3,239 0.002 0.204 1,161 0.005 12.479∗∗∗
finance 0.025 562 0.001 0.012 366 0.001 6.509∗∗∗
Cumulative average 0.007 18,563 0.001 0.011 6,061 0.001 15.225∗∗∗
p>0.050;∗∗p>0.010;∗∗∗ p>0.001.
Table 2. Correlation Matrix
(1) (2) (3) (4) (5) (6) (7) (8)
(1) GDP growth 1.000
(2) colocated 0.000 1.000
(3) employees 0.000 0.197 1.000
(4) revenue 0.000 0.267 0.607 1.000
(5) elite alum 0.000 0.050 0.041 0.024 1.000
(6) comp. density 0.000 0.096 0.041 0.012 0.012 1.000
(7) post reform 0.226 0.000 0.000 0.000 0.000 0.000 1.000
(8) primary 0.000 0.041 0.010 0.035 0.012 0.065 0.000 1.000
(9) manufacturing 0.000 0.047 0.083 0.031 0.011 0.145 0.000 0.029
(10) sales 0.000 0.048 0.081 0.036 0.012 0.258 0.000 0.064
(11) service 0.000 0.003 0.083 0.063 0.007 0.038 0.000 0.039
(12) construction 0.000 0.132 0.033 0.049 0.013 0.423 0.000 0.051
(13) finance 0.000 0.019 0.025 0.040 0.008 0.023 0.000 0.015
(14) technology 0.000 0.047 0.029 0.049 0.005 0.035 0.000 0.007
(15) consulting 0.000 0.016 0.046 0.075 0.030 0.026 0.000 0.015
(9) (10) (11) (12) (13) (14) (15)
(9) manufacturing 1.000
(10) sales 0.247 1.000
(11) service 0.151 0.247 1.000
(12) construction 0.196 0.438 0.268 1.000
(13) finance 0.058 0.013 0.079 0.103 1.000
(14) technology 0.067 0.156 0.095 0.058 0.015 1.000
(15) consulting 0.056 0.057 0.078 0.101 0.030 0.015 1.000
year fixed effects.) The positive and significant coeffi-
cient on the post reform variable (p<0.001) in all models
indicates that bankruptcy reform increased the like-
lihood of bankruptcy after reform. The positive and
significant coefficient on the post reform×elite alum vari-
able (p<0.001) confirms that the reform particularly
affected elite-founded firms. In models 3-5 and 3-6, we
regress the elite and nonelite populations separately
to directly compare their likelihood of bankruptcy. We
use t-tests to confirm that the post reform coefficient
for elite firms (model 3-5) is significantly greater than
that for nonelite firms (model 3-6) (p<0.001). Over-
all, the piecewise hazard analysis results support H1.
As a robustness check, Table 1 in the online appendix
presents the CEM panel logit analysis for H1. Overall,
these panel logit results add robust support to H1.
We next examine whether bankruptcy reform
increased founding likelihood, especially among elite-
founded firms (H2). Table 4shows the results using a
piecewise hazard analysis. Similar to Table 3, the coef-
ficients are hazard ratios, and as such, values greater
than 1 indicate an increased hazard of bankruptcy.
Because we are comparing the likelihood of found-
ing across pre- and post-reform time spans, we check
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS 13
Table 3. Piecewise Exponential Models on Transition to Bankruptcy
3-1 3-2 3-3 3-4 3-5 3-6
Dependent variable bankruptcy
Piecewise regressions: Controls Reform Reform Reform Elite-founded Nonelite-
Variables only variable interaction interaction firms founded firms
Firm age
0–3 years 0.004∗∗∗ 0.013∗∗∗ 0.014∗∗∗ 0.013∗∗∗ 0.100∗∗ 0.012∗∗∗
(0.001) (0.004) (0.004) (0.004) (0.000) (0.004)
4–6 years 0.028∗∗∗ 0.010∗∗∗ 0.011∗∗∗ 0.107∗∗∗ 0.153 0.009∗∗∗
(0.009) (0.003) (0.003) (0.003) (0.169) (0.002)
7–9 years 0.022∗∗∗ 0.011∗∗∗ 0.012∗∗∗ 0.011∗∗∗ 0.311 0.009∗∗∗
(0.009) (0.005) (0.004) (0.005) (0.393) (0.004)
New firm entry rate
0–3 years 1.888∗∗∗ 1.262∗∗ 1.268∗∗ 1.267∗∗ 1.071 1.280∗∗
(0.140) (0.105) (0.096) (0.106) (0.322) (0.110)
4–6 years 1.220∗∗ 1.377∗∗∗ 1.398∗∗∗ 1.381∗∗∗ 1.035 1.415∗∗∗
(0.091) (0.104) (0.096) (0.104) (0.271) (0.102)
7–9 years 1.322∗∗ 1.339∗∗ 1.348∗∗ 1.344∗∗ 0.944 1.388∗∗
(0.138) (0.140) (0.124) (0.140) (0.258) (0.151)
Local environment controlsa
colocated 2.212∗∗∗ 2.126∗∗∗ 2.024∗∗∗ 2.096∗∗∗ 1.491∗∗∗ 2.103∗∗∗
(0.049) (0.049) (0.044) (0.049) (0.132) (0.051)
comp. density 0.999 1.000 0.999 0.999 0.999 0.999
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
years_count 0.978∗∗∗ 0.912∗∗∗ 0.965∗∗∗ 0.910∗∗∗ 0.964∗∗∗ 0.909∗∗∗
(0.013) (0.014) (0.013) (0.014) (0.008) (0.014)
Firm-level variables
revenue 0.807∗∗∗ 0.808∗∗∗ 0.916∗∗∗ 0.806∗∗∗ 0.916∗∗∗ 0.800∗∗∗
(0.004) (0.005) (0.021) (0.005) (0.021) (0.004)
employees 0.988∗∗∗ 0.942∗∗∗ 0.942∗∗∗ 0.939∗∗∗ 1.037 0.933∗∗∗
(0.012) (0.012) (0.012) (0.012) (0.050) (0.012)
Reform variables
post reform 2.101∗∗∗ 2.063∗∗∗ 2.063∗∗∗ 2.479∗∗∗ 2.088∗∗∗
(0.126) (0.116) (0.125) (0.566) (0.129)
elite alum 1.3361.336
(0.174) (0.179)
post reform ×elite alum 1.34561.344
(0.181) (0.191)
Observations 320,086 320,086 320,086 320,086 12,220 307,866
Firms 24,622 24,622 24,622 24,622 940 23,682
Wald χ262,825∗∗∗ 50,303∗∗∗ 60,596∗∗∗ 50,136∗∗∗ 3,306∗∗∗ 57,537∗∗∗
Founding year effects Yes Yes Yes No Yes Yes
Industry effects Yes Yes Yes Yes Yes Yes
Region effects Yes Yes Yes Yes Yes Yes
Notes. Model 3-4 is without founding year fixed effects. The difference in the post reform coefficients in models 3-5 and 3-6 is significant
aIncludes GDP time-varying controls (not shown).
p<0.05;∗∗p<0.01;∗∗∗ p<0.001 (two-sided tests).
the difference between the coefficient of the 1998–2002
observation span and the coefficient of the 2003–2007
observation span. We note that because founding is
rare, these coefficients are small.
Model 4-1 presents controls only, while models 4-2
and 4-3 add the interaction variables of interest in-
cluding post reform ×elite alum. (Model 4-3 is without
founding year fixed effects.) The significantly higher
coefficient in the post-reform observation span on
founding (2003–2007) compared with the pre-reform
observation span indicates that founding is more likely
post reform in all Table 4models (p<0.001). While
these direct effects are helpful, our H2test focuses
on the positive and significant coefficient of the post
reform ×elite alum variable (p<0.001) in both mod-
els 4-2 and 4-3. This result confirms that bankruptcy
reform motivates more elite-led firm founding. Finally,
we turn to direct comparisons in models 4-4 and 4-5,
where we regress elite and nonelite populations sep-
arately. The difference between the coefficients of the
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
14 Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS
Table 4. Piecewise Exponential Models on Transition to Founding
4-1 4-2 4-3 4-4 4-5
Dependent variable found
Piecewise regressions: Controls Reform Reform Elite-founded Nonelite-
Variables only interaction interaction firms founded firms
Observation span
1998–2002 0.000∗∗∗ 0.000∗∗∗ 0.000∗∗∗ 0.000∗∗∗ 0.000∗∗∗
(0.000) (0.000) (0.000) (0.000) (0.000)
2003–2007 0.000∗∗∗ 0.000∗∗∗ 0.002∗∗∗ 0.001∗∗∗ 0.000∗∗∗
(0.000) (0.000) (0.000) (0.000) (0.000)
GDP growth
1998–2000 0.082∗∗∗ 0.082∗∗∗ 0.107∗∗∗ 0.107∗∗∗ 0.082∗∗∗
(0.002) (0.002) (0.002) (0.002) (0.002)
2001–2003 0.847∗∗∗ 0.847∗∗∗ 0.575∗∗∗ 1.449 0.833∗∗∗
(0.027) (0.027) (0.027) (0.023) (0.287)
Local environment controls
colocated 1.011 1.013 1.007 0.994 1.010
(0.023) (0.023) (0.024) (0.118) (0.024)
comp. density 1.000 1.000 1.000 0.999 1.000
(0.000) (0.000) (0.000) (0.000) (0.000)
years_count 1.533∗∗∗ 1.533∗∗∗ 1.179∗∗∗ 1.740∗∗∗ 1.527∗∗∗
(0.016) (0.016) (0.004) (0.100) (0.016)
Reform variables
elite alum 0.881 0.879
(0.066) (0.176)
post reform ×elite alum 1.375∗∗∗ 1.374∗∗∗
(0.137) (0.138)
Observations 320,086 320,086 320,086 12,324 307,866
Firms 24,624 24,624 24,624 948 23,684
Wald χ2117,049∗∗∗ 116,769∗∗∗ 112,762∗∗∗ 4,563∗∗∗ 112,402∗∗∗
Founding year fixed effects Yes Yes No Yes Yes
Industry fixed effects Yes Yes Yes Yes Yes
Region fixed effects Yes Yes Yes Yes Yes
Notes. Model 4-3 is without founding year fixed effects. The difference between coefficients of observations spans in models 4-4 and 4-5 is
significant (p<0.001).
+p<0.10;p<0.05;∗∗p<0.01;∗∗∗ p<0.001 (two-sided tests).
observation spans in models 4-4) and 4-5 were com-
pared with t-tests and found to be significantly dif-
ferent (p<0.001) from each other. Thus, t-tests of the
piecewise analyses confirm that founding pre versus
post reform (model 4-4) increased significantly more
for elites than that for nonelites (model 4-5). In sum,
these piecewise hazard results support H2.
We add robustness to these piecewise results with a
panel logit analysis in Table 1 of the online appendix
for H2. The panel logit results add robustness in
support of H2. To further test for robustness to the
reform year, we ran placebo-reform panel logit tests
by using 2000, 2001, and 2002 as main effects, and we
interacted them with our elite variable. We confirmed
that we have insignificant, or even opposite, effects
for bankruptcy and founding (H1and H2) in these
“placebo-reform” years, as we would expect (Tables 3
and 4 in the online appendix). In addition, we find
robust results using a fixed-effects negative binomial
model. We recast our data set such that the number of
observations equals the number of prefectures multi-
plied by the number of years, and then we conducted
negative binomial regression analyses with prefectures
and years modeled as fixed effects. We ran these regres-
sions for both elite and nonelite founder populations,
and we examined the coefficient of the variable indi-
cating whether a firm was founded after reform. We
find results consistent with our earlier analyses (Table 5
in the online appendix). The bankruptcy reform vari-
able for nonelites has a nonsignificant coefficient com-
pared with positive and significant results for the elite
In Table 5, we assess whether elite-led ventures
founded after bankruptcy reform grew faster than those
founded prior to reform (H3). We use a difference-
in-differences analysis. Model 5-1 includes control
variables only and indicates that venture growth is
positively related to GDP growth, as expected. In
model 5-2, we establish that our control group, consult-
ing, is affected differently than the treated group—that
is, there is a negative and significant coefficient for the
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS 15
Table 5. Growth Effects
5-1 5-2 5-3 5-4 5-5 5-6
DV growth
GLM DID Elite founder Elite founder
(CEM matching—industry, Controls Elite founder interaction interaction Elite founder Nonelite
capital, region): only interaction (Primary control) (DV profit) only founder only
GDP growth 0.184∗∗∗ 0.065∗∗ 0.065∗∗ 0.109∗∗∗ 0.191 0.172∗∗∗
(0.044) (0.024) (0.023) (0.010) (0.535) (0.041)
comp. density 0.014 0.000 0.000 0.000 0.000 0.005
(0.013) (0.000) (0.000) (0.000) (0.000) (0.017)
firm age 0.249∗∗∗ 0.237∗∗∗ 0.145∗∗∗
(0.015) (0.015) (0.006)
consulting 0.133∗∗∗ 0.127 0.030 0.565∗∗ 0.054
(0.037) (0.119) (0.054) (0.246) (0.044)
primary 0.029
post reform founding 0.993∗∗∗ 0.936∗∗∗ 0.515∗∗∗ 7.820∗∗∗ 2.912∗∗∗
(0.226) (0.226) (0.114) (1.184) (0.507)
elite alum 0.001 0.004 0.109
(0.182) (0.182) (0.077)
post reform founding × −0.883∗∗∗ 0.102 1.769∗∗∗ 0.330∗∗∗
consulting (0.236) (0.113) (0.240) (0.079)
post reform founding × −0.428
primary industry (0.587)
post reform founding ×1.502∗∗∗ 1.398∗∗∗ 0.331∗∗∗
elite alum (0.322) (0.321) (0.115)
Constant 2.396∗∗∗ 4.737∗∗∗ 4.768∗∗∗ 1.667∗∗∗ 13.124∗∗∗ 2.266∗∗∗
(0.209) (0.389) (0.266) (0.172) (2.241) (0.261)
Observations 21,544 21,544 21,544 9,680 1,015 20,529
Founding year fixed effects Yes Yes Yes Yes Yes Yes
Industry fixed effects Yes Yes Yes Yes Yes Yes
Region fixed effects Yes Yes Yes Yes Yes Yes
Notes. The difference in the post reform founding coefficient in models 5-5 and 5-6 is significant (p<0.001). DV, dependent variable.
p<0.05;∗∗p<0.01;∗∗∗ p<0.001 (two-sided tests).
interaction variable, post reform ×consulting (p<0.001).
While the coefficient on post reform is positive and
significant (p<0.001), consulting firms are negatively
influenced by the reform such that their growth after
the reform is 11% lower than the omitted category.
We also used primary resources as our control group
(model 5-3) and obtained similar results for the inter-
action variable, post reform ×primary (p<0.001).
In models 5-3 and 5-4, we examine the primary
variable of interest, elite alum ×post reform founding,
to assess whether bankruptcy reform particularly ben-
efited the growth of elite-founded firms (H3). The
positive and significant interaction (p<0.001) con-
firms that, after reform, elite entrepreneurs founded
particularly high-growth firms. Indeed, elite-founded
firms launched after the reform grew more than 80%
faster than firms without an elite founder after reform.
In addition, overall venture growth increased after
bankruptcy reform, as shown by the variable post reform
founding (p<0.001).
In models 5-5 and 5-6, we regress elite and nonelite
populations separately to compare their post-reform
growth directly. We use t-tests to confirm that the coeffi-
cient of post reform founding for elite firms (model 5-5) is
significantly greater than that for nonelites (model 5-6)
(p<0.001). In summary, these findings are consistent
with H3and support that bankruptcy reform dispro-
portionately increased the growth of elite-led ventures
founded after the reform. To test for robustness, we
reran our general linear models of performance with
placebo-reform years. While we confirmed our pre-
vious positive and significant effect for the pseudo
reform years of 2002 and 2001, t-tests show the interac-
tion is significantly smaller than the 2003 effect (Table 5
in the online appendix).
Discussion and Conclusion
Our central insight is that a significant institu-
tional change that reduces failure barriers via bank-
ruptcy reform can spur the formation of high-growth
firms. Prior research linking institutional theory to
entrepreneurship recognizes the key role of legiti-
macy, uncertainty, and risk for new firm outcomes
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
16 Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS
(Tolbert et al. 2011). We add to institutional theory by
indicating how our focal institutional change provides
a potent lever for creating high-growth firms, not just
more bankruptcies or foundings (Hiatt et al. 2009, Sine
et al. 2005). Shifts in legitimacy, uncertainty, and per-
ceptions of risk appear to underlie these results. We
also pinpoint the disproportionate effects of this insti-
tutional change on elite individuals—i.e., those who
are most likely to start high-growth firms. Thus, by
lowering the legitimacy-based and financial costs of
bankruptcy, capable people are particularly encour-
aged to leave marginal firms, become entrepreneurs,
and launch high-growth firms. We turn now to specific
implications and contributions.
Implications at the Nexus of Institutional Theory
and Entrepreneurship
First, we add to institutional theory by highlighting
the effects of a significant institutional change, lowering
failure barriers via bankruptcy reform. Prior institutional
research shows that lowered entry barriers encourage
more individuals to establish firms (Aldrich and Fiol
1994, Sine and David 2003). Related work indicates that
lenient bankruptcy regulation also encourages firm
founding (Lee et al. 2007). While we confirm that this
institutional change does stimulate firm founding, we
further contribute to institutional theory by also show-
ing that this change leads to more high-growth firms.
Overall, we find that an institutional change that makes
failure easier via bankruptcy reform has unexpectedly
broad implications for new firms—i.e., increasing their
exit rate, founding rate, and growth.
Second, we contribute to institutional theory by pro-
viding insights into how institutional change affects
different types of individuals. Prior research indicates
that institutional changes that lower barriers to entry
(Sine et al. 2005) and barriers to growth (Eesley 2016)
differentially affect specific individuals. For example,
Eesley et al. (2016) showed that China’s Project 985
altered the university institutional environment such
that the rate of high-tech entrepreneurship increased.
We extend this line of work by showing that making
exit easier via bankruptcy reform (i.e., lowering barri-
ers to failure) also distinctively affects types of individ-
uals. Moreover, we add to institutional theory by intro-
ducing a focus on elites versus nonelites, a distinction
that is particularly relevant for entrepreneurial firms.
While there are other sources of heterogeneity, we indi-
cate that the elite versus nonelite distinction is central
for understanding the implications of how the institu-
tional context shapes legitimacy, uncertainty, and per-
ceptions of risk. We suggest that these, in turn, affect
entrepreneurial behaviors and new firm outcomes. In
particular, we show the disproportionate implications
of an institutional change that lowers failure barriers
on elite individuals. While elite individuals launch rel-
atively few firms, they are particularly influenced by
such change. By reducing the financial and legitimacy
costs (perceived and actual) of exit via bankruptcy
reform, elite individuals are especially likely to leave
marginal firms and start new firms with greater likeli-
hood of high growth.
Third, we contribute at the nexus of institutional
theory and entrepreneurship research by bringing a
critical emphasis on venture growth. Beginning with
neo-institutionalism’s foundations (Meyer and Rowan
1977), interest in performance, including growth, has
been limited within institutional theory research.
Rather, this work often explores topics such as con-
formity to institutional logics (Thornton 2002), institu-
tional sources of organizational practices (Eisenhardt
1988), and tactics of social movements (Sine and Lee
2009). Even institutional research on ventures often
centers on topics such as founding (Sine et al. 2005),
innovation (Pahnke et al. 2015), and strategies for legit-
imating nascent markets (Navis and Glynn 2010) rather
than performance per se. Yet examining performance
can yield unanticipated and powerful results. For
instance, Eesley et al. (2016) found that China’s efforts
to encourage more technology-based entrepreneurship
succeeded in that goal and yet yielded firms with lower
financial performance as a result of institutional incon-
sistency. Instead, work in institutional theory rarely
considers performance, and when it does, it is usually
in terms of survival (Hiatt and Sine 2014, Hiatt et al.
2009). While useful, survival is a coarse performance
measure that does not distinguish between, for exam-
ple, the “living dead” and truly successful ventures.
We contribute by bringing venture growth to center
stage. This puts institutional theory into closer align-
ment with the aims of many actual entrepreneurs and
policy makers in superior growth, not just survival. We
provide insight into how a key institutional change,
lowering failure barriers via bankruptcy reform, is a
significant lever for generating high-growth ventures,
and it ultimately addresses broader policy aims of eco-
nomic expansion and job creation.
More broadly, we contribute insight into how insti-
tutional change can affect the rich interrelationships
among failure, founding, and growth—relationships
that are essential to the dynamics of creative destruction
in flourishing economies. Past research uncovers some
of these interrelationships (e.g., Hoetker and Agarwal
2007). For example, Hiatt et al. (2009) found that fail-
ures among breweries led to more foundings in the
emerging soft drink industry. We add to institutional
theory by indicating that an institutional change—i.e.,
easing exit via bankruptcy reform—increases failures,
spurs foundings, and stimulates new firm growth,
particularly in elite-led firms. Thus, a deeper under-
standing of how bankruptcy—that which occurs at the
end of a firm’s life—influences the birth and growth
at the beginning of another firm’s life emerges from
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS 17
our work. Beyond contributions to institutional theory,
these insights also suggest implications for public pol-
icy directed at economic growth and job creation.
A final question is whether our results generalize
beyond Japan. Since our hypotheses do not rely on
restrictive assumptions, they seem likely to generalize.
That said, the size of the empirical effects may vary
i.e., there may be larger effects in countries such as
Sweden with punitive bankruptcy laws and smaller
effects in nations such as the United States, where
bankruptcy laws are already lenient. Our results also
suggest that institutional change may also alter the
organization and strategies of ventures. Thus, future
work may explore these issues further as well as
whether such institutional effects are particularly influ-
ential for certain industries, strategies, and founding
teams (Eesley et al. 2014). Overall, we expect that our
findings are likely to be especially relevant for nations
with punitive bankruptcy laws (e.g., Sweden, Thailand,
Argentina) (Lee et al. 2011) and failure-averse cultures
(e.g., Chile). Finally, Japan is likely a conservative test of
our hypotheses given its societal preference for estab-
lished firms and its traditional stance against failure
and risk taking (Vogel 2006).
We began by observing the importance of under-
standing how a significant institutional change, lower-
ing failure barriers via bankruptcy reform, affects new
firm growth. We argued that such a change shifts the
legitimacy, uncertainty, and perception of risk related
to bankruptcy, particularly among elite individuals.
By examining an instance of this institutional change,
we find that elite individuals are disproportionately
encouraged to leave marginal firms, begin new firms,
and lead them to high growth. Overall, if entrepreneur-
ship is part of the gale of creative destruction, then the
institutions that regulate the financial and legitimacy
costs of exit via bankruptcy play a pivotal role in deter-
mining the “strength” of that gale.
The authors gratefully acknowledge the financial support of
the Alan Miner Foundation, Michael Alfant of Fusion Sys-
tems, Inc., the Schulze Family Foundation, and Cisco Systems
of Japan, who made this research possible. The authors also
particularly want to thank the students and faculty at Stan-
ford Technology Ventures Program; the participants of the
Society, Technology, and Japanese Entrepreneurship confer-
ences at Stanford University; and the Kauffman Foundation
International Roundtable on Policy Studies, as well as partici-
pants at Stanford’s SCANCOR seminar. Their grateful thanks
go to Riitta Katila, Woody Powell, Stephen Barley, John Gun-
nar Carlsson, Martin Kenney, Glenn Hoetker, Steve Vogel, Jes-
per Sørenson, Ramana Nanda, Christina Ahmadjian, George
Foster, Anne Marie Knott, Renee Rottner, Nydia MacGregor,
Jennifer Woolley, Robert Kneller, Daniel Armanios, William
Miller, Keiko Ohara, Robert Sutton, and Ulrike Schaede for
their helpful comments and encouragement. Any errors in
this paper are the sole responsibility of the authors.
This work is in fond memory of Henry Rowen and
Masahiko Aoki.
1Bankruptcy reform is largely neutral for resource stakeholders, at
least in this study. For example, they retained control rights such as
the ability to remove poor executives. The key negative was loss of
access to executives’ personal assets. These assets were usually small
relative to the debt or investment. Consistent with this, we have no
evidence (e.g., interview data) that this effect was relevant to our
findings, although it may have lowered stakeholders’ willingness
to fund nonelites. If so, this further supports H2. We appreciate a
reviewer’s advice to consider the resource stakeholder lens.
2Also supporting generalizability, we have worked with the Chilean
and Swedish governments regarding the applicability of these
results to their national economies.
3From this initial sample, we eliminated 665 banks because Japanese
banks reorganized in the 1990s and appear in the database as new
firms when they are not and are often decades old. We also excluded
722 incorporated government entities, such as prisons.
4As a prominent venture attorney in Tokyo summarized, “We only
began to advise our clients that reorganization [i.e., bankruptcy]was
feasible from 2003 because that’s when the courts lost their authority
to block it.”
5Since we have a national-level reform, our difference-in-differences
approach is comparable to similar work on national reforms, which
uses variation in the extent that a subpopulation was influenced by
the reform—that is, groups vary on “dosage,” where some get more
of the treatment than others (Card 1992, Flores et al. 2012, Thompson
2009, Weber 2014). This approach depends on the assumption that
the treatment and control groups are changing similarly over time
(Abadie 2003) and that there is no unobserved factor that may be
driving selection into either the treatment or control group condi-
tion, which would violate the strict exogeneity assumptions of DID
approaches. We appreciate the advice of an anonymous reviewer to
use an alternative control group as a robustness check.
6While this is a relatively small percentage of founders, elites have
a disproportionate and important effect on Japanese business. For
example, several prominent Japanese firms that have recently under-
gone an initial public offering (such as Morpho, DeNA, Uniqlo, and
Globis) have elite university-educated founders. Similarly, 60% of
Japan’s largest companies have CEOs with elite university educa-
tions (Benes 2010).
Abadie A (2003) Semiparametric instrumental variable estimation of
treatment response models. J. Econometrics 113(2):231–263.
Adler PS, Kwon S-W (2002) Social capital: Prospects for a new con-
cept. Acad. Management Rev. 27(1):17–40.
Ahlstrom D (2010) Innovation and growth: How business con-
tributes to society. Acad. Management Perspect. 24(3):11–24.
Ahuja G, Katila R (2004) Where do resources come from? The role of
idiosyncratic situations. Strategic Management J. 25(8–9):887–907.
Aldrich H, Fiol CM (1994) Fools rush in? The institutional context of
industry creation. Acad. Management Rev. 19(4):645–670.
Aldrich H, Zimmer S (1986) Entrepreneurship through social
networks. Sexton D, Smilor R, eds. The Art and Science of
Entrepreneurship (Ballinger, Cambridge, MA), 3–23.
Altman EI (1993) Evaluating the Chapter 11 bankruptcy-
reorganization process. Columbia Bus. Law Rev. (1):1–25.
Aoki M (1994) The contingent governance of teams: Analysis of insti-
tutional complementarity. Internat. Econom. Rev. 35(3):657–676.
Arikawa Y, Miyajima H (2007) Understanding the M&A boom in
Japan: What drives Japanese M&A? RIETI Discussion Paper
07-E-042, Research Institute of Economy, Trade and Industry,
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
18 Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS
Armanios DE, Eesley CE, Li J, Eisenhardt KM (2016) How
entrepreneurs leverage institutional intermediaries in emerg-
ing economies to acquire public resources. Strategic Management
J., ePub ahead of print October 12,
Armour J, Cumming D (2008) Bankruptcy law and entrepreneurship.
Amer. Law Econom. Rev. 10(2):303–350.
Baum J, Bird B (2010) The successful intelligence of high-growth
entrepreneurs: Links to new venture growth. Organ. Sci. 21(2):
Beckman CM, Burton MD (2008) Founding the future: Path depen-
dence in the evolution of top management teams from founding
to IPO. Organ. Sci. 19(1):3–24.
Benes N ed. (2010) New growth strategies for Japan. Report, Ameri-
can Chamber of Commerce in Japan, Tokyo.
Berkowitz J, White MJ (2004) Bankruptcy and small firms’ access to
credit. RAND J. Econom. 35(1):69–84.
Bharath ST, Shumway T (2008) Forecasting default with the Merton
distance to default model. Rev. Financial Stud. 21(3):1339–1369.
Blackwell M, Iacus S, King G, Porro G (2009) cem: Coarsened exact
matching in Stata. Stata J. 9(4):524–546.
Blossfeld H, Rohwer G (2002) Techniques of Event History Modeling:
New Approaches to Causal Analysis (Lawrence Erlbaum Asso-
cioates, Mahwah, NJ).
Brookhart MA, Schneeweiss S, Rothman KJ, Glynn RJ, Avorn J,
Stürmer T (2006) Variable selection for propensity score models.
Amer. J. Epidemiol. 163(12):1149–1156.
Burton MD, Beckman C (2007) Leaving a legacy: Position imprints
and successor turnover in young firms. Amer. Sociol. Rev.
Campbell DT (1969) Reforms as experiments. Amer. Psych. 24(4):
Card D (1992) Using regional variation in wages to measure the
effects of the federal minimum wage. Indust. Labor Relations Rev.
Cerqueiro G, Hegde D, Penas MF, Seamans RC (2016) Debtor rights,
credit supply, and innovation. Management Sci., ePub ahead of
print August 11,
Chatterji AK (2009) Spawned with a silver spoon? Entrepreneurial
performance and innovation in the medical device industry.
Strategic Management J. 30(2):185–206.
Cooper AC, Gimeno-Gascon FJ, Woo CY (1994) Initial human and
financial capital as predictors of new venture performance.
J. Bus. Venturing 9(5):371–395.
DiMaggio PJ, Powell WW (1983) The iron cage revisited: Institutional
isomorphism and collective rationality in organizational fields.
Amer. Sociol. Rev. 48(2):147–160.
Dobrev SD, Barnett WP (2005) Organizational roles and transition to
entrepreneurship. Acad. Management J. 48(3):433–449.
Eesley C (2016) Institutional barriers to growth: Entrepreneur-
ship, human capital and institutional change. Organ. Sci. 27(5):
Eesley CE, Roberts EB (2012) Are you experienced or are you
talented? When does innate talent versus experience explain
entrepreneurial performance? Strategic Entrepreneurship J. 6(3):
Eesley CE, Hsu D, Roberts EB (2014) The contingent effects of top
management teams on venture performance: Aligning founding
team composition with innovation strategy and commercializa-
tion environment. Strategic Management J. 35(12):1798–1817.
Eesley C, Li JB, Yang D (2016) Does institutional change in uni-
versities influence high-tech entrepreneurship? Evidence from
China’s Project 985. Organ. Sci. 27(2):446–461.
Eisenhardt KM (1988) Agency- and institutional-theory explana-
tions: The case of retail sales compensation. Acad. Management J.
Eisenhardt K, Schoonhoven C (1990) Organizational growth: Linking
founding team, strategy, environment, and growth among U.S.
semiconductor ventures, 1978–1988. Admin. Sci. Quart. 35(3):
Fan W, White M (2003) Personal bankruptcy and the level of
entrepreneurial activity. J. Law Econom. 46(October):543–567.
Flores CA, Flores-Lagunes A, Gonzalez A, Neumann TC (2012) Esti-
mating the effects of length of exposure to instruction in a
training program: The case of job corps. Rev. Econom. Statist.
Folta T, O’Brien J (2003) Entry in the presence of dueling options.
Strategic Management J. 25(2):121–138.
Foote DH, ed. (2011) Law in Japan: A Turning Point (University of
Washington Press, Seattle).
Forman C, Ghose A, Goldfarb A (2009) Competition between local
and electronic markets: How the benefit of buying online
depends on where you live. Management Sci. 55(1):47–57.
Georgellis Y, Wall HJ (2002) Entrepreneurship and the policy envi-
ronment. Working Paper 2002-019B, Research Division, Federal
Reserve Bank of St. Louis, St. Louis.
Geroski PA, Mata J, Portugal P (2010) Founding conditions and the
survival of new firms. Strategic Management J. 31(5):510–529.
Gimeno J, Folta T, Cooper A, Woo C (1997) Survival of the fittest?
Entrepreneurial human capital and the persistence of underper-
forming firms. Admin. Sci. Quart. 42(4):750–783.
Granovetter M (1985) Economic action and social structure: The
problem of embeddedness. Amer. J. Sociol. 91(3):481–510.
Gulati R, Higgins MC (2003) Which ties matter when? The contin-
gent effects of interorganizational partnerships on IPO success.
Strategic Management J. 24(2):127–144.
Hallen B (2008) The causes and consequences of the initial network
positions of new organizations: From whom do entrepreneurs
receive investments? Admin. Sci. Quart. 53(4):685–718.
Hallen BL, Eisenhardt KM (2012) Catalyzing strategies and efficient
tie formation: How entrepreneurs obtain venture capital ties.
Acad. Management J. 55(1):35–70.
Helwege J, Packer F (2003) Determinants of the choice of bankruptcy
procedure in Japan. J. Financial Intermediation 12(1):96–120.
Hernán MA (2010) The hazards of hazard ratios. Epidemiology 21(1):
Hiatt SR, Sine WD, Tolbert PS (2009) From Pabst to Pepsi: The
deinstitutionalization of social practices and the creation of
entrepreneurial opportunities. Admin. Sci. Quart. 54(4):635–667.
Hiatt SR, Sine WD (2014) Clear and present danger: Planning and
new venture survival amid political and civil violence. Strategic
Management J. 35(5):773–785.
Hirose S (2009) Effects of the bankruptcy laws reform on banks: The
examination of recent Japanese experience. Public Policy Rev.
Hmieleski K, Baron R (2009) Entrepreneurs’ optimism and new ven-
ture performance: A social cognitive perspective. Acad. Manage-
ment J. 52(3):473–488.
Hoetker G (2007) The use of logit and probit models in strategic
management research: Critical issues. Strategic Management J.
Hoetker G, Agarwal R (2007) Death hurts, but it isn’t fatal: The pos-
texit diffusion of knowledge created by innovative companies.
Acad. Management J. 50(2):446–467.
Hogarth RM, Karelaia N (2012) Entrepreneurial success and failure:
Confidence and fallible judgment. Organ. Sci. 23(6):1733–1747.
Hsu DH (2007) Experienced entrepreneurial founders, organiza-
tional capital, and venture capital funding. Res. Policy 36(5):
Hsu D, Roberts E, Eesley C (2007) Entrepreneurial ventures from
technology-based universities: Evidence from MIT. Res. Policy
Iacus SM, King G, Porro G (2012) Causal inference without balance
checking: Coarsened exact matching. Political Anal. 20(1):1–24.
Imai Y, Kawagoe M (2000) Business start-ups in Japan: Problems and
policies. Oxford Rev. Econom. Policy 16(2):114–123.
Kahneman D, Tversky A (1979) Prospect theory: An analysis of deci-
sion under risk. Econometrica 47(2):263–292.
Knott AM, Posen HE (2005) Is failure good? Strategic Management J.
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS 19
La Porta R, López-de-Silanes F (2001) Creditor protection and
bankruptcy law reform. Resolution of Financial Distress: An Inter-
national Perspective on the Design of Bankruptcy Laws (World Bank,
Washington, DC), 65–90.
Lee S-H, Peng MW, Barney JB (2007) Bankruptcy law and entrepre-
neurship development: A real options perspective. Acad. Man-
agement Rev. 32(1):257–272.
Lee S-H, Yamakawa Y, Peng MW, Barney JB (2011) How do bank-
ruptcy laws affect entrepreneurship development around the
world? J. Bus. Venturing 26(5):505–520.
Liang K-Y, Zeger SL, Qaqish B (1992) Multivariate regression analy-
ses for categorical data. J. Roy. Statist. Soc. Ser. B (Methodological)
Lounsbury M, Glynn MA (2001) Cultural entrepreneurship: Stories,
legitimacy, and the acquisition of resources. Strategic Manage-
ment J. 22(6–7):545–564.
Lowe RA, Ziedonis AA (2006) Overoptimism and the performance
of entrepreneurial firms. Management Sci. 52(2):173–186.
Maurer I, Ebers M (2006) Dynamics of social capital and their per-
formance implications: Lessons from biotechnology start-ups.
Admin. Sci. Quart. 51(2):262–292.
Meek W, Pacheco D, York J (2010) The impact of social norms on
entrepreneurial action: Evidenece from the environmental en-
trepreneurship context. J. Bus. Venturing 25(5):493–509.
Meyer JW, Rowan B (1977) Institutionalized organizations: Formal
structure as myth and ceremony. Amer. J. Sociol. 83(2):340–363.
Miller D, Shamsie J (1996) The resource-based view of the firm in two
environments: The Hollywood film studios from 1936 to 1965.
Acad. Management J. 39(3):519–543.
Miyamoto T, Rexha N (2004) Determinants of three facets of customer
trust: A marketing model of Japanese buyer–supplier relation-
ship. J. Bus. Res. 57(3):312–319.
Mizruchi MS, Fein LC (1999) The social construction of organiza-
tional knowledge: A study of the uses of coercive, mimetic, and
normative isomorphism. Admin. Sci. Quart. 44(4):653–683.
Mori M (2006) Japan finalizes insolvency law reform. Internat. Finance
Law Rev. (January 1),
Navis C, Glynn MA (2010) How new market categories emerge: Tem-
poral dynamics of legitimacy, identity, and entrepreneurship in
satellite radio, 1990–2005. Admin. Sci. Quart. 55(3):439–471.
Ohara K (2015) Personal communication by email and phone with
Eberhart R. Kamiyacho, November 19. International Law Office,
Pacheco DF, York JG, Hargrave TJ (2014) The coevolution of indus-
tries, social movements, and institutions: Wind power in the
United States. Organ. Sci. 25(6):1609–1632.
Pahnke EC, Riitta K, Eisenhardt KM (2015) Who takes you to the
dance? How funding partners influence innovative activity in
young firms. Admin. Sci. Quart. 60(4):596–633.
Peng M, Yamakawa Y, Lee S (2008) Entrepreneurship and the bar-
rier to exit: How does an entrepreneur-friendly bankruptcy law
affect entrepreneurship development at a societal level. Work-
ing Paper, No. 08slyymp, U.S. Small Business Administration,
Washington, DC.
Peng M, Yamakawa Y, Lee S (2009) Bankruptcy laws and entrepre-
neur friendliness. Entrepreneurship Theory Practice 34(3):517–530.
Powell WW, Koput KW, Smith-Doerr L (1996) Interorganizational
collaboration and the locus of innovation: Networks of learning
in biotechnology. Admin. Sci. Quart. 41(1):116–145.
Schaede U (2008) Choose and Focus: Japanese Business Strategies for the
21st Century (Cornell University Press, Ithaca, NY).
Scott WR (2008) Institutions and Organizations, 3rd ed. (Sage, Thou-
sand Oaks, CA).
Simon M, Houghton S, Aquino K (1999) Cognitive bias, risk percep-
tion, and venture formation: How individuals decide to form
companies. J. Bus. Venturing 15(2):113–134.
Sine WD, David R (2003) Environmental jolts, institutional change,
and the creation of entrepreneurial opportunity in the U.S. elec-
tric power industry. Res. Policy 32(2):185–207.
Sine WD, Lee B (2009) Tilting at windmills? The environmental
movement and the emergence of the U.S. wind energy sector.
Admin. Sci. Quart. 54(1):123–155.
Sine WD, Haveman HA, Tolbert PS (2005) Risky business? Entre-
preneurship in the new independent-power sector. Admin. Sci.
Quart. 50(2):200–232.
Singleton W, Globerman S (2002) The changing nature of financial
disclosure in Japan. Internat. J. Accounting 37(1):95–111.
Statistics Bureau of Japan (2011) Statistical abstract of Japan.
Accessed February 24, 2016,
Stuart T, Sorenson O (2003) The geography of opportunity: Spatial
heterogeneity in founding rates and the performance of biotech-
nology firms. Res. Policy 32(2):229–253.
Stuart T, Hoang H, Hybels R (1999) Interorganizational endorse-
ments and the performance of entrepreneurial ventures. Admin.
Sci. Quart. 44(2):315–349.
Suchman M (1995) Managing legitimacy: Strategic and institutional
approaches. Acad. Management Rev. 20(3):571–610.
Sutton R, Callahan A (1987) The stigma of bankruptcy: Spoiled
organizational image and its management. Acad. Management J.
Thaler RH, Tversky A, Kahneman D, Schwartz A (1997) The effect of
myopia and loss aversion on risk taking: An experimental test.
Quart. J. Econom. 112(2):647–661.
Thompson JP (2009) Using local labor market data to re-examine the
employment effects of the minimum wage. Indust. Labor Rela-
tions Rev. 62(3):343–366.
Thornhill S, Amit R (2003) Learning about failure: Bankruptcy, firm
age, and the resource-based view. Organ. Sci. 14(5):497–509.
Thornton PH (2002) The rise of the corporation in a craft industry:
Conflict and conformity in institutional logics. Acad. Manage-
ment J. 45(1):81–101.
Times Higher Education (2011) Times Higher Education World University
Rankings 2010–2011 (Thomson Reuters, London).
Tolbert PS, David RJ, Sine WD (2011) Studying choice and change:
The intersection of institutional theory and entrepreneurship
research. Organ. Sci. 22(5):1332–1344.
Tuma NB, Hannan MT, Groeneveld LP (1979) Dynamic analysis of
event histories. Amer. J. Sociol. 84(4):820–854.
Uzzi B (1999) Embeddedness in the making of financial capital: How
social relations and networks benefit firms seeking financing.
Amer. Sociol. Rev. 64(4):481–505.
Vogel SK (2006) Japan Remodeled: How Government and Industry
Are Reforming Japanese Capitalism (Cornell University Press,
Ithaca, NY).
Weber CE (2014) Toward obtaining a consistent estimate of the elas-
ticity of taxable income using difference-in-differences. J. Public
Econom. 117(September):90–103.
Wiesenfeld BM, Wurthmann KA, Hambrick DC (2008) The stigma-
tization and devaluation of elites associated with corporate fail-
ures: A process model. Acad. Management Rev. 33(1):231–251.
Wooldridge JM (2010) Econometric Analysis of Cross Section and Panel
Data, 2nd ed. (MIT Press, Cambridge, MA).
Xu P (2007) Corporate governance in financial distress: The new role
of bankruptcy. Aoki M, Jackson G, Miyajima H, eds. Corporate
Governance in Japan: Institutional Change and Organizational Diver-
sity (Oxford University Press, Oxford, UK), 179–204.
Yanagida K (2014) Personal insolvency law in Japan. Report,
Yanagida & Partners, Tokyo.
York JG, Lenox MJ (2014) Exploring the sociocultural determinants
of de novo versus de alio entry in emerging industries. Strategic
Management J. 35(13):1930–1951.
Robert N. Eberhart is an assistant professor of organiza-
tional analysis and entrepreneurship at the Leavey School
of Business at Santa Clara University. He received his Ph.D.
from Stanford University. His research interests include the
influence of institutional change on entrepreneurship and
the effect of those changes on the types of ventures that are
founded, their performance, and the effect of entrepreneur-
ship on society.
Eberhart, Eesley, and Eisenhardt: Institutional Change, Entrepreneurial Risk, New Firm Growth
20 Organization Science, Articles in Advance, pp. 1–20, ©2017 INFORMS
Charles E. Eesley is an assistant professor and Morgen-
thaler Faculty Fellow in the Department of Management Sci-
ence and Engineering at Stanford University. His research
focuses on the role of the institutional and university envi-
ronment in high-growth, technology entrepreneurship. His
research contexts include China, Japan, Chile, and the United
States. He received his Ph.D. in management from the MIT
Sloan School of Management.
Kathleen M. Eisenhardt is the S.W. Ascherman M.D. Pro-
fessor at Stanford University and codirector of the Stan-
ford Technology Ventures Program. Her recent book (with
Don Sull) is Simple Rules: How to Thrive in a Complex
World. Her research emphasizes the strategy and organiza-
tion of technology-based companies in high-velocity markets
including current work on ecosystems, venture boards, and
strategy in two-sided markets.
... Response heterogeneity adds an extra layer of complexity in deciphering the effects of policy interventions since individuals exposed to the same stimuli can decide on different courses of action (e.g. Eberhart, Eesley & Eisenhardt, 2017;Greene, Han & Marlow, 2011). In addition, Bjørnskov & Foss (2016, 301) find that many "studies assume that the responses to institutional and policy differences are approximately homogeneous across different types of industries, businesses, and countries and institutional settings". ...
... When an individual takes action to engage in entrepreneurship, their ability to access necessary resources is affected by their existing financial position (Armour & Cumming, 2008;Bassetto et al., 2015;Estrin et al., 2017) and networks (Sarasvathy, 2001). Individuals predicted to be successful and engage in high quality entrepreneurship, but who are unable to meet expectations are more likely to be stigmatised, which can create extra external barriers in further attempts to engage in entrepreneurship (Eberhart et al., 2017). ...
... A commitment to supporting entrepreneurs with social protection can cushion against some of the negative effects of failure and support an entrepreneur's transition to new ventures. For instance, in developed countries, corporate and individual bankruptcy laws that allow quick recovery from failure are associated with higher rates of entrepreneurship because the institutional context encourages individuals to take on the risks of new venture creation not only initially but also repeatedly over the longer term (Armour & Cumming, 2008;Eberhart et al., 2017). Social protection signals a willingness to invest in the human capital of all individuals in the labour market, including entrepreneurs. ...
Full-text available
Entrepreneurship has the potential to drive economic development and social advancement. The European Commission implemented entrepreneurship policy as a pragmatic response to its economic and social challenges, especially after the 2008 financial crisis. Institutional changes to promote entrepreneurship and enable individuals to directly contribute to economic growth, job creation and society were introduced to create an entrepreneurial Europe. This study undertakes a systematic review to examine the implications of entrepreneurship policy within Europe. Examining and understanding the impacts of entrepreneurship policy and institutional changes are particularly relevant because of the billions of Euros invested and the impacts on the working lives of European citizens. By examining a broad range of existing literature, the study finds that the entrepreneurial Europe envisioned by policymakers has not been fully realised. Instead, entrepreneurship activity has skewed towards poor quality, necessity entrepreneurship. The European institutional context has also shifted away from the social model on which it was founded, increasing the exposure of European workers to social risks. To promote sustainable growth, wellbeing and well-functioning labour markets, researchers and policymakers are reconsidering the role of social protection. Social protection also has the potential to promote quality entrepreneurship. Based on the review of literature, seven testable propositions about how social protection can promote quality entrepreneurship have been developed for future empirical testing. This study advances knowledge in entrepreneurship research and contributes to debates in policymaking and practice. It also provides a sound basis for subsequent empirical research.
... Indeed, few are the studies that attempted to examine how a specific policy can influence different types of individuals to enter the entrepreneurship arena, and none has a specific focus on innovative entrepreneurship. Unlike the preceding stream of literature addressing barriers to entry (Branstetter et al., 2014;Sine & David, 2003;Sine & Lee, 2009), which is inherently limited to observing the founding rate as the outcome, these more recent studies emphasize the impact of lowering various barriers to entrepreneurship (Eberhart et al., 2017;Eesley, 2016). We build on this idea that institutional changes that reduce different types of barriers to entrepreneurship may impact not only entrepreneurial quantity (Branstetter et al., 2014) but also entrepreneurial quality, i.e., the skills and competencies of the individuals who become founders. ...
... In this stream, there is a conspicuous body of evidence that previous experience of founders (e.g., Ardichvili et al., 2003;Åstebro & Thompson, 2011;Gruber et al., 2012;Shane, 2000;Shepherd & DeTienne, 2005;Ucbasaran et al., 2008Ucbasaran et al., , 2009), their demographics (e.g., ethnicity, gender, income level), and other personal (e.g., ambition, risk propensity, motivations) traits play significant roles on all the above mentioned dimensions related to entrepreneurship (DeTienne & Chandler, 2007;Douglas & Shepherd, 2002;Lévesque & Minniti, 2007;Van Praag & Cramer, 2001;Waldinger et al., 1990). Nonetheless, apart some notable exceptions (e.g., Eberhart et al., 2017;Eesley, 2016), studies in this realm have typically not considered whether (and eventually how) changes in the institutional environment, and in particular the issue of a new specific formal law, may impact the relationship between the characteristics of individuals and their decision to become entrepreneurs in the new institutional landscape. ...
... Previous studies highlight entry barriers as a pivotal mechanism influencing founding rates (Klapper et al., 2006;Sine & Lee, 2009). Eesley (2016) and Eberhart et al. (2017) do introduce and test alternative mechanisms (growth and exit barriers, respectively). However, they do not confront these mechanisms within the same policy reform. ...
Full-text available
Prior research shows that entrepreneurship enhances economic development. However, it is becoming increasingly evident that it is not the number of new startups that matter but rather their quality. This study investigates the effect of a comprehensive industrial policy intervention targeting innovative startups, i.e., the Italian Startup Act, on the composition of innovative entrepreneurs in terms of their human capital endowment. By decomposing the impact of lowering entry and growth barriers and by comparing the “before” and the “after” of the reform, we explore if the industrial policy has modified the composition of innovative entrepreneurs in terms of their human capital characteristics. The findings indicate that the reform, and in particular lowering growth barriers, was particularly able to push individuals with a relatively higher level of industry-specific, managerial, and entrepreneurial experience towards the creation of a new innovative venture. Overall, we show that a policy reform that decreases barriers to innovative entrepreneurship may attract entrepreneurs endowed with greater specific human capital than what occurred before the reform.
... Various scholars have linked control of corruption with firm growth aspirations (e.g., Estrin et al., 2013), suggesting a negative correlation between low-quality institutions, indicating extant corruption, and high-growth entrepreneurship (Bowen and Clercq, 2008). Eberhart et al. (2017) found that, as institutional quality increased in Japan (i.e., the shift to lenient bankruptcy reform), graduates of top Japanese management universities tend to establish high-growth firms. ...
... Further, this study focused on firm growth in the 2000s. Indeed, institutions change as economies develop (Baumol, 1990;Eberhart et al., 2017). Firms respond to such changes by reallocating resources and modifying strategies (Parnell, 2008), especially when they encounter a crisis (Coleman, 2004;Jakubanecs et al., 2018;Lalonde, 2007;Robert and Lajtha, 2002). ...
Full-text available
The growth of small and medium-sized enterprises (SMEs) is a multilevel phenomenon. However, scant research has examined its antecedents at both the country and firm levels. This study examines the effects of both countrylevel institutional quality and firm characteristics (i.e., firm age, state ownership, and management experience) on firm growth. The sample includes nearly 1,800 SMEs from 12 countries in the World Bank’s Management, Organization, and Innovation (MOI) database. Government effectiveness, political stability, and the absence of violence were positively associated with firm growth. Unexpectedly, control of corruption was negatively associated with firm growth. Younger and non-state-owned enterprises grew faster than older state-owned firms. Management experience did not influence firm growth. This study underscores the need for governments to provide political, economic, and social stability as a precursor to SME growth. Keywords: SMEs; growth; institutions; firm age; management experience; state ownership
... However, whether scope is conceived of as range only or also as intensity, the total impact and entrepreneurially enabling impact are often unlikely to be closely aligned. For instance, one EE may be a regulatory change intended to facilitate entrepreneurship and has that effect without affecting much else such as when Japan drastically reduced the stringency of bankruptcy regulations (Eberhart et al., 2017); another may be an environmental change that has effects across all industries but entrepreneurially enabling effects of meaningful magnitude only to ventures in some industries such as novel accounting regulations that affect all firms in a country but only act as an enabler for ventures that help other organizations to cope with the new regulations (cf. Kimjeon & Davidsson, 2021: 9). ...
"External enabler" (EE) denotes nontrivial changes to the business environment-such as new technology, regulatory change, demographic and sociocultural trends, macroeconomic swings, and changes to the natural environment-that enable entrepreneurial pursuits. The EE framework was developed to increase knowledge accumulation in entrepreneurship and strategy research regarding the influence of environmental factors on entrepreneurial endeavors. The framework provides detailed structure and carefully defined terminology to describe, analyze, and explain the influence of changes in the business environment on entrepreneurial pursuits. EE characteristics specify the environmental changes' range of impact in terms of spatial, sectoral, sociocultural, and temporal scope as well as the degree of suddenness and predictability of their onset. EE mechanisms specify the types of benefits individual ventures may derive from EEs. Among others, these include cost saving, resource provision, making possible new or improved products/services, and demand expansion. EE roles situate these (anticipated) mechanisms in entrepreneurial processes as triggering and/or shaping and/or outcome-enhancing. EE's influence is conceived of as mediated by entrepreneurial agency that-in addition to agent characteristics-is contingent on the opacity (difficulty to identify) and agency-intensity (difficulty to exploit) of EE mechanisms, with the ensuing enablement being variously fortuitous or resulting from strategic deliberation.
... Analysis of the impact of state regulatory institutions on entrepreneurship is perhaps one of the more developed areas of inquiry within the institutional approach to entrepreneurship. Scholars have analyzed how the removal of regulatory barriers to failure and growth can increase both the number and the quality of nascent ventures (Eesley 2016, Eberhart et al. 2017. Alternatively, scholars have also highlighted how state regulations can spur entrepreneurial activity via the creation of new entrepreneurial opportunities (Eckhardt and Shane 2003, David et al. 2013, Georgallis et al. 2019) and via the conferral of legitimacy (Assenova andSorenson 2017, Armanios andEesley 2021). ...
The institutional context, which includes the normative, regulative, and cognitive dimensions of social life within the various constitutive spheres of society, has a strong influence on entrepreneurial processes and outcomes. Institutions shape who becomes an entrepreneur, opportunity creation, identification, and evaluation, as well as how entrepreneurs attempt to start new firms. We introduce a novel framework that unifies the two dominant perspectives in sociological neoinstitutionalism, the institutional logics and the institutional pillars typologies, and apply this unified framework to examine the existing research at the nexus of entrepreneurship and institutional theory while outlining a set of entrepreneurial phenomena to which the framework can be applied. We analyzed the citation pattern of all 77 articles published since 1999 in top management journals (Academy of Management Journal, Academy of Management Review, Administrative Science Quarterly, Organization Science, and Strategic Management Journal) that used institutional theory to examine entrepreneurial phenomena, and we demonstrate how the unified framework effectively organizes past research while also pointing to new and important areas for development.
... Too stringent bankruptcy legislation therefore curtails knowledge flows (Holbrook et al. 2000;Armour and Cumming 2006). By contrast, research suggests that when lenient, such legislation leads to more business creation (Fan and White 2003;Peng et al. 2009) and may even improve venture quality (Eberhart et al. 2017). Entrepreneurship will always be about risk-taking, but ...
This study examines how social safety nets providing paid family leave (PFL) benefits to employees influence subsequent business performance for entrepreneurial ventures. A multilevel framework guided by qualitative interviews proposes two competing mechanisms (pre‐hiring recruitment gains via prospective employees and post‐hiring operation losses via incumbent employees) and a firm‐level contingency (venture innovation type). Leveraging the 2009‐implemented New Jersey PFL program in a difference‐in‐differences design, I show that employee access to state‐provided PFL benefits adversely affects the profitability for noninnovative new ventures but increases profitability for innovative ventures. Exploration of treatment timing and intermediate venture outcomes supports a dominating pre‐hiring mechanism for innovative ventures (in which the ventures become more attractive to joiners) but a dominating post‐hiring mechanism for noninnovative ventures (due to employee leave). What are the business implications of employee access to paid family leave (PFL) benefits for nascent firms? Leveraging evidence from a U.S. state PFL program, the study shows that employee access to state PFL benefits hurts the profitability of noninnovative ventures but increases the profitability of innovative ventures. Results suggest that while noninnovative ventures experience operation losses as incumbent employees increase family leave use or quit their jobs, innovative ventures reap recruitment gains by attracting more and better prospective employees. For founders and managers, the provision of employee benefits is more central to firms’ ability to create superior human capital pools than often thought and should be aligned with organizational incentive design and HR systems to prepare for productivity shocks related to employee leave use. This article is protected by copyright. All rights reserved.
Full-text available
Among contemporary economists, Mariana Mazzucato stands out for her emphasis on the importance of innovation to solving pressing challenges and achieve a greater quality of life. However, the type of mission-oriented innovation policies she promotes usually rely on an overly mechanical view of innovation and economic growth. We employ an ecosystem perspective to demonstrate that innovative entrepreneurship takes place in a collaborative innovation bloc consisting of a myriad of nodes. Entrepreneurs, inventors, early- and later-stage financiers, key personnel, and customers are all actors whose skills and abilities are necessary to realize an entrepreneurial project. When mission-oriented policies play a large role in an industry’s production or financing, connections between actors in the innovation bloc risk being severed, severely curtailing the scope for actors to play their requisite roles. Thus, there is a risk that such policies do more harm than good for innovation and economic growth.
This study explores how bribery affects firm growth by focusing on the asymmetric dependence of firms on government resources and services. We conceptualize bribery as relationships through which bribery requests require firms to frequently interact with rent-seeking government officials. Through bribery relationships, such officials extort firms beyond the exchange of bribe money for preferential treatment, depriving acquiescing firms of time and effort and thereby imposing hidden costs that could be otherwise used for firm growth. We find that bribery relationships damage firm growth. Firm status such as introducing new products or not affects how rent-seeking government officials check on firms and calculate their extortion schemes in bribery relationships; bribery relationships damage firm growth more significantly for firms without new products. The damage of bribery relationships to firm growth is also contingent on institutional environments. Under pervasive corruption, firms in bribery relationships may increase their acquiescence to the extortion of rent-seeking government officials. In countries with high-quality governance, however, firms can depend on sound regulations and rules of law and government officials experience high moral costs of corruption. Thus, the negative effect of the bribery relationship on firm growth will be strengthened under pervasive corruption and weakened under high-quality governance. Using Business Environment and Enterprise Performance Survey and World Governance Indicators data for 28 Eastern European countries from 2002–2014, we demonstrate the multifaceted features of the bribery relationship and its interaction with country-level institutional environments.
Full-text available
Milton Friedman once argued that profits are the chief purpose of business. Profits do matter, but today we know more about how business contributes to society. Good firms bring innovation to the marketplace, which facilitates their growth. Innovative, growing firms generate economic growth and employment, which, in turn, greatly improves people’s lives. In this paper I argue that the main goal of business is to develop new and innovative goods and services that generate economic growth while delivering important benefits to society. Steady economic growth generated through innovation plays a major role in producing increases in per capita income. Small changes in economic growth can yield very large differences in income over time, making firm growth particularly salient to societies. In addition to providing growth, innovative firms can supply important goods and services to consumers, particularly those at the base of the pyramid. Through innovation and growth firms can do untold good for society.
Full-text available
Research Summary: Governments in emerging economies often use institutional intermediaries to promote entrepreneurship, and bridge the void between ventures and public funding. While prior literature describes what institutional intermediaries do, it leaves open how intermediaries support different types of entrepreneurs. By comparing science park and non-science park firms in Beijing and across China, we distinguish which entrepreneurs benefit from certification v. capability-building through the introduction of two new constructs: skill adequacy and context relevance. Broadly, our study adds insights at the nexus of emerging economies and entrepreneurship research, and to the tie formation and institutional intermediaries literatures. Managerial Summary: A key dilemma facing entrepreneurs is how to finance their ventures. While entrepreneurs in developed economies can seek VC or angel investment, entrepreneurs in emerging economies often need to pursue potential government funding opportunities. Our study highlights three strategies for acquiring government funding. Well-connected entrepreneurs can leverage their political ties to acquire such funding. Less-connected entrepreneurs can leverage science parks that in emerging markets are designed to help governments to identify promising ventures. For returnees whose ample experience abroad may not fit with local ways of doing business, gaining science park admission can certify quality and so ease the path to government funding. For technically skilled local entrepreneurs who lack business skills, science parks can help build such skills which then ease the path to government funding.
Prior research often focuses on how many entrepreneurial firms are created, rather than on institutions that encourage specific types of firms or entrepreneurs. This paper identifies institutional changes that reduce barriers to growth as an important factor influencing the propensity of individuals to start a business. The findings suggest that the impact of lower barriers to growth is shaped by the extent of the reduction in barriers to growth and the level of human capital of the individual. Only a large reduction in barriers to growth has a stronger impact in increasing the likelihood of founding at higher levels of human capital. I capitalize on two reforms lowering barriers to growth as natural experiments. One reform, in 1988, only slightly lowered barriers to growth. The second reform, in 1999, more strongly lowered barriers to growth with an amendment to the Chinese constitution reversing regulations that favored firms with foreign investors. This made it easier for domestic entrepreneurs to compete. I collected data through a survey of 2,966 alumni who graduated from a top Chinese university. Results show that reducing the institutional barriers to growth differently affects college-educated individuals with different levels of human capital.
Firms’ innovative activities can be sensitive to public policies that a ect the availability of capital. In this paper, we investigate the e ects of regional and temporal variation in U.S. personal bankruptcy laws on firms’ innovative activities. We find that bankruptcy laws that provide stronger debtor protection decrease the number of patents produced by small firms. Stronger debtor protection also decreases the average quality, and variance in quality, of firms’ patents. We find evidence that the negative e ect of stronger debtor protection on experimentation and innovation may be due to the decreased availability of external financing in response to stronger debtor rights, an e ect amplified in industries with a high dependence on external financing. Hence, while it is typically assumed that stronger debtor protection encourages innovation by reducing the cost of failure for innovators, we show that it can instead dampen innovative activities by tightening the availability of external financing to innovative firms.