程序代写代做代考 Excel database finance This article appeared in a journal published by Elsevier. The attached

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:

http://www.elsevier.com/authorsrights

http://www.elsevier.com/authorsrights

Author’s personal copy

Banking deregulation and innovation$

Sudheer Chava a, Alexander Oettl a,n, Ajay Subramanian b, Krishnamurthy
V. Subramanian c

a Scheller College of Business, Georgia Institute of Technology, Atlanta, GA 30308, USA
b Robinson College of Business, Georgia State University, Atlanta, GA 30303, USA
c Indian School of Business, Gachibowli, Hyderabad 500032, India

a r t i c l e i n f o

Article history:
Received 1 November 2012
Received in revised form
23 January 2013
Accepted 22 February 2013
Available online 6 April 2013

JEL classifications:
G28
L43
O31
O43
K23

Keywords:
Banking
Innovation
Growth
Young firms
Private firms

a b s t r a c t

We document empirical support for a key micro-level channel—innovation by young,
private firms—through which financial sector deregulation affects economic growth. We
find that intrastate banking deregulation, which increased the local market power of
banks, decreased the level and risk of innovation by young, private firms. In contrast,
interstate banking deregulation, which decreased the local market power of banks,
increased the level and risk of innovation by young, private firms. These contrasting
effects on innovation also translated into contrasting effects on economic growth. Our
study suggests that the nature of financial sector deregulation crucially affects its potential
benefits to the real economy.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

Although the financial economics literature provides
robust evidence supporting the Schumpeterian view that

financial development fosters economic growth,1 evidence
on the micro-level channels through which this relation
manifests remains relatively sparse. In this paper, we build
on the premise of endogenous growth theory (e.g., Aghion
and Howitt, 1992) to show that financial sector deregula-
tion impacts firm-level innovation and, thereby, economic
growth. Interestingly, however, we find that the effects of
financial deregulation crucially depend on the nature of
deregulation and how it affects competition in credit
markets. We focus on the impact of intrastate and inter-
state banking deregulations, and show that they had
contrasting effects on innovation by young, private firms

Contents lists available at SciVerse ScienceDirect

journal homepage: www.elsevier.com/locate/jfec

Journal of Financial Economics

0304-405X/$ – see front matter & 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jfineco.2013.03.015

☆ Ajay Subramanian is very grateful for the hospitality of the Indian
School of Business, Hyderabad, where a part of this research was carried
out. Krishnamurthy Subramanian would like to thank the Center for
Innovation, Leadership and Change at the Indian School of Business for
research support. We thank the anonymous referee, Heitor Almeida, Bo
Becker, Tom Chemmanur, Brent Goldfarb, Radhakrishna Gopalan, Bill
Kerr, Elena Loutskina, Debarshy Nandy, Ramana Nanda, Vikram Nanda,
Alminas Zaldokas, and seminar participants at the 12th Annual Round-
table for Engineering Entrepreneurship Research (REER) conference, the
University of Florida (Gainesville), the Georgia Institute of Technology,
and the Corporate Finance conference at Washington University in St.
Louis for valuable comments and suggestions. Hyun Jung provided
excellent research assistance.

n Corresponding author.
E-mail address: alexander.oettl@scheller.gatech.edu (A. Oettl).

1 King and Levine (1993a,b), Demirgüç-Kunt and Maksimovic (1998),
Rajan and Zingales (1998), and Beck and Levine (2004) provide cross-
country evidence, while Jayaratne and Strahan (1996, 1998), Cetorelli and
Strahan (2006), and Beck, Levine, and Levkov (2010) provide evidence
within the U.S.

Journal of Financial Economics 109 (2013) 759–774

Author’s personal copy

that led to corresponding effects on economic growth. Our
results highlight that innovation by young, private firms is
a key channel through which finance affects economic
growth.

We focus on the impact of banking deregulation on the
innovative activity of young, private firms for two impor-
tant reasons. First, banks play an important role in finan-
cing young, private firms (e.g., Berger, 2010; Nanda and
Nicholas, 2012). Even with the growth of venture capital in
recent years, empirical and anecdotal evidence suggests
that banks continue to play an important role in the
financing of innovation. Robb and Robinson (forth-
coming) find that firms in the Kauffman Firm Survey—
primarily young, private firms—rely extensively on bank
financing. Berger and Udell (1998) find that commercial
bank loans provide 19% of all financing for small busi-
nesses in the National Survey of Small Business Finances
sample.2 Cetorelli and Strahan (2006) document that 70%
of the surveyed firms in the 1998 Survey of Small Business
Finance (SSBF) use banks for credit.3 Second, innovation is
a fountainhead of economic growth and a significant body
of empirical evidence shows that young, private firms are
key drivers of path-breaking innovation (e.g., Akcigit and
Kerr, 2011; Acs and Audretsch, 1987, 1988, 1993; Zucker,
Darby and Brewer, 1998; Kortum and Lerner, 2000; Samila
and Sorenson, 2010; Darby and Zucker, 2003).

We disentangle the effects of two disparate forms of
banking deregulation in the U.S.: intrastate deregulation
that allowed banks to expand within states, and interstate
deregulation that allowed banks to expand beyond state
boundaries. Using the results of previous studies, we
hypothesize that intrastate and interstate deregulations
had contrasting effects on the local market power of banks
and, thereby, innovation by young, private firms. Consis-
tent with our hypotheses, we show that while intrastate
deregulation decreased the level and risk of innovation by
young, private firms, interstate deregulation increased
both. Further, the contrasting effects of intrastate and
interstate deregulation on innovation resulted in corre-
sponding effects on economic growth. Our findings, there-
fore, highlight that the nature of financial deregulation
crucially impacts its benefits to the real economy.

Existing evidence suggests that intrastate deregulation
increased banks’ bargaining power with young, private
firms, while interstate deregulation decreased the same.
Though both intra- and interstate banking deregulation
increased consolidation in the banking industry, they had
contrasting effects on the bargaining power of young,
private firms vis-à-vis their lenders. Mergers following
intrastate deregulation often involved some overlap in
local banking markets and served to increase

concentration in local markets. Further, after intrastate
deregulation, efficient banks acquired inefficient banks
within the same state. For example, Jayaratne and
Strahan (1998) document that, after intrastate deregula-
tion, banks’ profitabilities improved as reflected in
decreased non-interest costs and, crucially, reduced loan
losses. Moreover, the market share of small banks within a
state decreased sharply (e.g., Strahan, 2002). Compared
with small banks, large banks lend disproportionately less
to small firms (e.g., Berger, Kashyap, and Scalise, 1995;
Cole, Goldberg, and White, 2004; Berger, Miller, Petersen,
Rajan, and Stein, 2005). Therefore, we argue that intrastate
deregulation increased the bargaining power of banks vis-
à-vis young, private firms by creating more efficient banks
and decreasing the proportion of small banks within
a state.

In contrast, interstate deregulation led to an active
market for corporate control among banks (e.g., Berger,
Kashyap, and Scalise, 1995; Hubbard and Palia, 1995;
Berger, 2010). Annual acquisition rates across states
increased significantly (e.g., Strahan, 2002; Stiroh and
Strahan, 2006). The lower entry barriers following inter-
state deregulation increased competition from strong, out-
of-state banks (in contrast to intrastate deregulation,
where the competition from in-state banks reduced). This
expanded the set of banks from which firms could borrow
and thereby reduced banks’ bargaining power.

Prior literature generates ambiguous predictions for the
impact of banks’ market power on the availability of credit
to young, private firms (e.g., Berger, 2010) and, therefore,
innovation by these firms. Under the traditional structure-
conduct-performance hypothesis (e.g., Berger, Hasan, and
Klapper, 2004), a decrease in competition restricts the
supply of credit and, thereby, decreases innovation. In
addition, because innovation is a risky activity, the number
of “very good” and “very bad” projects with payoffs in the
“right” and “left” tails decrease. The incomplete contract-
ing literature (e.g., Grossman and Hart, 1986; Hart and
Moore, 1990; Hart, 1995) delivers similar predictions.
Intuitively, an increase in the bargaining/holdup power
of banks lowers entrepreneurs’ ex post rents from innova-
tion and thereby dampens potential entrepreneurs’ incen-
tives to invest in innovation ex ante. Therefore, an increase
in banks’ bargaining power lowers the overall amount of
innovation.4

In contrast, studies in the relationship banking litera-
ture (e.g., Rajan 1992; Petersen and Rajan, 1994, 1995)
predict that an increase in the bargaining power of banks
could lead to more innovation by young, private firms.
An increase in banks’ bargaining power increases the

2 Other outside funding sources are trade credit at 16%, finance
company loans at 5%, venture capital investments at 2%.

3 While the venture capital market has grown considerably since the
late 1990s, it has largely focused on the high-tech and biotech sectors.
Innovative firms, especially young, private ones, in other industries
continue to depend on banks as primary sources of credit. For example,
The Austin Business Journal (May 25, 1997) reports that banks such as
Imperial Bank, Silicon Valley Bank, Texas Commerce Bank, and Bank One
have created a niche somewhere between conventional lending and
venture capital by lending to high-tech companies.

4 These arguments can be formalized in a simple framework where
risk-averse entrepreneurs obtain financing from banks for their innova-
tive projects (see Chava, Oettl, Subramanian, and Subramanian, 2012 for
details). In this framework, an entrepreneur’s expected utility declines
with project risk. Therefore, she chooses to invest in innovation if and
only if the risk is below a threshold. An increase in banks’ bargaining
power decreases the entrepreneurs ex post rents and, thereby, lowers the
threshold level of risk below which projects are undertaken ex ante.
Consequently, the amount of innovation as well as the number of projects
with payoffs in the left and right tails decrease with an increase in banks’
bargaining power.

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774760

Author’s personal copy

availability of credit to relationship borrowers by encoura-
ging banks to invest in relationships that provide future
benefits. Reduced competition helps banks to enforce
implicit contracts in which borrowers receive cheaper
loans in the short-term, but expect to pay higher rates in
the long-term. A decrease in bank competition, therefore,
encourages innovation by enhancing credit availability to
young, private firms.

The contrasting predictions for the effects of changes in
banks’ market power on innovation motivate our two
refutable hypotheses. First, an increase in banks’ bargain-
ing/market power leads to a decrease in the level of
innovation by young, private firms. Second, an increase
in banks’ bargaining/market power leads to a decrease in
the number of innovative projects undertaken by young,
private firms that have payoffs in the “right” and “left”
tails. Since intrastate deregulation increased the local
market power of banks, we hypothesize that intrastate
deregulation decreased the overall level and risk of inno-
vation by young, private firms. In contrast, since interstate
deregulation decreased banks’ bargaining power, we
hypothesize that interstate deregulation increased the
overall level and risk of innovation by young, private firms.

As proxies for the level of innovation, we use the
number of patents filed by firms as well as the cumulative
number of citations to these patents, which provide a
measure of the quality of innovation (e.g., Griliches, 1990;
Hall, Jaffe, and Trajtenberg, 2001).5 As a proxy for the risk
of innovation, we use the number of patents with “high”
citations and those with “low” citations. In our main
specifications, we include state and year fixed effects to
control for time-invariant determinants of our dependent
variables at the state level as well as time-dependent
determinants that are constant across states. In additional
specifications, we separately control for state-specific time
trends and industry-specific time trends. Our results are
both quantitatively and qualitatively unchanged by their
inclusion. These time trends allow us to more precisely
distinguish the effects of banking deregulation on innova-
tion using deviations from state- and industry-level mean
trends due to confounding factors.6

We find results consistent with our hypotheses. Intras-
tate deregulation reduced the number of patents filed by
young, private firms by 23%, while interstate banking
deregulation increased the same by 17%. In addition,

intrastate deregulation led to a decrease in the variance
of the quality of patents while interstate deregulation led
to an increase in the same. Specifically, intrastate dereg-
ulation decreased the number of good and bad patents
(patents with citations in the top and bottom quartiles of
the citation distribution) by 33% and 19%, respectively. In
contrast, interstate deregulation increased the number of
good and bad patents by 21% and 22%, respectively.

To address potential concerns about reverse causality,
we examine the dynamic effects of intrastate and interstate
deregulation on the level and risk of innovation. We find
that there were no effects prior to deregulation. While the
effects manifested within a year, they were much stronger
three years after deregulation. Apart from mitigating the
possibility of reverse causality, the time lag in the effects of
banking deregulation is consistent with the fact that
innovative projects involve long gestation periods.

In unreported results, we undertake the tests sepa-
rately for mature, private firms as well as public firms.
Such firms have alternate financing avenues, and are likely
to be less significantly affected by changes in the nature of
bank competition. Consistent with these arguments, we do
not find a significant effect of intrastate or interstate
deregulation on innovation by such firms. These falsifica-
tion tests provide further support for our hypotheses.

We also conduct tests that distinguish between
explorative and exploitative innovation (e.g., Sørensen
and Stuart, 2000) as well as product and process innova-
tion. Explorative innovation is likely to be more path-
breaking and riskier than exploitative innovation, which
builds on prior innovation by the firm (e.g., March, 1991).
We find that the changes in the level and risk of innovative
activity following deregulation were contributed largely by
explorative rather than exploitative innovation. Apart from
introducing these innovation measures into the finance
literature, a key contribution of our study is to show that
finance possibly contributes to economic growth by foster-
ing mold-breaking, explorative innovation by young, pri-
vate firms. Product innovation, which involves the creation
of new products, is likely to be riskier than process
innovation. Consistent with our hypotheses, we find that
banking deregulation impacted product innovation, but
not process innovation.

Next, we exploit interstate differences in the proportion
of large firms in a state before deregulation to further
probe the channels through which banking deregulation
affected innovation by young, private firms. Because small
firms are likely to have lower bargaining power vis-à-vis
banks, a change in bank bargaining power would dispro-
portionately affect innovation by small firms compared to
large firms. Consequently, we would expect that the
impact of deregulation was relatively muted in states with
relatively large firms when compared to states with
relatively small firms. Consistent with these arguments,
we find that the impact of banking deregulation on
innovation was, indeed, relatively muted in states where
firms were relatively large.

Motivated by the literature on endogenous growth and
evidence in Jayaratne and Strahan (1996), we finally
examine whether the differential impacts of intrastate
and interstate banking deregulations on innovation led

5 Chava, Chong, and Nanda (2012) provide evidence that lenders
value both the level and quality of patents of borrowers. They document
that ex ante firms with significant patent activity receive cheaper bank
loans as compared to firms with low patent activity. Lenders, especially
experienced banks, provide cheaper loans when patents are of high
quality, i.e., well cited and more general patents. Ex post, when covenants
are violated and control rights pass to lenders, experienced lenders cut
research and development (R&D) significantly, specifically when the
violating firm has a lower R&D efficiency. Consistent with these results,
stock markets react less negatively to technical covenant violations of
innovative firms when the bank is experienced.

6 The inclusion of a time trend for each state ensures that any
differential pre-trends in innovation across states are controlled for (e.g.,
Angrist and Pischke, 2008). Although industry-specific trends in innova-
tion are unlikely to be correlated with the timing of banking deregulation
in a state, the tests employing industry-specific time trends nevertheless
control for this possibility.

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774 761

Author’s personal copy

to similar contrasting effects on economic growth.7 For
this purpose, we construct an industry-level measure of
the propensity of young, private firms in an industry to
innovate. We classify an industry as “innovative” if the
value of the “innovation” proxy for young, private firms in
that industry is greater than the median value of the proxy
for young, private firms across all industries. By interacting
this proxy for innovative industries with the dummies for
intrastate and interstate deregulations, we find that the
positive effect of interstate deregulation on innovation led
to a positive effect on economic growth. The effect lasted
at least eight years after interstate deregulation and was
economically large. Growth due to interstate deregulation
in the innovative industries was approximately 0.40%
greater per annum than other industries. The coefficients
representing the effects of intrastate deregulation on
economic growth through innovation by young, private
firms are not statistically significant at conventional levels,
but are all uniformly negative. This result is also consistent
with intrastate deregulation having a negative effect on
economic growth because of its negative effect on innova-
tion by young, private firms.

Our results suggest that policies that are aimed at
developing financial markets can have a positive extern-
ality on the economy by boosting innovative activity and,
thereby, long-term economic growth. The negative effects
of intrastate deregulation and the positive effects of inter-
state deregulation on innovation, however, suggest that
the manner in which financial sector reform is carried out
is important to realize its potential benefits to the real
economy. To the best of our knowledge, ours is the first
study to: (i) investigate innovation by young, private firms
as a channel through which the financial sector affects
growth; (ii) show the contrasting effects of intrastate and
interstate deregulations on real outcomes; and (iii) distin-
guish between explorative and exploitative innovation and
show that financial deregulation primarily impacts mold-
breaking, explorative innovation.

The rest of the paper proceeds as follows. We briefly
review related literature in Section 2. Section 3 describes
the hypotheses we test in the paper and explains the
sources and construction of the data used in the empirical
analysis. Our empirical results are presented in Section 4.
We provide concluding thoughts in Section 5.

2. Related literature

Our study relates to the literature that examines the
real effects of financial development (e.g., Jayaratne and
Strahan, 1996; Black and Strahan, 2002; Kerr and Nanda,
2009; Beck, Demirgüç-Kunt, Laeven, and Levine, 2008;
Beck, Demirgüç-Kunt, and Maksimovic, 2004). Our study
contributes to this literature by focusing on an important
micro-level channel, innovation, through which the finan-
cial sector could affect economic growth. Our results
suggest that financial development influences both the

level and risk of innovative activities and, thereby, eco-
nomic growth.

Rice and Strahan (2010) use the differences in state
openness to interstate branching as an instrument for
variation in credit competition. They find that in states
that were more open to branching, small firms borrowed
at significantly lower interest rates than firms operating in
states that were less open. However, while more firms
used bank debt in states open to interstate branching, this
increase in bank debt did not translate into more total
borrowing, higher rates of credit approval, or changes in
debt maturity. Their results suggest that competition may
increase credit rationing even as the price of credit falls.
Our results complement Rice and Strahan (2010) by high-
lighting the differential effects of intrastate and interstate
banking deregulation and the consequent change in local
market power of lenders on the innovative performance of
small, young, private firms.

Our paper also contributes to the literature that studies
the impact of banking consolidation on measures of credit
availability and economic performance. As Berger (2010)
highlights, the empirical evidence on this issue is mixed so
far. Some studies find favorable effects of concentration
and other restrictions on competitiveness on measures of
credit availability (e.g., Petersen and Rajan, 1995; Cetorelli
and Gambera, 2001; Bonaccorsi di Patti and Dell’Ariccia,
2004; Cetorelli, 2004). Some other studies, however,
document unfavorable effects (e.g., Black and Strahan,
2002; Berger, Hasan, and Klapper, 2004; Karceski,
Ongena, and Smith, 2005; Cetorelli and Strahan, 2006).
Black and Strahan (2002) also find contrasting effects of
intrastate and interstate deregulation on economic out-
comes.8 Our results complement the findings in the
aforementioned studies by showing that intrastate and
interstate deregulations had contrasting effects on innova-
tion and economic growth.

A number of studies find that banks became larger after
deregulation. Cole, Goldberg, and White (2004) and
Berger, Miller, Petersen, Rajan, and Stein, (2005) find that
large banks are likely to lend less often to young, private
firms. On the other hand, Berger, Saunders, Scalise, and
Udell (1998) find that other banks in the local market—
particularly small banks that have a comparative advan-
tage in lending to young, private firms—substitute for
lower lending to young, private firms by large banks.
Therefore, the effects of the change in bank size due to
deregulation on innovation is not clear. As our tests that
include state-specific trends control for the effects due to

7 We replicate the results in Jayaratne and Strahan (1996) and find
that banking deregulation increased economic growth by 1.354%, which
is very close to the 1.4% that Jayaratne and Strahan (1996) estimate.

8 In Table 5 (columns 3 and 4) of their study, Black and Strahan
(2002) find that interstate deregulation had a positive effect on business
incorporations, while intrastate deregulation had a negative, statistically
significant effect on business incorporations. Cetorelli and Strahan (2006)
find that firm size decreased and firm entry increased after interstate
deregulation. However, in Table 5 (columns 1, 3, and 5) they find that the
effects of interstate and intrastate deregulations on the number of
establishments were opposite in sign to each other even though the
coefficients for the interaction of financial dependence with intrastate
deregulation are insignificant. Similarly, in Table 8 as well (columns 1, 2,
3, and 5) they find opposing effects of intra- and interstate deregulations.

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774762

Author’s personal copy

changes in bank size, our results suggest that the effects of
changes in banks’ market power dominated those of any
changes in bank size.

Three contemporaneous papers also examine the
impact of banking deregulation on innovation. Two studies
examine the effects of interstate deregulation on innova-
tion by public firms (e.g., Amore, Schneider, and Zaldokas,
2012; Cornaggia, Tian, and Wolfe, 2012); whereas one
examines the effects of intrastate deregulation on innova-
tion by all firms (e.g., Hombert and Matray, 2012). Amore,
Schneider, and Zaldokas (2012) provide evidence that
interstate banking deregulation had a beneficial impact
on innovation by public firms, whereas Cornaggia, Tian,
and Wolfe (2012) find the opposite. Hombert and Matray
(2012) find that intrastate deregulation decreased innova-
tion by all firms. We differ from these studies in five
important ways. First, we examine the effects of both
intrastate and interstate banking deregulation on innova-
tion and find contrasting effects between them. Our results
show that it is important to disentangle the effects of
intrastate and interstate deregulation on innovation. Sec-
ond, we focus on the effects of banking deregulation on
innovation by young, private firms. As we discussed earlier,
young, private firms are more likely to depend on bank
debt to finance innovation. Further, private firms are the
dominant source of path-breaking innovation. Mature,
private, and public firms can access financing from sources
other than bank financing, and innovation by such firms is
often due to spillover effects of innovation by young,
private firms. Third, we examine the effects of banking
deregulation not only on the level of innovation, but also
on the risk of innovation. Fourth, we show that the effects
of deregulation on innovation by young, private firms were
associated with corresponding effects on economic growth
as well. Finally, we distinguish between explorative and
exploitative innovations and show that changes in innova-
tion due to deregulation stemmed primarily from rela-
tively more path-breaking and riskier, explorative
innovation. As the concept of “creative destruction” as
enunciated by Schumpeter (1942) suggests that explora-
tive/path-breaking product innovation is the main driver
of economic growth, our study contributes to the literature
by studying such nuanced differences among different
types of innovation.

Our work also relates to the emerging literature exam-
ining the effects of laws and regulations on innovation.
Acharya and Subramanian (2009) find that debtor-friendly
bankruptcy laws foster innovation and economic growth,
while Acharya, Baghai, and Subramanian (2012) argue
theoretically and provide empirical evidence that laws
that impose restrictions on dismissal of employees
encourage innovation and entrepreneurship. Chemmanur
and Tian (2012) find that firm-level anti-takeover provi-
sions encourage innovation. Sapra, Subramanian, and
Subramanian (2013) develop a theory of the effects of
external and internal corporate governance mechanisms
on innovation. They show empirical support for the
testable prediction of their theory that the degree of
innovation varies in a U-shaped manner with the severity
of external takeover pressure as measured by the strin-
gency of state-level anti-takeover laws.

3. Hypotheses and data

3.1. Testable hypotheses

Our discussion in Section 1 leads to the following
refutable hypotheses:

Hypothesis 1. Intrastate deregulation led to a decrease in the
level of innovation by young, private firms.

Hypothesis 2. Intrastate deregulation led to a decrease in the
number of innovative projects undertaken by young, private
firms that have payoffs in the “high” and “low” tails.

Hypothesis 3. Interstate deregulation led to an increase in
the level of innovation by young, private firms.

Hypothesis 4. Interstate deregulation led to an increase in
the number of innovative projects undertaken by young,
private firms that have payoffs in the “high” and “low” tails.

3.2. USPTO data on patents and citations

To construct our innovation measures, we use patents
filed by U.S. firms with the United States Patent and
Trademark Office (USPTO) and the citations to these
patents, compiled in the National Bureau of Economic
Research (NBER) Patents File (e.g., Hall, Jaffe, and
Trajtenberg, 2001).9 We date our patents according to
the year in which they were applied for to avoid anomalies
that may be created due to the lag between the date of
application and the date of granting of the patent (e.g.,
Hall, Jaffe, and Trajtenberg, 2001). Note that although we
use the application year as the relevant year for our
analysis, the patents appear in the database only after
they are granted. Hence, we use the patents actually
granted (rather than the patent applications) for our
analysis.

The unit of analysis is the state-year. We include all 51
states (including the District of Columbia) and the years
from 1975 to 2005 following Beck, Levine, and Levkov,
(2010), which leads to a final sample of 1,581 (51�31)
observations. In some specifications, we measure innova-
tion at the level of the one-digit NBER technology class (e.
g., Hall, Jaffe, and Trajtenberg, 2001), which results in
9,486 observations (51 states�31 years�6 technology
classes). We drop firms with fewer than two lifetime
(between 1975 and 2011) patents in a given state.

Patents have long been used as indicators of innovative
activity in both micro- and macro-economic studies (e.g.,
Pakes and Griliches, 1980; Griliches, 1990). Although
patents provide an imperfect measure of innovation, there
is no other widely accepted method that can be applied to
capture technological advances. An an alternative to
patents, R&D spending at the firm/industry level could

9 The NBER patent data set provides (among other items) annual
information on patent assignee names, the number of patents, the
number of citations received by each patent, the technology class of the
patent, and the year that the patent application is filed. The USPTO
defines “assignee” as the entity to which a patent is assigned. To link the
patent data with Compustat, we exploit the fact that each assignee in the
NBER patent data set is given a unique and time-invariant identifier.

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774 763

Author’s personal copy

be a potential proxy for innovation. However, this presents
several challenges. First, accounting norms—particularly
whether R&D is capitalized or is expensed—would have a
mechanical effect on R&D spending. Because such prac-
tices may vary across firm/industries, these mechanical
effects may influence the measures of R&D spending.10

Moreover, R&D spending represents the input to innova-
tion while patents and citations capture the output of
innovation. In any case, Griliches (1990) emphasizes that
there is a strong relationship in the U.S. between R&D and
the number of patents received at the cross-sectional level
across firms and industries. The median R2 is of the order
of 0.9.

3.3. Proxies for the level and risk of innovation: dependent
variables

We measure the level of innovation using the
following:

� Number of patents: This variable captures a simple
count of the number of distinct patents applied for
(and subsequently granted) by assignees in state s in
year t.

� Number of citations: This variable captures the number
of citations to patents applied for (and subsequently
granted) by assignees in state s in year t. Citations
capture the importance of a patent (e.g., Pakes and
Griliches, 1980).11 While we only examine patenting
activity that takes place between 1975 and 2005, we
use patent-citation data up through and including 2011
that we obtain directly from the USPTO.

We measure the risk of innovation using the following:

� Number of patents with “high” citations: the number of
patents applied for (and subsequently granted) by
assignees in state i in year t whose citations are above
the 75th percentile (4th quartile) of year t’s citation
distribution.

� Number of patents with “low” citations: the number of
patents applied for (and subsequently granted) by
assignees in state i in year t whose citations are below
the 25th percentile (1st quartile) of year t’s citation
distribution.

We identify a patent assignee as a private firm if there
is no GVKEY match for the assignee in the NBER patent
database. In our main tests, we classify a private firm as
young if it has three or fewer years of patenting experience
in the focal state. In robustness tests in Table 7, we also

consider break points of five and ten years to define young
firms among the class of private firms.

3.4. Banking deregulation: independent variables

Given our hypotheses, the two independent variables of
interest are the following:

� Intra is a dummy variable that switches to one the year
after the focal state implemented intrastate banking
deregulation. Since intrastate banking deregulation
consisted of both de novo and mergers and acquisitions
(M&A) deregulation, we follow previous literature (e.g.,
Jayaratne and Strahan, 1996, 1998) by classifying a state
as “intrastate deregulated” the year after either de novo
or M&A deregulation.

� Inter is a dummy variable that switches to one the year
after the focal state implemented interstate banking
deregulation (e.g., Kerr and Nanda, 2009; Jayaratne and
Strahan, 1996, 1998).

3.5. Descriptives

Table 1 presents summary statistics of our key vari-
ables. The mean state applies for just under 850 patents in
an average year and receives 12,539 citations to these
patents. Younger firms (three years or less experience) in a
state account for fewer patents, but their average patent
quality as measured by citations is higher (16.5 citations
per patent) than older firms (14.3 citations per patent).

3.6. Graphical evidence

Fig. 1 presents plots from spline regressions of the
relationship between intrastate and interstate deregula-
tion on young, private firms. The graphs represent coeffi-
cients plots (and their associated 95% confidence intervals
as represented by the vertical bars) from a Poisson
regression of the number of patents on a series of dummy
variables corresponding to pre-treatment leads (years up
to and including t−3, and t−2) and post-treatment lags (t0,
…,t5, and years t6 and all subsequent years). The omitted
variable in these splines is the year before intrastate
(interstate) deregulation, which implies that the coeffi-
cients in these plots capture the rate of change in patent-
ing in any year vis-à-vis the year before intrastate
(interstate) deregulation.

In Fig. 1, we present the results from the spline
regressions for young, private firms after intrastate dereg-
ulation and interstate deregulation, respectively. In these
graphs, we first notice that the rate of change in patenting
before the deregulation is insignificant. If bank deregula-
tion caused a change in innovation but not vice versa, then
the rate of patenting in the year before deregulation
should be statistically indistinguishable from all other
years prior to deregulation. This is indeed what we
observe, which suggests that reverse-causal effects of
changes in innovation in a state on the deregulation are
not very plausible in our setting. Second, when compared

10 It is also possible that financing constraints at the firm/industry
level influence the decision to expense/capitalize R&D; if that were the
case, the impact on financing constraints due to banking deregulation
would introduce a source of endogeneity that would mar identification.

11 Pakes and Schankerman (1984) show that the distribution of the
importance of patents is extremely skewed; i.e., most of the value is
concentrated in a small number of patents. Hall, Jaffe, and Trajtenberg
(2001) (among others) demonstrate that patent citations are a good
measure of the value of innovations.

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774764

Author’s personal copy

to the year before intrastate deregulation, we notice a
statistically significant decrease in patenting in the year of
intrastate deregulation. Third, when compared to the year
before interstate deregulation, we notice a statistically
significant increase in patenting one year after interstate
deregulation. Thus, while the effects of intrastate dereg-
ulation manifested in the year of deregulation itself, the
effects of interstate deregulation manifested one year later.
Fourth, we find that the decrease (increase) in patenting
following intrastate (interstate) deregulation continues to
remain statistically significant for at least six years after
deregulation occurs.

4. Results

4.1. Empirical strategy

Our main econometric model focuses on the relation-
ship between indicator variables for intrastate and inter-
state banking deregulation and our proxies for the level
and risk of innovative activity. The empirical specification
we estimate is as follows:

E½Yit � ¼ expðβnIntrait þ γnInterit þ δt þ ϕiÞ; ð1Þ
where Yit is one of our four dependent variables measured
in state i in year t, δt represents a set of year dummies, and
ϕi a set of state dummies. The inclusion of state dummies
results in the identification of β and γ solely from within-
state variation across time. All time-invariant characteris-
tics of the state that may influence its innovative output
are controlled for with state dummies. For example, in a
specification without state fixed effects, a state’s (time-
invariant) comparative advantage/disadvantage in innova-
tive activity may influence the estimated effect of banking
deregulation on innovation if deregulation occurs earlier/
later in a systematically correlated manner with such

state-level, unobserved factors. All state-invariant time
trends are controlled for with the time dummies. Thus,
any secular (i.e., across the U.S.) shocks in innovation
coinciding with the timing of banking deregulation (which
varies from state to state) are controlled for using the time
fixed effects.

Due to the count-based nature of our dependent vari-
ables, we employ a fixed-effects Poisson estimator (e.g.,
Hausman, Hall, and Griliches, 1984). This estimator is
computationally straightforward and also has strong
robustness features when estimated by quasi-maximum
likelihood (QML). We report “Wooldridge” robust standard
errors (e.g., Wooldridge, 1999) that are robust to over-
dispersion, are valid under any variance assumption, and
allow for arbitrary serial correlation that is often found in
difference-in-differences settings (e.g., Bertrand, Duflo,
and Mullainathan, 2003).

Since deregulation occurred gradually and states
deregulated at different times, we can use states that
had not deregulated at a point in time to control for
potentially confounding effects and thereby estimate a
difference-in-differences: the difference in the level of
innovation in a state before and after the deregulation
compared to this difference for states that did not undergo
a deregulation during the same period. Furthermore,
because interstate deregulation followed intrastate dereg-
ulation after a gap of a few years, and the gap between
intrastate and interstate deregulation varied across states,
we are able to disentangle the effects of these two forms of
deregulation.

4.2. Effects on innovation by young, private firms

We focus our empirical analysis on patenting activity
by young, private firms. As discussed earlier, since young,
private firms depend primarily on bank debt for external

Table 1
Descriptive statistics.

The sample consists of 1,581 observations during 1970–2005 with the unit of analysis as state–year. Panel A reports the descriptive statistics for all firms
while Panel B and Panel C report descriptive statistics for subsamples that consist of private and public firms, respectively. “Intra dereg” (Inter dereg) is a
dummy variable that turns to one the year after the focal state implemented intrastate (interstate) banking deregulation. “Patents” is the number of distinct
patents applied for (and subsequently granted) by assignees in state s in year t. “Citations” is the number of citation-weighted patents applied for (and
subsequently granted) by assignees in state s in year t. “1st Quartile cites” (4th Quartile cites) is the number of patents applied for (and subsequently
granted) by assignees in state s in year t that are in the 1st (4th) quartile of year t’s citation distribution.

Panel A: all firms Panel B: private firms Panel C: public firms

Variables Mean Std. dev. Min Max Mean Std. dev. Min Max Mean Std. dev. Min Max

Patents 849.15 1,718.73 0 19,535 296.57 522.2 0 6,485 559.15 1,266.32 0 13,541
Patents≤3 yrs 184.77 383.77 0 5,332 125.15 243.96 0 3,332 61.71 149.5 0 2,072
Patents43 yrs 664.38 1,379.32 0 14,329 171.42 304.95 0 3,153 497.44 1,144.56 0 11,498
Citations 12,539.45 29,322.47 0 378,168 4,283.3 9,157.76 0 116,817 8,349.62 21,056.84 0 263,879
Citations≤3 yrs 3,048.53 8,141.2 0 118,574 2,011.74 4,953.52 0 70,637 1,068.78 3,362.21 0 52,033
Citations43 yrs 9,490.92 22,000.61 0 259,594 2,271.55 4,501.08 0 53,217 7,280.83 18,272.08 0 211,846
1st Quartile cites 243.17 473.64 0 4,945 86.19 147.67 0 1,569 159.2 353.34 0 3,450
1st Quartile cites ≤3 yrs 47.81 93.36 0 1,320 32.08 57.88 0 820 16.38 38.99 0 534
1st Quartile cites 43 yrs 195.36 393.55 0 3,736 54.11 99.57 0 890 142.82 323.11 0 2,937
4th Quartile cites 204.45 477.38 0 5,844 69.75 149.35 0 1,987 136.22 342.18 0 3,979
4th Quartile cites ≤3 yrs 50.4 131.49 0 1,757 33.45 82.74 0 1,124 17.46 51.39 0 678
4th Quartile cites 43 yrs 154.05 357.62 0 4,087 36.29 71.17 0 883 118.76 298.31 0 3,310
Intra dereg. 0.71 0.45 0 1 0.71 0.45 0 1 0.71 0.45 0 1
Inter dereg. 0.59 0.49 0 1 0.59 0.49 0 1 0.59 0.49 0 1

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774 765

Author’s personal copy

financing, we expect the effects of banking deregulation to
manifest primarily for these firms.

Table 2 shows the effects of banking deregulation on
innovation by young, private firms (firms that do not have
a Compustat GVKEY assigned in the NBER patent data set
and that have three or fewer years of patenting experience
in state i in year t). In specifications 1–4, we estimate Eq.
(1) where we include state and year fixed effects. In
specifications 5–8, we control for the presence of time
trends that could differ across states by employing the
following specification:

E½Yit � ¼ expðβnIntrait þ γnInterit þ δt þ ϕi þ t � ϕiÞ; ð2Þ
where Yit is one of our four dependent variables measured
in state i in year t. Doing so enables to not only allow states

to have different intercepts ðϕiÞ, but also different slopes
across time ðt � ϕiÞ.

In specifications 9–12, we include fixed effects corre-
sponding to the technology-class as well as time trends
that could vary by technology-class by employing the
following specification:

E½Yikt � ¼ expðβnIntrait þ γnInterit þ δt þ ϕi þ ηk þ ηk � tÞ; ð3Þ
where the difference compared to Eq. (1) pertains to the
inclusion of technology-class effects ðηkÞ and trends ðt � ηkÞ
in addition to the state and year fixed effects. By account-
ing for state-specific and technology-specific time trends
in specifications 5–8 and 9–12, respectively, we identify
the hypothesized effect using deviations (at the
technology-class level) from the average time trend for
each state and that for each technology class. Since other
state- or technology-level factors accompanying banking
deregulation could lead to state-specific as well as
technology-specific time trends, these controls enable us
to isolate precisely the pure effect of deregulation on
innovation.

Specifications that control for state-specific time trends
are particularly important because we attempt to estimate
causal effects of banking deregulation on innovation
through a difference-in-differences framework. As
Angrist and Pischke (2008) describe “(difference-in-differ-
ences) strategies punt on comparisons in levels, while
requiring the counterfactual trend behavior of treatment
and control groups to be the same.” Inclusion of a time
trend for each state ensures that any differential pre-
trends in innovation in the treatment and control groups
are controlled for.

Across the three different sets of specifications
described above, we find the coefficient of intrastate
deregulation to be negative and the coefficient of inter-
state deregulation to be positive. The coefficients are
statistically significant at the 5% level or lower in all the
specifications. The results presented in Table 2 are, there-
fore, consistent with our four hypotheses. In particular, the
coefficients of intrastate and interstate deregulation do not
change significantly across columns 1, 5, and 9. This
pattern is repeated for each of the four dependent vari-
ables that we employ in Table 2. If differential time trends
in innovation across states were correlated with banking
deregulation, then the resulting omitted-variable bias
would alter the coefficient of intrastate and interstate
deregulation when we include the state-specific time
trends. However, the fact that the coefficient remains
unchanged across the three different specifications for
each of the four dependent variables suggests that the
estimates are unlikely to be affected by state-specific nor
technology-specific trends. Our preferred specifications,
therefore, exclude the more computationally intensive
year trends and base statistical inference on the inclusion
of state and year fixed effects as in columns 1–4.

Quantitatively, intrastate banking deregulation resulted
in a 23% ðe−0:266−1¼ −0:23Þ decrease in patents and a 32%
decrease in citation-weighted patents being filed by
young, private firms. Moreover, the mass of patents in
the left and right tails of the citation distribution
decreased by 19% and 33%, respectively, reflecting a

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

P
oi

nt
e

st
im

at
es

r
el

at
iv

e
to

y
ea

r
be

fo
re

de

re
gu

la
tio

n

-3 -2 -1 0 1 2 3 4 5 6
Years before/after intrastate deregulation

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

P
oi

nt
e

st
im

at
es

r
el

at
iv

e
to

y
ea

r
be

fo
re

de
re

gu
la

tio
n

-3 -2 -1 0 1 2 3 4 5 6
Years before/after interstate deregulation

Fig. 1. This figure presents patenting rates for private firms≤3 years of
age before and after intra- (Panel a) and interstate banking deregulation
(Panel b). This figure plots point estimates for leading and lagging
indicators of banking deregulation of the following specification:
Patentsst ¼ expðα−3 Intrast−3þ α−2 Intrast−2 þ α0Intrastþ ⋯þ α6Intrastþ6þ
β−3 Interst−3 þ β−2 Interst−2 þ β0Interst þ⋯þ β6Interstþ6 þ μs þ δt Þ. “Patents”
is the number of distinct patents applied for (and subsequently granted)
by assignees in state s in year t. “Intrast” (Interst) is a dummy variable set
to one if state s deregulated in year t and zero otherwise.
Intrast−3=Interst−3 is set to one for years up to and including three years
prior to intrastate/interstate banking deregulation and zero otherwise.
Intrastþ6=Interstþ6 is set to one for all years six years after intrastate/
interstate banking deregulation and zero otherwise. μs and δt are state
and year fixed effects, respectively. The vertical bars correspond to 95%
confidence intervals with state-clustered standard errors.

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774766

Author’s personal copy

reduction in the risk of innovative activity undertaken by
young, private firms. In contrast, the level of innovative
activity of young, private firms increased significantly after
interstate banking deregulation. Patenting increased 17%
after interstate banking deregulation while citation-
weighted patenting increased by 16%. The mass of patents
in the left and right tails of the citation distribution also
increased 22% and 21%, respectively.

Overall, the results in Table 2 support our hypotheses.
We find that banking deregulation significantly affected
the level and risk of innovative activity by young, private
firms. Further, intrastate and interstate deregulations had
contrasting effects on the level and risk of innovation by
young, private firms.

4.3. Falsification tests

As we discussed earlier, we would not expect to find a
significant effect of banking deregulation on the level and
risk of innovation by mature, private, or public firms. To
investigate this, we estimate Eq. (1) on the innovative
activity of three different samples of firms: mature private
firms, public firms, and the aggregate sample of all patents
in the NBER patent data set. We do not find a significant
effect of intrastate or interstate deregulation on either the
level or the risk of innovative activity undertaken by either
set of these firms that are less dependent on bank finan-
cing. We omit these results in the interest of brevity; they
are available from the authors on request.

Our investigation of the effects of banking deregulation
on innovation by mature, private, or public firms serves as
a useful “placebo” test of our hypotheses. The presence of a
significant impact of banking deregulation exactly where it
is predicted to have one—innovation by young, private
firms—and not for other firms provides support for the
channels through which intrastate and interstate banking
deregulations impacted innovation.

4.4. Temporal dynamics of innovation by young, private
firms

Kroszner and Strahan (1999) suggest that state-level
factors that manifested differently across states could have
affected the timing of deregulation in different states. If
the states also differed in their innovation intensities, it
could be the case that such differences triggered the
deregulation, thereby suggesting the presence of a
reverse-causal relationship between deregulation and
innovation. Our earlier tests, in which we controlled for
state-specific time trends (that is, time trends in innovation
that differ across states) significantly mitigate such con-
cerns. To further explore the possibility of reverse caus-
ality, we examine the dynamics of innovation by young
firms. If reverse causality is indeed present, we should see
changes in innovation prior to the deregulation events.

In Table 3, we examine the dynamics of innovation by
young, private firms following banking deregulation. To
investigate the temporal dynamics, we introduce a series
of timing dummies. Intrað≤−2Þ=Interð≤−2Þ is a dummy
variable set to one for all years up to and including two
years prior to intrastate/interstate banking deregulation.
Intra(0)/Inter(0) is set to one the year intrastate/interstate
deregulation occurs. Intra(1,3)/Inter(1,3) is set to one for
years 1, 2, and 3 after intrastate/interstate deregulation.
Intrað43Þ=Interð43Þ is set to one for years that are more
than three years after intrastate/interstate deregulation.
The omitted category is the year before bank deregulation.

The coefficients of Intrað≤−2Þ=Interð≤−2Þ are all insig-
nificant, indicating that the level and risk of innovative
activity of young, private firms were not significantly
affected prior to intrastate or interstate deregulation. The
results of our earlier tests that include state-specific time
trends in addition to the finding that there were no effects
on innovation by young, private firms prior to deregulation
substantially alleviate concerns about reverse causality.

Table 2
Effect of bank deregulation on innovation by young, private firms.

The following table reports regression results for young, private firms. “Intra” (Inter) is a dummy variable that turns to one the year after the focal state
implemented intrastate (interstate) banking deregulation. “Patents” is the number of distinct patents applied for (and subsequently granted) by assignees
in state s in year t. “Citations” is the number of citation-weighted patents applied for (and subsequently granted) by assignees in state s in year t. “1Q Cites”
(4Q Cites) is the number of patents applied for (and subsequently granted) by assignees in state s in year t that are in the first (fourth) quartile of year t’s
citation distribution. Results in columns 1–4 include year fixed effects and state fixed effects. Results in columns 5–8 include state x year trends in addition
to the year and state fixed effects. Columns 9–12 include technology class x year trends in addition to the year and state fixed effects. State-clustered robust
standard errors are reported in parentheses. n, nn, and nnn indicate significance better than 10%, 5%, and 1%, respectively.

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
Dependent variable Patents Citations 1Q Cites 4Q Cites Patents Citations 1Q Cites 4Q Cites Patents Citations 1Q Cites 4Q Cites

Intra −0.266nn −0.381nnn −0.213nn −0.397nnn −0.260nn −0.374nnn −0.209nn −0.388nnn −0.267nn −0.391nnn −0.206nn −0.405nn

(0.115) (0.137) (0.099) (0.154) (0.113) (0.134) (0.098) (0.151) (0.117) (0.139) (0.101) (0.158)

Inter 0.160nnn 0.150nn 0.198nnn 0.192nnn 0.158nnn 0.148nn 0.197nnn 0.189nnn 0.168nnn 0.157nn 0.205nnn 0.206nnn

(0.058) (0.070) (0.069) (0.072) (0.057) (0.069) (0.070) (0.071) (0.062) (0.074) (0.073) (0.075)

Year FE ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
State FE ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
State time trend ✓ ✓ ✓ ✓
Tech. class FE ✓ ✓ ✓ ✓
Tech. time trend ✓ ✓ ✓ ✓

Observations 1,581 1,581 1,581 1,581 1,581 1,581 1,581 1,581 9,486 9,486 9,486 9,486
Num. states 51 51 51 51 51 51 51 51 51 51 51 51

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774 767

Author’s personal copy

Across all the specifications we see the effects of bank
deregulation on the level and risk of innovative activity of
young, private firms increase as the time since deregula-
tion increases. The results show that the effects of intras-
tate deregulation manifested in the first year of deregu-
lation, while the effects of interstate deregulation take
more than one year to commence.

4.5. Evidence on the channels

4.5.1. Explorative and exploitative innovation
We now conduct further tests to probe the channels

through which banking deregulation affected innovation
by young, private firms. Explorative innovation is likely to
be riskier and characterized by a greater degree of asym-
metric information than exploitative innovation, which
builds on prior innovation by the firm. The strategy
literature contends that exploitative innovation involves
the refinement and extension of existing technologies and

leads to incremental innovation as it is based on “localized
learning.” In contrast, explorative innovation involves
experimentation with new alternatives to the firm and
contributes to radical innovation since it is based on
“learning-by-experimentation” (e.g., Henderson, 1993;
Levinthal and March, 1993). Moreover, exploitative inno-
vation increases the efficiency of existing technologies
while explorative innovation is more likely to lead to the
discovery of new high-quality and high-impact technolo-
gies (e.g., Ahuja, 2000).

Following Sørensen and Stuart (2000), we define
exploiting patents as those that include at least one self-
citation by the assignee. In other words, if the assignee of a
focal patent includes a citation to a patent previously
issued to the same assignee, we call this an exploiting
patent. Intuitively, a patent that builds upon a firm’s prior
innovations exploits previous innovations by the firm.
Conversely, an exploring patent is one that does not include
any self-citations. Consistent with anecdotal evidence, and
the evidence discussed above that young, private firms are
more likely to indulge in path-breaking innovation,
approximately 83% of patents filed by young, private firms
are exploring, while 63% of those filed by older, private
firms are exploring.

In Table 4, we present results from our classification of
patents into exploring and exploiting patents. Consistent
with the intuition that exploratory innovation is likely to
be riskier and more path-breaking than exploitative

Table 4
Effect of bank deregulation on exploring/exploiting patents and product/
process patents.

The following table reports regression results of exploration/exploita-
tion patents, and product/process patents of young, private firms. “Intra”
(Inter) is a dummy variable that turns to one the year after the focal state
implemented intrastate (interstate) banking deregulation. “Exploring
patents” is the number of distinct process patents applied for (and
subsequently granted) by assignees in state s in year t that do not include
any citations to the assignee’s prior patents. These patents can thus be
seen as exploring new technological domains. “Exploiting patents” is the
number of distinct patents applied for (and subsequently granted) by
assignees in state s in year t that include one or more citations to the
assignee’s prior patents. These patents thus build upon the firm’s prior
innovative activities. “Product patents” is the number of distinct product
patents applied for (and subsequently granted) by assignees in state s in
year t. “Process patents” is the number of distinct process patents applied
for (and subsequently granted) by assignees in state s in year t. State-
clustered robust standard errors are reported in parentheses. n, nn, and
nnn indicate significance better than 10%, 5%, and 1%, respectively.

(1) (2) (3) (4)

Dependent variable Exploring Exploiting Product Process
patents Patents Patents Patents

Intra −0.273nn −0.164 −0.270nn −0.112
(0.111) (0.146) (0.116) (0.120)

Inter 0.151nnn 0.165 0.170nnn −0.135
(0.056) (0.104) (0.058) (0.160)

Year FE ✓ ✓ ✓ ✓
State FE ✓ ✓ ✓ ✓

Observations 1581 1581 1581 1581
Num. states 51 51 51 51

Table 3
Dynamic effects of bank deregulation and innovation.

The following table explores the temporal dynamics of innovation by
young, private firms. “Intrað≤−2Þ” (Interð≤−2) is a dummy variable set to
one for all years up to and including two years prior to intrastate
(interstate) banking deregulation. “Intra(0)” (Inter(0)) is set to one the
year intrastate (interstate) deregulation occurs. “Intra(1,3)” (Inter(1,3)) is
set to one for years 1, 2, and 3 after intrastate (interstate) deregulation.
“Intrað43Þ” (Interð43Þ) is set to one for all years three years after
intrastate (interstate) deregulation. “Patents” is the number of distinct
patents applied for (and subsequently granted) by assignees in state s in
year t. “Citations” is the number of citation-weighted patents applied for
(and subsequently granted) by assignees in state s in year t. “1Q Cites” (4Q
Cites) is the number of patents applied for (and subsequently granted) by
assignees in state s in year t that are in the first (fourth) quartile of year t’s
citation distribution. State-clustered robust standard errors are reported
in parentheses. n, nn, and nnn indicate significance better than 10%, 5%,
and 1%, respectively.

(1) (2) (3) (4)
Dependent variable Patents Citations 1Q Cites 4Q Cites

Intrað≤−2Þ −0.160 −0.025 −0.246 −0.108
(0.180) (0.139) (0.226) (0.168)

Intra(0) −0.175nnn −0.097nn −0.239nnn −0.130nn

(0.054) (0.045) (0.072) (0.058)

Intra(1,3) −0.238nnn −0.238nnn −0.292nnn −0.289nnn

(0.090) (0.073) (0.112) (0.091)

Intrað43Þ −0.484nnn −0.512nnn −0.484nnn −0.585nnn
(0.151) (0.138) (0.169) (0.162)

Interð≤−2Þ −0.105 −0.139n −0.129 −0.139
(0.081) (0.072) (0.111) (0.098)

Inter(0) 0.000 0.026 −0.038 0.097n

(0.048) (0.062) (0.070) (0.057)

Inter(1,3) 0.127nn 0.121n 0.135n 0.186nn

(0.061) (0.074) (0.082) (0.079)

Interð43Þ 0.212nn 0.186n 0.264nnn 0.323nnn
(0.087) (0.110) (0.093) (0.109)

Year FE ✓ ✓ ✓ ✓
State FE ✓ ✓ ✓ ✓

Observations 1,581 1,581 1,581 1,581
Num. states 51 51 51 51

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774768

Author’s personal copy

innovation, we observe a 24% decrease in exploring
patents by young, private firms after intrastate deregula-
tion, and an increase of 16% after interstate deregulation in
specification 1. In contrast, in specification 2, both intras-
tate and interstate deregulation do not have a significant
impact on exploiting innovative activity. In unreported
results, we find that banking deregulation did not affect
explorative or exploitative innovation by mature and
public firms. The results of Table 4 provide further support
for our central premise that banking deregulation affected
both the level and quality of innovations produced by
young, private firms. Our findings show that banking
deregulation largely did so by affecting riskier, path-break-
ing, exploratory innovation by young, private firms.

4.5.2. Product and process innovation
Product innovation, which leads to the creation of new

products, is likely to involve a higher level of innovation
and more risk than process innovation, which typically
involves improving the efficiency of existing production
processes. As a result, we should see the effects of dereg-
ulation manifest more strongly for product innovation
rather than process innovation.12

In columns 3 and 4 of Table 4, we show the results of
tests in which we classify patents according to whether
they are product patents or process patents. To construct
our product and process patent measures, we focus on the
International Patent Classification (IPC) category that lar-
gely focuses on Physical or Chemical Processes, IPC cate-
gory B01. We construct a count of all patents that fall into
IPC category B01 as process patents, and all others as
product patents. Using this proxy, we find that approxi-
mately 3% of the patents filed correspond to process
patents.

Consistent with the intuition discussed above, we see
that intrastate (interstate) deregulation had a significant
negative (positive) effect on product innovation by young,
private firms, but no effect on process innovation. In
unreported results, we also find that banking deregulation
did not have a significant effect on either product or
process innovation by mature, private firms or public
firms.

While the above results provide further support for our
hypotheses, we should mention that our classification of
patents into product and process patents is not entirely
clean: patents in other IPC categories may also include
process patents. These results, therefore, should be viewed
as primarily supportive in nature.

4.5.3. Differential impact based on percentage of small firms
in the state

We exploit interstate differences in the proportion of
large firms in a state before deregulation to further probe
the channels through which banking deregulation affected
innovation by young, private firms. Because small firms are

likely to have lower bargaining power vis-à-vis banks, a
change in bank bargaining power would disproportio-
nately affect innovation by small firms relative to large
firms. Therefore, if the changes in banks’ bargaining power
due to intrastate and interstate deregulations indeed
explains the above results, then the impact of the dereg-
ulations would be relatively muted in states where firms
were relatively large before the deregulations when com-
pared to states where firms were relatively small.

In Table 5, we consider the impact of banking dereg-
ulation on innovation in states that are sorted based on the
share of small firms in the state before deregulation. We
employ the following specification:

E½Yikt � ¼ expðβ � Intrait þ γ � Interit þ ½η � Intrait þ ρ � Interit �
nFew Small Firmsi þ δt þ ϕiÞ: ð4Þ

In the above, “Few small firms” is a dummy set to one if the
share of small firms in the focal state was below the mean
of the year’s small firm share the year prior to deregula-
tion, and zero otherwise. This variable is time-invariant
within states. We find that intrastate banking deregulation
has a negative effect, and interstate banking deregulation
has a positive effect on the level and riskiness of innova-
tion of young, private firms. Crucially, we notice that the
coefficient of the interaction of interstate deregulation
with the dummy for states where the share of small firms
was low is consistently negative and statistically signifi-
cant. We can, therefore, infer that the impact of interstate
deregulation on innovation by young, private firms is
much lower in states where the share of small firms is
lower as compared to the states where the share of small
firms is higher. Similarly, we notice that the coefficient of
the interaction of intrastate deregulation with the dummy
for states where the share of small firms was low is
consistently positive and statistically significant in two of
the four specifications. We can, therefore, infer that the
impact of intrastate deregulation on innovation by young,
private firms is much lower in states where the share of
small firms is lower as compared to the states where the
share of small firms is higher. These results provide
additional support for our argument that banking dereg-
ulation affected innovation by altering banks’ bargaining
power vis-à-vis young, private firms.

4.6. Effects on economic growth through innovation by
young, private firms

As the literature on endogenous growth (e.g., Romer,
1990; Grossman and Helpman, 1991; Aghion and Howitt,
1992) posits that firm-level innovation is an essential
ingredient for economic growth, we now investigate
whether the differential impacts of intrastate and inter-
state banking deregulations on innovation led to corre-
sponding contrasting effects on economic growth. This
investigation assumes importance given the evidence in
Jayaratne and Strahan (1996) that banking deregulation
fostered economic growth as measured by the Gross State
Product (GSP). To begin with, we replicate the results in
Jayaratne and Strahan (1996). For this purpose, we

12 In a recent paper, Benfratello, Schiantarelli, and Sembenelli (2008)
investigate the effect of local banking development on firms’ innovative
activities using a rich data set on innovation for a large number of Italian
firms. They present evidence that banking development affects the
probability of process innovation, particularly for firms that are small.

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774 769

Author’s personal copy

undertake the following regression:

yit ¼ δt þ ϕi þ βnDeregulationit þ εit ; ð5Þ
where yit ¼ gspit=gspi;t−1−1 equals the growth rates in GSP
in state i in year t, Deregulationit equals one for states
permitting M&A and zero otherwise as in Jayaratne and
Strahan (1996). Column 1 in Table 6 shows that deregula-
tion increased economic growth by 1.354%, which is very
close to the 1.4% that Jayaratne and Strahan (1996)
estimate.

To examine the effect of banking deregulation on
economic growth through the innovative activity of young,
private firms, we construct an industry-level measure of
the propensity of young, private firms in an industry to
innovate. For this purpose, we match the industry classi-
fication employed by the Bureau of Economic Analysis
(BEA) to the technology classes employed in the USPTO
data. We perform this match by first matching the BEA
industry classification to the two-digit Standard Industrial
Classification (SIC) and then using the matching of tech-
nology classes in the USPTO data to the two-digit SIC
match. We classify a BEA industry as “innovative” if the
value of the innovative proxy for young, private firms in
that industry is greater than the median value of the
innovative proxy for young, private firms across all BEA
industries over the entire sample period. For example,
using the number of patents filed by young, private firms
as the proxy for innovation, we define the “Innovative
industry dummy” for BEA industry j as follows:

Innovative industry dummyj

¼
1 if ∑

t
Patentsjt4Median

j
∑
t
Patentsjt

� �

0 if ∑
t
Patentsjt≤Median

j
∑
t
Patentsjt

� �
:

8>>>< >>>:

ð6Þ

Defining innovative industries in this time-invariant man-
ner avoids any endogeneity in the classification of indus-
tries due to the effect of the regulation.

Since the effect of banking deregulation on economic
growth through the innovation channel would manifest
over time, we employ the following specification that
accounts for dynamic effects:

yijt ¼ ϕi þ μj þ δt ½β−1 � Intrað−4;−1Þit þ β0 � Intrað0;3Þit
þβ1 � Intrað≥4Þit �nInnovative industry dummyj
þ½γ−1 � Interð−4;−1Þit þ γ0 � Interð0;3Þit
þγ1 � Interð≥4Þit �nInnovative industry dummyj
þ½β′−1 � Intrað−4;−1Þit þ β0′ � Intrað0;3Þit
þβ′1 � Intrað≥4Þit � þ ½γ′−1 � Interð−4;−1Þit
þγ′0 � Interð0;3Þit þ γ′1 � Interð≥4Þit � þ εit : ð7Þ

In the above, β−1; β0; β1 ðγ−1; γ0; γ1Þ together capture the
dynamic effects of intrastate (interstate) banking dereg-
ulation on economic growth through innovation by young,
private firms. β−1ðγ−1Þ captures the effect of intrastate
(interstate) four years prior to one year prior to deregula-
tion, β0ðγ0Þ captures the effect from the year of intrastate
(interstate) deregulation to three years after, while β1ðγ1Þ
captures the effect of intrastate (interstate) deregulation
four years after and beyond. The omitted category is more
than four years prior to intrastate (interstate) deregulation.
ϕi, μj, and δt capture state, industry, and time fixed effects,
respectively.

Columns 2–5 in Table 6 show the results from estimat-
ing Eq. (7) using each of the four different proxies for
innovation by young, private firms (Number of patents,
Number of citations, Number of patents with “high”
citations, and Number of patents with “low” citations) to
generate the “Innovative industry dummy.” Across all four
specifications, we notice that the coefficients γ0 and γ1,

Table 5
Effect of deregulation on innovation in states with fewer small firms.

The following table reports regression results of interactions between bank deregulation and states that had fewer small firms prior to deregulation on
younger, private firm innovation. “Intra” (Inter) is a dummy variable that turns to one the year after the focal state implemented intrastate (interstate)
banking deregulation. “Few small firms” is a dummy set to one if the share of small firms in the focal state was below the mean of the year’s small firm share
the year prior to deregulation, and zero otherwise. This variable is time-invariant within states. “Patents” is the number of distinct patents applied for (and
subsequently granted) by assignees in state s in year t. “Citations” is the number of citation-weighted patents applied for (and subsequently granted) by
assignees in state s in year t. “1Q Cites” (4Q Cites) is the number of patents applied for (and subsequently granted) by assignees in state s in year t that are in
the first (fourth) quartile of year t’s citation distribution. State-clustered robust standard errors are reported in parentheses. n, nn, and nnn indicate
significance better than 10%, 5%, and 1%, respectively.

(1) (2) (3) (4)
Dependent variable Patents Citations 1Q Cites 4Q Cites

Intra −0.284nn −0.355nnn −0.278n −0.374nn

(0.141) (0.135) (0.144) (0.161)

Intra n Few small firms 0.264n 0.233n 0.275 0.285
(0.153) (0.142) (0.168) (0.174)

Inter 0.315nnn 0.325nnn 0.330nnn 0.377nnn

(0.072) (0.069) (0.087) (0.078)

Inter n Few small firms −0.441nnn −0.518nnn −0.360nn −0.556nnn

(0.124) (0.123) (0.142) (0.133)

Year FE ✓ ✓ ✓ ✓
State FE ✓ ✓ ✓ ✓

Observations 1,581 1,581 1,581 1,581
Num. states 51 51 51 51

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774770

Author’s personal copy

which capture the effects after interstate deregulation on
economic growth through innovation by young, private
firms, are positive and statistically significant. This result is
consistent with interstate deregulation having a positive
effect on innovation by young, private firms and, thereby,
fostering economic growth over at least the next eight
years. Further, even though the coefficients of the interac-
tion of intrastate deregulation with the Innovative indus-
try dummy are not significant, they are all uniformly
negative, which is also consistent with intrastate dereg-
ulation having a negative effect on economic growth

because of its negative effect on innovation by young,
private firms. The effects are not statistically significant
at conventional levels possibly because of the fact that
interstate deregulations were passed about two years after
intrastate deregulations, which was perhaps not long
enough for the effects on innovation to translate into
economic growth. The growth effects through innovation
by young, private firms is economically large. Growth due
to interstate deregulation in the innovative industries is
approximately 0.40% greater per annum than other indus-
tries. We also notice that the direct effects of intrastate and

Table 6
Growth effects of banking deregulation through innovation by young, private firms.

The following table explores the growth effects of banking deregulation through innovation by young, private firms. In column 1, we replicate the results
in Jayaratne and Strahan (1996) by estimating the following specification using Ordinary Least Squares (OLS): yit ¼ ϕi þ δt þ β′IntrastateDeregulationit þ εit ,
where yit ¼ gspit=gspi;t−1−1 equals the growth rates in GSP in state i in year t, Deregulationit equals one for states permitting M&A and zero otherwise as in
Jayaratne and Strahan (1996). In columns 2–5, we estimate the following specification using OLS: yijt ¼ ϕi þ μj þ δt ½β−1 � Intrað−4;−1Þit þ β0 � Intrað0;3Þitþ
β1 � Intrað≥4Þit �nInnovative industry dummyj þ ½γ−1 � Interð−4;−1Þit þ γ0 � Interð0;3Þit þ γ1 � Interð≥4Þit �nInnovative industry dummyj þ ½β′−1 � Intrað−4;−1Þitþ
β′0 � Intrað0;3Þit þ β′1 � Intrað≥4Þit � þ ½γ′−1 � Interð−4;−1Þit þ γ′0 � Interð0;3Þit þ γ′1 � Interð≥4Þit � þ εit where β−1 ; β0; β1 ðγ−1 ; γ0 ; γ1Þ together capture the dynamic
effects of intrastate (interstate) banking deregulation on economic growth through innovation by young, private firms. β−1ðγ−1Þ captures the effect of
intrastate (interstate) four years prior to one year prior to deregulation, β0ðγ0Þ captures the effect from the year of intrastate (interstate) deregulation to
three years after, while β1ðγ1Þ captures the effect of intrastate (interstate) deregulation four years after and beyond. The omitted category is more than four
years prior to intrastate (interstate) deregulation. ϕi , μj , and δt capture state, industry, and time fixed effects, respectively. The Innovative industry dummies
in columns 2–5 are set to one for industries that are above the mean patent, citations, 1st quartile citations, and 4th quartile citations, respectively. State-
clustered robust standard errors are reported in parentheses. n, nn, and nnn indicate significance better than 10%, 5%, and 1%, respectively.

(1) (2) (3) (4) (5)
Innovative industry dummy is based on: Patents Citations 1Q Cites 4Q Cites

Intrastate deregulation 1.354nn

(0.342)
Intra(−4,−1) n Innovative industry dummy −0.234 −0.123 −0.118 −0.102

(0.220) (0.234) (0.229) (0.208)

Intra(0,3) n Innovative industry dummy −0.338 −0.315 −0.372 −0.331
(0.226) (0.245) (0.233) (0.210)

Intrað≥4Þ n Innovative industry dummy −0.173 −0.225 −0.202 −0.136
(0.183) (0.190) (0.185) (0.175)

Inter(−4,−1) n Innovative industry dummy 0.127 0.281 0.256 0.269
(0.226) (0.247) (0.237) (0.214)

Inter(0,3) n Innovative industry dummy 0.394n 0.419n 0.400n 0.353n

(0.226) (0.249) (0.230) (0.205)

Interð≥4Þ n Innovative industry dummy 0.334n 0.461nn 0.409nn 0.229
(0.195) (0.205) (0.194) (0.183)

Intra(−4,−1) −0.269 −0.359 −0.363 −0.377n

(0.219) (0.234) (0.229) (0.205)

Intra(0,3) 0.800nnn 0.784nnn 0.833nnn 0.791nnn

(0.227) (0.246) (0.234) (0.210)

Intrað≥4Þ 1.011nnn 1.057nnn 1.036nnn 0.981nnn
(0.198) (0.205) (0.201) (0.190)

Inter(−4,−1) 2.638nnn 2.505nnn 2.527nnn 2.519nn

(0.249) (0.270) (0.260) (0.238)

Inter(0,3) 3.041nnn 3.009nnn 3.029nnn 3.068nnn

(0.242) (0.261) (0.245) (0.226)
Interð≥4Þ 2.812nnn 2.694nnn 2.744nnn 2.894nnn

(0.231) (0.238) (0.231) (0.223)

State FE ✓ ✓ ✓ ✓ ✓
Industry FE ✓ ✓ ✓ ✓
Year FE ✓ ✓ ✓ ✓ ✓

Observations 1,173 20,456 20,456 20,456 20,456
R2 0.469 0.451 0.451 0.451 0.451

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774 771

Author’s personal copy

interstate deregulation on economic growth are positive
and extend through time. Note that the positive estimate
for γ′−1; which captures growth effects four years
prior to interstate deregulation, could be due to the effect
of intrastate deregulation as it preceded interstate
deregulation.

The above tests enable us to conclude that the con-
trasting effects of intrastate and interstate banking dereg-
ulation on innovation by young, private firms led to similar
contrasting effects on economic growth.

4.7. Robustness tests and alternative interpretations

4.7.1. Robustness to definition of young, private firms
Banking deregulation primarily impacted innovation by

young, private firms. So far, we have classified a private
firm as young if it has three or fewer years of patenting
experience in the focal state. In Table 7, we check for the
robustness of our results where we use different cutoffs to
define young and mature firms. We see that the results
continue to hold when the cutoff is set at five years. Not
surprisingly, the results are less statistically significant
when the cutoff is set to ten years.

4.7.2. Alternative interpretations
Having documented a series of strong and robust

results that highlight the relationship between banking
deregulation innovative activity by young, private firms,
we now address possible alternative explanations for our
findings.

Mean reversion: We find that intrastate deregulation
had a negative effect, while interstate deregulation had a
positive effect on innovation by young, private firms. Since
interstate deregulation typically followed intrastate dereg-
ulation with the lag of a few years, a possible alternative
interpretation may be that these results are due to a
combination of: (i) a decline in innovation coinciding with
the passage of interstate deregulation due to other extra-
neous factors and; and (ii) subsequent mean reversion in
innovation with a lag equal to the time lag between
intrastate and interstate deregulations.

First, we should note that “mean reversion” cannot
strictly be assumed to exist; there must be an underlying
economic explanation for the phenomenon. Suppose that
we were to assume that mean reversion mechanically
exists. If the mean reversion were to manifest secularly
across all states, the year dummies would control for such
secular patterns in innovation. In fact, in order for mean
reversion to be a plausible alternative explanation, we
would have to observe differing reversion rates between
the treatment and control groups of states in order to
explain the results from our difference-in-differences
results. Further, mean reversion of this type must satisfy
three additional conditions to account for our results in
their entirety. First, the mean reversion should be present
in young, private firms but not in mature, private firms, or
public firms. Second, the differential rates of mean rever-
sion in the treatment and control groups of states must not
be captured by the state-specific time trends that we
included in our other specifications. Third, given the fact
that we did not observe any effects on innovation prior to
interstate deregulation, the time period within which
mean reversion occurs must equal the average time lag
between intrastate and interstate deregulation. It is quite
implausible that these requirements hold together, which
suggests that our results are robust to this alternative
interpretation.

Creation of the court of appeals of the federal circuit: The
U.S. Court of Appeals of the Federal Circuit (CAFC) was
created by Congress in 1982 with a jurisdiction over
appeals made pertaining to U.S. patent law. Following
the establishment of the court, there was a large surge in
patenting in the U.S. that was commonly ascribed to the
creation of the court although Kortum and Lerner (1999)
attribute the same to other factors such as changes in the
management of research. The spur in patenting activity
also overlaps with the period when banking deregulation
occurred. Our results are, however, unlikely to be driven by
this phenomenon for the following reasons. First, the
creation of CAFC possibly led to a secularly increasing
trend in patenting by U.S. firms. As we find a negative
effect on innovation due to intrastate deregulation and a

Table 7
Robustness tests: different age cutoffs to classify private firms into young and mature.

The following table reports regression results exploring the robustness of age cutoffs in classifying firms as young and mature. “Intra” (“Inter”) is a
dummy variable that turns to one the year after the focal state implemented intrastate (interstate) banking deregulation. “Patents” is the number of distinct
patents applied for (and subsequently granted) by assignees in state s in year t. State-clustered robust standard errors are reported in parentheses.
n, nn, and nnn indicate significance better than 10%, 5%, and 1%, respectively.

Dependent variable: patents

(1) (2) (3) (4)
Firm age ≤5 yrs Firm age45 yrs Firm age ≤10 yrs Firm age410 yrs

Intra −0.277nn 0.452 −0.220n 0.856
(0.116) (0.377) (0.125) (0.572)

Inter 0.177nnn 0.065 0.184nn −0.020
(0.061) (0.124) (0.072) (0.166)

Year FE ✓ ✓ ✓ ✓
State FE ✓ ✓ ✓ ✓

Observations 1,581 1,581 1,581 1,581
Num. states 51 51 51 51

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774772

Author’s personal copy

positive effect due to interstate deregulation, a uniformly
increasing trend in patenting by U.S. firms could not be the
omitted variable that leads to such disparate effects.
Second, if the creation of CAFC accounted for our results,
we should have found statistically significant effects for
innovation by mature, private firms, and public firms,
which is not the case. Therefore, it is unlikely that our
results are driven by the creation of CAFC in 1982.

5. Conclusions

We show that financial sector reform can affect long-
term economic growth by influencing innovation by
young, private firms. Motivated by prior studies in the
structure-conduct-performance, incomplete contracting,
and relationship banking literatures, we develop refutable
hypotheses for the effects of banks’ bargaining/market
power on innovation by young, private firms.

We find that the increase in local market power of
banks after intrastate deregulation had a negative effect on
the innovative activity of young, private firms. Both the
level of innovation (as measured by the number of patents
and the number of citation-weighted patents) and the risk
of innovation (as measured by the number of patents
in the tails of the citation distribution) decreased signifi-
cantly after intrastate banking deregulation. On the other
hand, the decrease in local market power of banks after
interstate deregulation had a positive effect on innovation
by young, private firms. Both the level and risk of innova-
tion increased after interstate banking deregulation.

While intrastate and interstate banking deregulations
affected innovation by young, private firms, neither had
any effect on innovation by mature, private firms, or public
firms. To provide further evidence consistent with the
channels through which deregulation affects innovation,
we show that the deregulations impacted explorative and
product innovation, but not exploitative and process
innovation. We document additional support for the
hypothesized channels through tests that exploit interstate
differences in initial conditions before deregulation.

Our findings suggest that financial development can
benefit economic growth by relaxing financial constraints
and boosting innovation. More importantly, the evidence
suggests that financial development benefits young firms
that are more likely to transform industries with their
technological breakthroughs than large firms that typically
undertake incremental and path-dependent innovation.
The contrasting effects of intrastate and interstate dereg-
ulations on innovation, however, suggest that the nature of
financial sector reform is crucial in unlocking its potential
benefits to the real economy.

References

Acharya, V., Subramanian, K.V., 2009. Bankruptcy codes and innovation.
Review of Financial Studies 22, 4949–4988.

Acharya, V., Baghai, R., Subramanian, K.V. Wrongful discharge laws and
innovation. Review of Financial Studies, http://dx.doi.org/10.1093/rfs/
hht009, forthcoming.

Acs, Z.J., Audretsch, D.B., 1987. An empirical examination of small firm
growth. Economics Letters 25, 363–366.

Acs, Z.J., Audretsch, D.B., 1988. Testing the Schumpeterian hypothesis.
Eastern Economic Journal 14, 129–140.

Acs, Z.J., Audretsch, D.B. (Eds.), 1993. Small Firms and Entrepreneurship.
Cambridge University Press, Cambridge.

Aghion, P., Howitt, P., 1992. A model of growth through creative destruc-
tion. Econometrica 60, 323–351.

Ahuja, G., 2000. Collaboration networks, structural holes, and innovation:
a longitudinal study. Administrative Science Quarterly 45, 425–455.

Akcigit, U., Kerr, W.R., 2011. Growth Through Heterogeneous Innovations.
Unpublished Working Paper, University of Pennsylvania and Harvard
University.

Amore, M., Schneider, C., Zaldokas, A., 2012. Credit Supply and Corporate
Innovations. Unpublished Working Paper, Bocconi University, Copen-
hagen Business School, and Hong Kong University of Science
+Technology.

Angrist, J.D., Pischke, J., 2008. Mostly Harmless Econometrics: An Empiri-
cist’s Companion. Princeton University Press, Princeton, NJ.

Beck, T., Demirgüç-Kunt, A., Laeven, L., Levine, R., 2008. Finance, firm size,
and growth. Journal of Money Credit and Banking 40, 1379–1405.

Beck, T., Demirgüç-Kunt, A., Maksimovic, V., 2004. Bank competition and
access to finance: international evidence. Journal of Money, Credit
and Banking 36, 627–648.

Beck, T., Levine, R., 2004. Stock markets, banks, and growth: panel
evidence. Journal of Banking and Finance 28, 423–442.

Beck, T., Levine, R., Levkov, A., 2010. Big bad banks? The winners and
losers from bank deregulation in the United States. Journal of Finance
65, 1637–1667.

Benfratello, L., Schiantarelli, F., Sembenelli, A., 2008. Banks and innova-
tion: microeconometric evidence on Italian firms. Journal of Financial
Economics 90, 197–217.

Berger, A.N., 2010. Small business lending. In: Berger, A., Molyneux, P.,
Wilson, J. (Eds.), The Oxford Handbook of Banking, Oxford University
Press, NY, pp. 531–549.

Berger, A.N., Udell, G., 1998. The economics of small business finance: the
roles of private equity and debt markets in the financial growth cycle.
Journal of Banking Finance 22, 613–673.

Berger, A.N., Kashyap, A.K., Scalise, J.M., 1995. The transformation of the
U.S. banking industry: what a long strange trip it’s been. Brookings
Papers on Economic Activity 2, 55–201.

Berger, A.N., Saunders, A., Scalise, J., Udell, G.F., 1998. The effects of bank
mergers and acquisitions on small business lending. Journal of
Financial Economics 50, 187–229.

Berger, A.N., Hasan, I., Klapper, L., 2004. Further evidence on the link
between finance and growth: an international analysis of community
banking and economic performance. Journal of Financial Services
Research 25, 169–202.

Berger, A.N., Miller, N., Petersen, M., Rajan, R.G., Stein, J., 2005. Does
function follow organizational form? Evidence from the lending
practices of large and small banks. Journal of Financial Economics 76,
237–269.

Bertrand, M., Duflo, E., Mullainathan, S., 2003. How much should we trust
differences-in-differences estimates? Quarterly Journal of Economics
119, 249–275.

Black, S.E., Strahan, P.E., 2002. Entrepreneurship and bank credit avail-
ability. Journal of Finance 57, 2807–2833.

Bonaccorsi di Patti, E., Dell’Ariccia, G., 2004. Bank competition and firm
creation. Journal of Money, Credit and Banking 36, 225–251.

Cetorelli, N., Gambera, M., 2001. Banking market structure, financial
dependence and growth: international evidence from industry data.
Journal of Finance 56, 617–648.

Cetorelli, N., 2004. Real effects of bank competition. Journal of Money,
Credit and Banking 36, 543–558.

Cetorelli, N., Strahan, P.E., 2006. Finance as a barrier to entry: bank
competition and industry structure in local U.S. markets. Journal of
Finance 61, 437–461.

Chava, S., Chong, X., Nanda, V., 2012. Funding Innovation: The Role of
Lender Expertise and Control Rights. Unpublished Working Paper,
Georgia Institute of Technology.

Chava, S., Oettl, A., Subramanian, A., Subramanian, K.V., 2012. Banking
Deregulation, Bargaining Power, and Innovation. Unpublished Work-
ing Paper, Georgia Institute of Technology, Georgia State University,
and Indian School of Business, Hyderabad.

Chemmanur, T., Tian, X., 2012. Do Anti-Takeover Provisions Spur Corpo-
rate Innovation? Unpublished Working Paper, Boston College and
Indiana University.

Cole, R.A., Goldberg, L.G., White, L.J., 2004. Cookie cutter vs. character: the
micro structure of small business lending by large and small banks.
Journal of Financial and Quantitative Analysis 39, 227–251.

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774 773

Author’s personal copy

Cornaggia, J., Tian, X., Wolfe, B., 2012. Does Banking Competition
Affect Corporate Innovation? Unpublished Working Paper, Indiana
University.

Darby, M.R., Zucker, L.G., 2003. Growing by leaps and inches: creative
destruction, real cost reduction, and inching up. Economic Inquiry 41,
1–19.

Demirgüç-Kunt, A., Maksimovic, V., 1998. Law, finance, and firm growth.
Journal of Finance 53, 2107–2137.

Griliches, Z., 1990. Patent statistics as economic indicators: a survey.
Journal of Economic Literature 28, 1661–1707.

Grossman, S.J., Hart, O.D., 1986. The costs and benefits of ownership: a
theory of vertical and lateral integration. Journal of Political Economy
94, 691–719.

Grossman, G.M., Helpman, E., 1991. Quality ladders in the theory of
growth. Review of Economic Studies 58, 43–61.

Hall, B.H., Jaffe, A.B., Trajtenberg, M., 2001. The NBER Patent Citation Data
File: Lessons, Insights and Methodological Tools. Unpublished Work-
ing Paper No. 8498, National Bureau of Economic Research.

Hart, O., 1995. Firms, Contracts, and Financial Structure. Oxford University
Press, New York.

Hart, O., Moore, J., 1990. Property rights and the nature of the firm.
Journal of Political Economy 98, 1119–1158.

Hausman, J.A., Hall, B.H., Griliches, Z., 1984. Econometric models for count
data with an application to the patents–R&D relationship. Econome-
trica 52, 909–938.

Henderson, R.M., 1993. Underinvestment and incompetence as responses
to radical innovation – Evidence from the photolithographic align-
ment equipment industry. Rand Journal of Economics 24, 248–270.

Hombert, J., Matray, A., 2012. The Real Effects of Hurting Lending
Relationships: Evidence from Banking Deregulation and Innovation.
Unpublished Working Paper, HEC Paris.

Hubbard, G.R., Palia, D., 1995. Executive pay and performance evidence
from the U.S. banking industry. Journal of Financial Economics 39,
105–130.

Jayaratne, J., Strahan, P.E., 1996. The finance-growth nexus: evidence from
bank branch deregulation. Quarterly Journal of Economics 111,
639–670.

Jayaratne, J., Strahan, P.E., 1998. Entry restrictions, industry evolution, and
dynamic efficiency: evidence from commercial banking. Journal of
Law and Economics 41, 239–273.

Karceski, J., Ongena, S., Smith, D.C., 2005. The impact of bank consolida-
tion on commercial borrower welfare. Journal of Finance 60,
2043–2082.

Kerr, W.R., Nanda, R., 2009. Democratizing entry: banking deregulations,
financing constraints, and entrepreneurship. Journal of Financial
Economics 94, 124–149.

King, R.G., Levine, R., 1993a. Finance, entrepreneurship and growth:
theory and evidence. Journal of Monetary Economics 32, 513–542.

King, R.G., Levine, R., 1993b. Finance and growth: Schumpeter might be
right. Quarterly Journal of Economics 108, 717–737.

Kortum, S., Lerner, J., 2000. Assessing the contribution of venture capital
to innovation. RAND Journal of Economics 31, 674–692.

Kroszner, R.S., Strahan, P.E., 1999. What drives deregulation? Economics
and politics of the relaxation of bank branching restrictions. Quar-
terly Journal of Economics 114, 1437–1467.

Levinthal, D.A., March, J.G., 1993. The myopia of learning. Strategic
Management Journal 14, 95–112.

March, J.G., 1991. Exploration and exploitation in organizational learning.
Organization Science 2, 71–87.

Nanda, R., Nicholas, T., 2012. Did Bank Distress Stifle Innovation During
the Great Depression? Unpublished Working Paper, Harvard Business
School.

Pakes, A., Griliches, Z., 1980. Patents and R&D at the firm level: a first
report. Economics Letters 5, 377–381.

Pakes, A., Schankerman, M., 1984. The rate of obsolescence of patents,
research gestation lags, and the private rate of return to research
resources. In: Griliches, Z. (Ed.), R&D, Patents and Productivity,
University of Chicago Press, Chicago, IL, pp. 98–112.

Petersen, M., Rajan, R.G., 1994. The benefits of lending relationships:
evidence from small business data. Journal of Finance 49, 3–37.

Petersen, M., Rajan, R.G., 1995. The effect of credit market competition on
lending relationships. Quarterly Journal of Economics 110, 407–444.

Rajan, Raghuram, G., 1992. Insiders and outsiders: the choice between
informed and arm’s-length debt. Journal of Finance 47, 1367–1400.

Rajan, R.G., Zingales, L., 1998. Financial dependence and growth. Amer-
ican Economic Review 88, 559–586.

Robb, A.M., Robinson, D.T. The capital structure decisions of new firms.
Review of Financial Studies, http://dx.doi.org/10.1093/rfs/hhs072,
forthcoming.

Romer, P., 1990. Endogenous technological change. Journal of Political
Economy 98, 71–102.

Rice, T., Strahan, P.E., 2010. Does credit competition affect small-firm
finance? Journal of Finance 65, 861–889.

Samila, S., Sorenson, O., 2010. Venture capital as a catalyst to commer-
cialization. Research Policy 39, 1348–1360.

Sapra, H., Subramanian, A., Subramanian, K.V. Corporate Governance and
Innovation: Theory and Evidence. Journal of Financial and Quantita-
tive Analysis, forthcoming.

Schumpeter, J., 1942. Capitalism, Socialism, and Democracy. Harper &
Row, New York.

Sørensen, J.B., Stuart, T.E., 2000. Aging, obsolescence, and organizational
innovation. Administrative Science Quarterly 45, 81–112.

Stiroh, K.J., Strahan, P.E., 2006. Competitive dynamics of competition:
evidence from U.S. banking. Journal of Money, Credit and Banking 35,
801–828.

Strahan, P.E., 2002. The Real Effects of U.S. Banking Deregulation.
Unpublished Working Paper 02-39, Center for Financial Institutions,
Wharton School Center for Financial Institutions, University of
Pennsylvania.

Wooldridge, J.M., 1999. Distribution-free estimation of some nonlinear
panel data models. Journal of Econometrics 90, 77–97.

Zucker, L., Darby, M., Brewer, M., 1998. Intellectual human capital and the
birth of U.S. biotechnology enterprises. American Economic Review
88, 290–306.

S. Chava et al. / Journal of Financial Economics 109 (2013) 759–774774