Journal of Regulatory Economics (2019) 55:172–192

https://doi.org/10.1007/s11149-019-09375-y

ORIGINAL PAPER

Entrepreneurial response to interstate regulatory

competition: evidence from a behavioral discrete choice
experiment

Trey Malone1 · Antonios M. Koumpias2 · Per L. Bylund3

Published online: 8 February 2019

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Abstract
Despite a developing literature exploring the relationship between regulation, taxa-
tion and business startups, few studies have utilized artefactual experimental meth-
ods to link the choices made by entrepreneurs to an underlying regulatory frame-
work or tax system. Using information collected from a discrete choice experiment
where 182 small business owners and entrepreneurs made eight start-up decisions,
we describe the effect of state-level government intervention in terms of an entrepre-
neur’s choice to start a business. The design allows the generation of data on entre-
preneurial choice of institutional setting for new business formation, which are diffi-
cult or impossible to observe in natural settings from surveys. We find that over 80%
of entrepreneurs are likely to respond adversely to regulatory and tax legislation
such as mandatory licensing, income taxes, and time to register a business. Results
confirm that, at least in the short run, highly regulated business environments are
less likely to foster entrepreneurial market entry. Additionally, a non-trivial fraction
of entrepreneurs will choose not to start a new business, even in the presence of low
taxation and regulatory burden.

Keywords Entrepreneurship · Experiments · Taxes · Occupational licensing ·

Business registration · Regulatory burden

JEL Classification C9 · D22 · L26

* Trey Malone
[email protected]
1
Department of Agricultural, Food, and Resource Economics, Michigan State University,
East Lansing, USA
2
Department of Social Sciences, University of Michigan-Dearborn, Dearborn, USA
3
School of Entrepreneurship, Oklahoma State University, Stillwater, USA

1Vol:.(1234567890)
3
Entrepreneurial response to interstate regulatory… 173

1 Introduction

Economists have long known that heavy regulatory burdens distort business deci-
sion-making (Stigler 1971) as the total consequences of legislation (such as taxation,
occupational licensing, and new business regulation) were acknowledged as early as
Bastiat (1848, pp. 6)1:
In the economic sphere an act, a habit, an institution, a law produces not only
one effect, but a series of effects. Of these effects, the first alone is immediate;
it appears simultaneously with its cause; it is seen. The other effects emerge
only subsequently; they are not seen; we are fortunate if we foresee them.
This is an important matter because of the fierce subnational fiscal competition to
attract new firms (Agrawal et al. 2015).2 Interstate regulatory competition has the
potential to lead to undesirable outcomes (Partridge and Olfert 2011), making it of
seminal importance for state governments to understand how entrepreneurs decide
between two competing locations. As states compete to induce new businesses to
locate in their jurisdiction, fiscal gains from new business formation may be miti-
gated by generous tax breaks and non-tax incentives offered. In some cases, employ-
ers with high bargaining power such as Amazon can secure deals that may largely
wipe the net fiscal gains from new business formation.3
Many studies have evaluated the influence of government regulations on eco-
nomic decision-making (Carruthers and Lamoreaux 2016). For example, Aidis et al.
(2012) use the Global Entrepreneurship Monitor to display a robust negative asso-
ciation between entrepreneurial entry through the creation of new business start-ups,
and the size of the government sector across 47 countries. Furthermore, Sobel et al.
(2007) find that politically granted protectionist policies stifle entrepreneurship in
developed economies. Using a cross-section of OECD countries, they estimate a
negative and significant effect of increased administrative burden for start-ups on
productive, market, and non-political entrepreneurship. Similarly, Klapper et al.
(2006) identify the role of European market entry regulations in slowing the creation

1
Our study of firm location choice is an analysis of institutional, not spatial characteristics of location.
Regulatory burdens can be contrasted with the spatial ones which might be explained via Ricardian com-
parative advantage.
2
Chirinko and Wilson (2017) attribute the decline in state corporate tax rates over time to synchronous
responses among states to common shocks rather than strategic tax competition. Using panel data set
covering the 48 contiguous U.S. states for the period 1965 to 2006, they find a negative response of home
state to foreign tax policy implying that tax competition may increase the provision of local public goods.
3
Such financial incentives are often increasing in the size of the investment. In the highly-publicized
case of Amazon’s HQ2, cities across the U.S. engaged in fierce competition in the form of tax breaks
and incentives to attract a single business promising up to 50,000 jobs. The chosen sites in New York
City’s Long Island City, NY and in Arlington, VA “provided total incentives worth $2.8 billion” when
other sites in Maryland and Newark, New Jersey offered $8.5 and $7 billion, respectively. Such incen-
tives imply a substantial cash transfer to new businesses: “The bulk of Virginia’s tax incentives come
in the form of a cash grant of $22,000 for each job added over the next 12 years, as long as the average
annual wage of those jobs is at least $150,000—a total of $550 million if Amazon adds the 25,000 jobs it
says it will.” https​://www.cnn.com/2018/11/13/busin​ess/amazo​n-hq2-subsi​dies/index​.html.

13
174 T. Malone et al.

of new limited-liability firms, causing start-ups to be larger when young and to grow
more slowly over time. Da Rin et al. (2011) use a panel of 17 European countries
from 1997 through 2004 to find a significant negative and concave effect of corpo-
rate income taxation on firm entry rates. Bruce and Mohsin (2006) show that the
federal top corporate income tax rate and payroll tax rate have a large negative effect
on entrepreneurship and self-employment rates. Bailey and Thomas (2017) combine
data from the Statistics of US Businesses and RegData, an index that measures the
intensity of regulation in 215 separate industries in the US, to find that fewer new
firms were created in relatively more regulated industries from 1998 to 2011.
Despite this research, the extant literature on the relationship between regulation
and entrepreneurship has been limited by the availability of data. Data constraints
make observing the total effect of regulations difficult to uncover. Prior research
on the impact of the regulatory burden on business location choice has typically
employed readily available secondary data such as surveys or administrative data,
which are limiting in their ability to observe people who choose not to start a busi-
ness; thus, suffering from a form of survivorship bias in entrepreneurship (Cassar
2004). As a result, studies have struggled to accurately identify the characteristics
of the entrepreneurs who decided against starting a business. Therefore, prior esti-
mates in the empirical literature might be understating the effect of taxation and
regulation on entrepreneurship by solely relying on successful business formation.
Furthermore, rather than being able to estimate utility functions and make predic-
tions soundly rooted in economic demand theory, most studies of entrepreneurial
decision-making have been constrained to drawing inferences about entrepreneurial
decision-making without directly observing the choice to start a business.
As such, this article utilizes a novel technique to assess how state regulation such
as licensing requirements, business registration processing times, and corporate tax
rates influence whether and where new businesses choose to locate. Initially, we
implement a discrete choice experiment (DCE) that elicits entrepreneurs’ prefer-
ences over new business formation in response to changes in the state regulatory
and tax burden).4 Stated simply, the DCE is a method firmly rooted in random util-
ity theory that allows researchers to test hypotheses against demand theory while
avoiding common econometric issues such as endogeneity and confounded variables
(Hensher et al. 2015). Using a “branded” orthogonal fractional factorial experimen-
tal design, we consider two localities separated by a state border as potential new
business formation locations; namely, Kansas City, Missouri (KCMO), and Kansas
City, Kansas (KCKS) as well as the option of no new business formation anywhere.
The DCE allows us to avoid some of key confounds inherent in many econometric
analyses. As an example, consider the decision process for tax rates and the regula-
tory constraints imposed on entrepreneurs. These are often jointly determined by
government expenditures and the incumbent political party. With a DCE, we can
avoid any such confounds as we can exogenously determine the characteristics of
the marketplace itself. It is important to note that we are posing an institutional

4
Following the typology of field experiments in Harrison and List (2004), the design could alternatively
be termed an artefactual field experiment.

13
Entrepreneurial response to interstate regulatory… 175

question about the firm location choice, not a spatial one. By conducting the analysis
within the same metropolitan area, we are abstracting away from discrepancies in
firm location attractiveness due to spatial differences and solely examine those due
to differences in the regulatory framework between localities. The inclusion of the
alternative to not start a business to the choice set of the entrepreneur-respondent is
what enables this study to overcome survivorship bias. By assuming that the num-
ber of entrepreneurial startups is a function of the demand of entrepreneurs to start
a new business, we can estimate entrepreneurs’ indirect utility at each one of the
three choices via a latent class logit model. Next, having generated primary data
from experimental methods, we derive two policy-relevant statistics that constitute
our main findings. First, we estimate the relative probabilities that an entrepreneur
would start a business in each of the two states along with the probability that an
entrepreneur will not start a business at all under the specified regulatory conditions.
This statistic measures the isolated effect of each policy on entrepreneurship sepa-
rately which allows us to assess the relative importance of the regulatory framework
to the tax system in entrepreneurs’ business formation decision. Second, we evaluate
the impact of changes in regulation on the likelihood the entrepreneur will choose
to start her business in each state or not start a business at all by computing point
elasticities of the probability of locating in state with respect to a marginal change in
one of the three choice attributes (the two regulatory and the one tax variable). Since
these are both “own-price” and “cross-price” in nature, we can infer the impact
of policy changes not only within the policy’s jurisdiction but across borders, too.
Thus, our second statistic is capturing the feedback loop effects of interstate regula-
tory and tax competition.
The rest of the paper is organized as follows. Section 2 discusses the theoretical
background of our experiment, and justifies our rationale for including occupational
licensing, corporate taxes, and registration-processing time in our DCE. We develop
and analyze a DCE to test how entrepreneurs’ location choice responds to changes in
government regulatory and tax burden. This introduces a novel approach in generat-
ing credible estimates about key questions in entrepreneurship such as the effect of
barriers to entry and taxation to new business formation. It should be noted that even
though our DCE estimates are based on hypothetical choices, our findings may still
suffer from survivorship bias as the surveyed population is still comprised of exist-
ing entrepreneurs. In Sect. 3, we explain the design of the DCE along with the data
and methodology we employ. Section 4 presents the results, which suggest that the
majority of entrepreneurs respond to changes in public policy in a manner consistent
with economic theory. Section 5 concludes with a discussion of the limitations of
our study as well as its implications for future research and policy.

2 Theoretical background

Countless attributes are likely to influence where an entrepreneur chooses to locate

her business. An entrepreneur may be interested in positioning close to the relevant
quality labor (Melo et al. 2009), target costumers (Ellison et al. 2010), or to ensure
proximity to other businesses (Acs and Varga 2005; Delgado et al. 2010). At the

13
176 T. Malone et al.

same time, entrepreneurs exhibit “local bias in entrepreneurship”, as they often start
their business at the location they were born (Michelacci and Silva 2007). Most per-
tinent to this study is the characteristic which state governments most directly influ-
ence, and that is the development of a regulatory environment that leads to high ease
of doing business.
Many state government interventions are likely to influence a startup’s loca-
tion decision, and the notion that entrepreneurs sort into localities based on how
conducive to business the underlying regulatory framework is, has been very well
documented. Carruthers and Lamoreaux (2016) provide an extensive review of
jurisdictional competition on regulatory standards across labor regulation, environ-
mental protection, corporate governance, and finance. They report firm migration
across state or county borders in the presence of differences in the cost and benefits
of regulation but rare incidence of races to the bottom.5 We choose to abstract away
from “pro-business” government interventions such as state and local startup reten-
tion programs since the latter are most effective among stagnant firms that do not
seek growth capital (Zhao 2013). Instead, we select three interventions whose prior
literature might be complemented by this novel experimental method.
First, we include a state corporate tax rate, as it has been well documented that
higher tax rates discourage entrepreneurial startups (Gentry and Hubbard 2005;
Curtis and Decker 2018; Serrato and Zidar 2016). Theoretically, corporate taxa-
tion has an adverse effect on private investment (Harberger 1962). However, Hau-
fler et al. (2014) warn that tax systems that systematically favor entrepreneurial
entry can lead to welfare losses due to inefficient quality choices. Empirically, it
has been extensively shown that entrepreneurial entry will be hampered by taxa-
tion; Henrekson and Sanandaji (2011) provide an extensive review. Djankov et al.
(2010) construct a novel dataset of comparable corporate income tax rates and offi-
cial entrepreneurial entry rates, defined as the number of newly registered limited
liability corporations as a percentage of the stock of such firms for 62 countries over
the 2000–2004 period. They find that a 10-percentage point increase in the first-
year effective corporate tax rate reduces the official entry rate by 1.4 percentage
points for an average entry rate of 8%. At the state level, Serrato and Zidar (2016)
use matched firm-establishment data across 490 U.S. public-use micro-data areas
(PUMAs), variation in state corporate tax rates, and apportionment rules to measure
the elasticity of local economic activity to local business tax changes. They estimate
that a 1% reduction in local business taxes leads to a 3 to 4% increase in the number
of local establishments over the course of 10 years. At the local level within the state
of New York, Mast (2017) finds that competition for new firms between city govern-
ments leads to increases in the tax breaks given.
Second, we consider occupational licensing as a requirement to enter the indus-
try. Approximately 30% of the workforce is required to hold some kind of license,
and the effect of occupational licensing is likely to be heterogeneous across states
(Kleiner and Krueger 2010, 2013). Previous literature suggests that the effect of

5
See Carruthers and Lamoreaux (2016) for a review of the literature on interstate regulatory competi-
tion. Notably, their review does not include any experiments.

13
Entrepreneurial response to interstate regulatory… 177

occupational licensing on the likelihood an entrepreneur will start a business in that

industry is somewhat ambiguous; occupational licensing has the potential to raise
wages by 10% to 20% (Kawaguchi et al. 2014; Kleiner and Krueger 2013; Kleiner
and Vorotnikov 2017; Pizzola and Tabarrok 2017). These higher wages might actu-
ally attract would-be entrepreneurs to a state (Redbird 2017). Alternatively, occupa-
tional licensing has the potential to reduce the supply of quality producers (Peter-
son et al. 2014; Sass 2015). Furthermore, some studies have suggested that these
licenses make it less likely for entrepreneurs to start a business (Thornton and Tim-
mons 2013).
Finally, we include registration-processing time (in days). Empirical work sug-
gests that changes in registration processing can alter the rate of new business start-
ups (De Soto 1989; Djankov et al. 2002). This metric is an integral component of
the World Bank’s annual, influential “Doing Business” report and measures of bar-
riers to entry (Teague 2016). The former has become a leading indicator of the ease
of doing business for academics, policymakers, and the media with coverage of 190
countries as well as selected cities at the subnational level as of 2018. For a review
of the literature employing the registration-processing time (in days) metric in the
Doing Business report, see Besley (2015).

3 Methods

Experimental methods have become increasingly popular in the entrepreneurship

literature (e.g. Arentz et al. 2013; Jiang and Capra 2016), and the DCE represents
a somewhat novel approach for solving economic development and entrepreneurial
questions that were once previously unanswerable (Malone and Lusk 2017, 2018a).6
The DCE has become a key empirical tool used to identify decision-making patterns
in fields such as agricultural economics, transportation, environmental economics,
and health economics. These experiments differ from traditional conjoint analyses
because the DCE is consistent with economic theory, and is more suitable for policy
analysis. As explained by Louviere et al. (2010, pp. 67), “[Conjoint analysis] lacks
a sound, theoretical relationship with real market choice behavior(s), which serves
to reinforce the ad hoc, predominantly statistical and methodical nature of CA
research and practice.” By contrast, the modern DCE builds from Lancaster (1966)
and McFadden (1974). These early studies described a framework consistent with
economic theory that allowed individuals to choose between multiple options in a
manner consistent with demand theory.
Discrete choice experiments represent a valuable complement to secondary data
techniques while maintaining their external validity (Brooks and Lusk 2010; Chang
et al. 2009). Previous studies have shown that preferences measured via “stated pref-
erence” choice experiments are consistent with those inferred from revealed prefer-
ence data (Hensher et al. 1998; Louviere et al. 2000). While there are concerns about
so-called “hypothetical bias,” previous research has shown that such bias is less of

6
See Hsu et al. (2017) for a thorough description of experiments in entrepreneurship research.

13
178 T. Malone et al.

a problem when estimating marginal changes (Carlsson and Martinsson 2001; Lusk
and Schroeder 2004).
The DCE can complement the extant literature on the effect of regulations on
entrepreneurship in several ways. Choice experiments provide increased control by
avoiding potential context-specific confounds. As discussed more below, by design-
ing experiments where attributes are uncorrelated with one another, it also avoids
endogeneity concerns within the econometric model. If we assume that an entre-
preneur’s choice to start a business is determined by her demand to start a business,
that decision can be thought of as composed of the entrepreneur’s idiosyncratic pref-
erences and beliefs (Foss and Klein 2012; Knight 1921; Mises 1949). The startup
decision is thus a matter of how, where, and whether to start a business, rather than
an automatic response to known cost structures (Coase 1937; Williamson 1975,
1985, 1996). As such, choice experiments allow the researcher to identify individ-
ual-specific information more clearly than secondary data. From a new institutional
economics literature standpoint, this DCE is, essentially, a comparative institutional
analysis in experimental form (Coase 1992).

3.1 DCE design

Our dataset comes from a DCE conducted on entrepreneurs that elicits how they
evaluate tradeoffs in their decision-making process. This study utilizes a main effects
orthogonal fractional factorial design where each choice attribute is uncorrelated
with the others, thereby removing any endogeneity concerns across decisions. For
example, it is reasonable to believe that policymakers make a simultaneous, conjoint
decision when jointly determining tax rates and the regulatory framework. By utiliz-
ing an experimental design where no attribute is correlated, we can guarantee that
there are no such endogenously determined parameter estimates. The key drawback
to this structure is that it assumes all attributes are completely orthogonal (or unre-
lated) with one another as well, which means that including interactions between
attributes is not advised. For our purposes, we focus on two choice options, each
with three attributes and two attribute levels. As such, we used the FACTEX proce-
dure in ­SAS® to determine how many questions to ask and what attribute levels to
include for each choice question.7 While the full design would require each partici-
pant to answer (2 × 3)2 = 36 questions, the orthogonal fractional factorial design only
requires each participant to answer eight questions.
Before collecting the data for this study, we first conducted a pilot test of 28 uni-
versity faculty and graduate students to calibrate and test the quality of the survey
instrument. The relevant portions were structured as follows. Prior to the choice
questions, the entrepreneurs were given the following prompt:
Imagine that you are considering to start a small business in Kansas City and
that you are trying to decide on what side of the state line to locate. Your busi-

7
While ­SAS® provides additional procedures for more complicated experimental designs, other pro-
grams such as Ngene® are also commonly employed for the most complex designs.

13
Entrepreneurial response to interstate regulatory… 179

Fig. 1  Example of a choice question

ness model and market access will be the same regardless of geographic loca-
tion. Assuming all other relevant characteristics are the same, please choose
where you would choose to locate based on the specific information provided
regarding licensing requirement, the tax rate, and estimated processing time to
register your business.
We chose to provide participants the option of Kansas City, as there are relatively
few differences in access and quality of life within the city. Furthermore, popular
press outlets have highlighted the ongoing controversies surrounding state gov-
ernment economic development incentive programs (Planet Money 2016; Vaupel
2017). Entrepreneurs were presented with the DCE questions. These choice ques-
tions followed a “branded” design where each of the options represented a state
that had a different regulatory framework. They were then presented eight discrete
choice question, which stated:
Please select which location you would choose to start your small business.
For each question, the entrepreneur could choose one of three options: Kansas,
Missouri, or a “none” option, that simply stated, “I would not start a business in
these conditions.” The question and choice option orders were randomized to elimi-
nate any concerns about order effects. Figure 1 displays an example of the choice
question.
|For this choice experiment, we included the three state government attributes
highlighted earlier, as they are likely to influence an entrepreneur’s decision to start
a business. Immediately following the eight choice questions, we asked three ques-
tions regarding the business environment in each of the two states. Recent research
has suggested that political uncertainty is likely to influence new business startups
(Bylund and McCaffrey 2017). As such, we asked participants to:
Consider the political climate of the two states Kansas and Missouri; how
certain or uncertain are you about each state’s continued business friendliness?
Choose on a scale from −5 (very uncertain) to +5 (very certain).

13
180 T. Malone et al.

Other studies have indicated that an entrepreneur’s perceptions of government inter-

vention is an important determinant of their decision-making (Teague 2016). To be
sure that perceptions of Kansas and Missouri are not confounding the entrepreneur’s
choice in the experiment, we then asked participants a question related to likely reg-
ulatory barriers to entry:
Consider how difficult political/regulatory measures make it to start and run
a business (so-called regulatory barriers to entry). How do you perceive the
regulatory barriers to entry for the following states, where −5 is very low and
+5 is very high?
The phrase “barriers to entry” is likely to be difficult to understand for non-academ-
ics. To make sure we were not picking up alternative barriers to entry such as state-
specific access to capital or proximity to suppliers, we also asked about “natural”
barriers to entry. Because these two variables are likely to be correlated, we simply
compared means of the two groups to test the understanding of the participants. This
question stated:
Consider how difficult it is to find a good location, to get access to required
resources, etc. (so-called natural barriers to entry). How do you perceive the
natural barriers to entry for the following states, where −5 is very low and +5
is very high?
We expect less negative association of “natural” barriers to market entry, since these
should reflect factors such as input costs that are very similar in KCKS relative to
KCMO given the geographical proximity. By virtue of separation between “natural”
and “regulatory” barriers, we ensure that the generated responses permit compari-
sons between KCKS and KCMO on an institutional basis, not a spatial one. Fol-
lowing the perceptions questions, we asked a series of questions about the entrepre-
neur’s actual business and individual demographics (Fig. 2).

3.2 Empirical model

When entrepreneurs choose what business they want to start, they reveal which mar-
ket characteristics they value. For our purposes, we specify entrepreneur i’s utility
of selecting state j, j = Kansas, Missouri, in choice option s, s = starting business in
KSKS, or KSMO, or do not start a business, as:
Uijs = 𝛽j + 𝛼1 ⋅ Licensej + 𝛼2 ⋅ Taxj + 𝛼3 ⋅ Registerj + 𝜀ijs (1)
where the alternative-specific constants βj indicate the utility of state j not explained
by the perceptions or preferences defined in the model, Licensej is the licens-
ing requirement of state j, Taxj is corporate income tax for state j, Registerj is the
registration processing time for state j, εijs is the unobserved portion of the utility
function, and all αh represent the marginal utilities associated with each variable
h = 1,2,3. Note that the specification of this model includes an error term, which
makes this equation random. As such, anything not included in the utility function is
captured by the error term.

13
Entrepreneurial response to interstate regulatory… 181

Institutional 10% 10% 80%

Neoclassical 36% 59% 4%

0% 20% 40% 60% 80% 100%

Missouri Kansas None

Fig. 2  Probability each type of entrepreneur would choose to start a business in Kansas City, Missouri or
Kansas City, Kansas relative to not starting a business at all

The model most commonly used is the traditional multinomial logit model (MNL),
where the dependent variable is the log of the odds ratio between choice alternative j
and some predetermined baseline decision (in our case, this baseline is “none”). The
estimated log odds ratio is commonly interpreted as participant i’s indirect utility of
choosing state j (Vij), or:
Vij = 𝛽j + 𝛼1 ⋅ Licensej + 𝛼2 ⋅ Taxj + 𝛼3 ⋅ Registerij , (2)
which reports the indirect utility of state j relative to not starting a business. For
this study, we conduct our statistical analysis using N ­ LOGIT®, which is particularly
powerful for estimating limited dependent variable models. It is likely that entrepre-
neurs will make decisions based on varying decision processes. One useful expan-
sion of the MNL model is the latent class model, which assumes there are underly-
ing, latent characteristics that are likely to make decision-makers respond differently
to the choice questions. Without imposing any strong assumptions on the model
structure, these latent class logit models effectively separate the data based on the
underlying choice structure of the data. Instead of estimating a single indirect utility
function (Eq. 2), these models use an endogenous sorting mechanism to estimate
multiple indirect utility functions.
Of primary interest to this study are the relative probabilities that an entrepreneur
would start a business in each of the states, and the effect of changes in the specified
variables on the likelihood the participant will choose to start that business. Empiri-
cally, the probability entrepreneur i chooses state j is:

eVij
Prob(j is chosen) = ∑J (3)
k=1 eVij

13
182 T. Malone et al.

where Vij is the systematic portion of the utility function determined by the govern-
mental attributes of state j, individual-specific characteristics, and individual-spe-
cific perceptions. We also calculate elasticities for each of the choice attributes.

4 Results

Data were collected in March 2017, by ­SSI®, who maintain an online survey panel
of entrepreneurs, small business owners, and franchisees. Where prior research has
focused on a specific industry such as technology-based start-ups (van Rijnsoever
and Cerutti 2017), we opted to focus on the decision to start a business more broadly.
Each participant was paid approximately $20 per completed survey, and the aver-
age time to complete the survey was 15 min and 37 s. Table 1 displays the sample
demographics. Nearly half of the small business owners in our sample were female,
and 85.2% of our sample identified as white/Caucasian. Our sample was politically
well-balanced, and 30.8% considered themselves moderate or middle of the road.
Half of our sample had a 4-year college degree or higher, and more than two-thirds
of the sample were between 25 and 64 years of age. Approximately two-thirds of our
sample also earn between $30,000 and $89,999 in annual income.
Table 2 reports the business demographics for every entrepreneur’s business.
Over half of the small business owners in our sample had been in operation for
fewer than 5 years, and 69.8% of our sample founded their company. The majority
reported a greater than 10% profit, and saw this profit margin to be as expected for
their own expectations and for their own industry. The average participant in our
sample employed 24.30 workers, but business sizes varied between one and 100
employees.
Table 3 displays the correlation matrix of the perceptions variables along with
the mean values for each perception. All questions were significantly correlated with
one another, although the barriers to entry questions were more correlated with each
other than were the political uncertainty questions. On average, participants consid-
ered the political climate in Kansas to be more likely to continue being business
friendly than in Missouri. Additionally, they considered the natural barriers to entry
(difficulty to find a good location, to get access to required resources, etc.) to be
higher in Missouri than in Kansas, but there was very little difference between opin-
ions regarding regulatory barriers to entry (how difficult political/regulatory meas-
ures make it to start and run a business).
Exploratory factor analysis confirmed that there are two underlying latent
classes in the perceptions data. We define small business owners as “neoclassi-
cal” when they prefer to avoid additional government constraints on their busi-
ness activity and “institutional” when they do not respond substantially to regula-
tory changes due to their indifference in their decision to start a business to the
three government interventions in our experiment. This distinction follows van
Rijnsoever and Cerutti (2017), who distinguish tech-based entrepreneurial deci-
sions into behavioral, neoclassical, and institutional classes. The factor associ-
ated with political uncertainty was still connected to the factor associated with
barriers to entry (Table 4). The barriers to entry factor explained 49.7% of the

13
Entrepreneurial response to interstate regulatory… 183

Table 1  Sample demographics

Demographic Category levels Percentage

Female 48.4
Race White/Caucasian 85.2
African American 6.6
Hispanic 4.4
Asian 2.7
Other 1.1
Politics Extremely liberal 12.1
Slightly liberal 19.2
Moderate or middle of the road 30.8
Slightly conservative 24.2
Extremely conservative 9.3
I don’t know or neither liberal or conservative 4.4
Education High school/GED 13.7
Some college 20.9
2-year college degree 11.0
4-year college degree 32.4
Master’s degree 15.4
Doctoral degree 3.3
Professional degree 0.5
Age 18 to 24 years 3.8
25 to 34 years 23.6
35 to 44 years 22.5
55 to 64 years 22.0
65 to 74 years 19.2
75 to 84 years 8.2
84 years or older 0.5
Annual income Less than $29,999 8.8
$30,000 to $59,999 23.6
$60,000 to $89,999 22.0
$90,000 to $119,999 14.8
$120,000 to $149,999 14.8
$150,000 to $179,999 5.5
More than $180,000 10.4
Number of participants 182

Presented as a percent of the survey sample

variation in the perceptions data, while the political uncertainty factor explained
20.3% of the variation in the perceptions data. The results of these analyses sug-
gest that the perceptions of Missouri and Kansas did not vary substantially within
our sample of entrepreneurs and small business owners.

13
184 T. Malone et al.

Table 2  Sample business characteristics

Demographic Category levels Percentage

Years in operation Less than 1 year 14.3

Between 1 and 5 years 38.5
More than 5 years but fewer than 15.4
10 years
10 years or more 31.9
Founder 69.8
Profitability Loss 7.1
Approximately break-even 13.7
0–5% profit 15.4
5–10% profit 25.8
Greater than 10% profit 37.9
Perception of profitability Below expectation 17.0
As expected 65.9
Above expectation 17.0
Compared to the industry profitability Below standard 13.2
Commensurate 63.7
Above standard 23.1
Prior experience? As a founder 37.9
As a manager 26.9
No 35.2
Number of workers employed 24.30 (32.02)
Number of participants 182

Presented as a percent of the total sample

Table 5 presents the latent class logit models along with the traditional MNL
parameter estimates. Because each participant answered eight discrete choice
questions, we base our analysis on 8 × 182 = 1456 individual choice observations.
The Akaike Information Criterion (AIC) for the latent class logit models was
substantially lower than the AIC for the traditional MNL model, which suggests
that there are indeed two different types of small business owners in our sam-
ple. Based on a lower AIC, the latent class logit models’ estimates serve as our
results.
Estimates from the latent class logit models suggest that 83.9% of the small
business owners in our sample would be regarded “neoclassical”. responded the
way economic theory might suggest. That is, their market entry decision was neg-
atively affected by tax and regulatory burdens. Specifically, they were averse to
starting a business that required a license to operate in their industry. The imposi-
tion of occupational licensing requirements in Missouri or Kansas decreases the
likelihood that a neoclassical entrepreneur will start a business by 1.41% rela-
tive to a regulatory environment that does not feature any licensing requirements
for their type of business. This class of participants also avoided higher income

13
 Table 3  Correlation matrix and means of perception variables
Political uncer- Political uncer- Natural barriers to Natural barriers to Regulatory barriers to Regulatory bar-
tainty in Missouri tainty in Kansas entry in Missouri entry in Kansas entry in Missouri riers to entry in
Kansas

Political uncertainty in Kansas 0.563*

Entrepreneurial response to interstate regulatory…

Natural BTE in Missouri 0.235* 0.209*

Natural BTE in Kansas 0.254* 0.231* 0.469*
Regulatory BTE in Missouri 0.244* 0.299* 0.595* 0.523*
Regulatory BTE in Kansas 0.207* 0.301* 0.482* 0.661* 0.543*
Variable Means 0.835 (2.354) 0.945 (2.447) 0.582 (2.363) 0.495 (2.278) 0.709 (2.419) 0.725 (2.322)

An asterisk denotes statistical significance at the α = 0.05 level

Numbers in parentheses are standard deviations
185

13
186 T. Malone et al.

Table 4  Factor patterns of perception variables calculated via exploratory factor analysis
Barriers to entry Political uncertainty

Political uncertainty in Missouri 0.525 0.714

Political uncertainty in Kansas 0.556 0.685
Natural barriers to entry in Missouri 0.735 − 0.257
Natural barriers to entry in Kansas 0.778 − 0.259
Regulatory barriers to entry in Missouri 0.791 − 0.210
Regulatory barriers to entry in Kansas 0.793 − 0.251

The first latent factor (barriers to entry) explained 49.7% of the variation in the perceptions, and the sec-
ond latent factor (political uncertainty) explained 20.3% of the variation in the perceptions data

Table 5  Discrete choice models

Neoclassical Institutional MNL

Alternative specific constants

Missouri 3.377* (0.176) − 1.568* (0.429) 1.898* (0.115)
Kansas 3.625* (0.175) − 1.509* (0.425) 2.146* (0.114)
Choice attributes
Licensing − 0.431* (0.091) − 0.498 (0.314) − 0.397* (0.079)
Corporate State Income Tax − 0.109* (0.008) 0.001 (0.028) − 0.087* (0.007)
Registration − 0.017* (0.002) − 0.006 (0.007) − 0.014* (0.002)
Probability of Class 0.839* (0.028) 0.161* (0.028) N/A
AIC 2316.5 2747.7

Estimation based 1456 individual choices made by 182 entrepreneurs and small business owners
An asterisk denotes statistical significance at the α = 0.05 level. Alternative specific constants indicate the
mean baseline utility level associated with starting a business in Missouri or Kansas relative to not start-
ing a business at all. Numbers in parentheses are standard errors

taxes, as well as lengthy registration processes for their business. These entrepre-
neurs decreased the probability they would start a business by 0.45% if both state
governments were to raise corporate income tax rates by a full percentage point.8
Finally, a 1-day increase in business registration time leads to a 0.07% decrease in
the likelihood an entrepreneur will start a business.9
However, the second class of small business owners in our sample did not signifi-
cantly adjust their small business decisions to marginal changes in licensing, busi-
ness registration, or corporate income tax rates. The effect of licensing on firm entry
is large and negative but not in a statistically significant way, whereas the effect
of taxation and business registration times is null and centered on zero. This class
comprised of 16.1% of our sample. In both classes of our entrepreneurs, however,

8
Estimated relative to a baseline of 4% in Kansas and 6.25% in Missouri.
9
Estimated relative to 5.6 days, as it is the number of registration days reported by Teague (2016).

13
Entrepreneurial response to interstate regulatory… 187

Table 6  Elasticity with respect to a 1% change in the corporate tax rate in the state listed in the row on
the probability of selecting the column state
Missouri Kansas None

Δ (Missouri) − 0.401 0.135 0.135

Δ (Kansas) 0.182 − 0.354 0.182

occupational licensing requirements clearly stand out as the most distortionary bar-
rier to entrepreneurship. The introduction of state corporate income taxation also
has a sizable negative impact on new business formation, but business registration
times seem the least distortionary barrier to firm market entry of the three.
The results of the MNL model confirm our baseline results. All estimated coef-
ficients are negative and statistically significant. By and large, our estimates using
the MNL model could be interpreted as a weighted average of the findings based on
the latent class logit models in terms of magnitude. That is not surprising since the
MNL model does not separate the sample into two classes of entrepreneurs. Licens-
ing remains the most important barrier to entry, taxation is still important, and busi-
ness registration times have a small, yet regular negative and significant effect on
firm entry.
Table 6 above shows the responsiveness of entrepreneurs’ decisions to start a new
business in Missouri, Kansas, or not starting a new business at all following a 1%
increase in the state corporate income tax rate in Missouri or Kansas. As predicted
by economic theory, we derive negative elasticities of entrepreneurial entry with
respect to in-state tax changes and positive “cross-price” elasticities of entrepreneur-
ial entry, with respect to out-of-state tax changes.10 Specifically, a one percentage
point increase in the state corporate income tax rate in Missouri will cause a reduc-
tion in entrepreneurial entry in Missouri by 41%, and an increase in entrepreneurial
entry in Kansas or no business formation anywhere by 13.5%. The negative effect of
an increase in state corporate taxation in Kansas on firm entry is slightly less pro-
nounced. A one percentage point increase in the state corporate income tax rate in
Kansas will cause a reduction in entrepreneurial entry in Kansas by 35.4%, and an
increase in entrepreneurial entry in Missouri or no business formation anywhere by
18.2%.
Table 7 presents our estimates of entrepreneurs’ elasticity of market entry with
respect to marginal changes in the number of business registration days in Missouri
or Kansas. A 1 day increase in the number of days required to register a business in
Missouri will cause a reduction in entrepreneurial entry in that state by 0.337%, and
an increase in entrepreneurial entry in Kansas or no business formation anywhere
by 0.150%. The negative effect of an increase in lengthier business registration days
in Kansas on firm entry slightly less pronounced. A 1 day increase in the business

10
The elasticities we computed are arc elasticities or mid-point elasticities, where we are using the aver-
age of the pre- and post-policy change values as opposed to a single point along entrepreneurs’ choice
set..

13
188 T. Malone et al.

Table 7  Elasticity with respect to a 1% increase in the number of days required to register a business in
the state listed in the row on the probability of selecting the column state
Missouri Kansas None

Δ (Missouri) − 0.337 0.150 0.150

Δ (Kansas) 0.198 − 0.289 0.198

registration times in Kansas will cause a reduction in entrepreneurial entry in Kansas

by 0.289%, and an increase in entrepreneurial entry in Missouri or business forma-
tion anywhere by 0.198%. In line with our discrete choice models’ results, registra-
tion times are slightly less distortive for market entry decisions than state corporate
income tax rates. Moreover, we find that entrepreneurial entry in Kansas is slightly
less sensitive to changes in the state corporate income tax rate and business registra-
tion times relative to Missouri. This location preference could be attributed to under-
lying factors that were unidentified in our design.

5 Conclusion

This study uses the role of state government intervention to highlight the value of
incorporating experimental methods into the entrepreneurship and economic devel-
opment literature. Our hypothetical choice experiment shows that the majority of
entrepreneurs and small business owners are likely to respond to changes as antici-
pated by economic theory. This suggests that similar experiments can complement
the literature by controlling for various endogeneity issues, and by allowing us to
collect individual-specific information such as psychographic and demographic
data. In the context of regulation, naturally occurring choices can be limiting in the
answers they can provide. An experiment such as ours yields data that can enable
researchers to explore hypotheses that are important and relevant but otherwise not
amenable to analysis using non-experimental data.
Specifically, we find that the introduction of occupational licensing in Kansas or
Missouri will reduce the probability of startup entry by 1.41%. Moreover, we esti-
mate that a one percentage point increase in the statutory state corporate income tax
rate in Kansas and Missouri will decrease entrepreneurial entry by 41% and 35.4%,
respectively. Our estimated elasticity of entrepreneurial entry, with respect to state
corporate income tax changes, is in-line with previous findings in the empirical lit-
erature. Gentry and Hubbard (2005) find that a five-percentage-point increase in the
convexity of the individual income tax schedule (difference in the weighted aver-
age of the marginal tax rates on self-employment income as entrepreneurial success
moves them across different tax brackets) reduces the probability of market entry by
21%, or approximately 0.67 percentage points from an average probability of mar-
ket entry of 3.26%. Rohlin et al. (2014) also provide evidence of state corporate
tax deterrent effects on firm location along a state border. They show that a one
percentage point reduction in the maximum corporate income tax rate on one side

13
Entrepreneurial response to interstate regulatory… 189

of the border will decrease (increase) the likelihood that new firms locate on that
side by 23 (0.2) percentage points if reciprocal tax agreements are (not) in place.
This also highlights the importance of accounting for such reciprocal tax agreements
between states that require employees to pay personal income tax to their state of
residence as opposed to their state of employment. While not directly comparable,
Curtis and Decker (2018) use a border discontinuity design to find that startup activ-
ity (defined as employment in startup firms) declines by 3.7% in response to a one
percentage point increase in the corporate tax rate. However, they also find no evi-
dence that corporate tax rates increased activity across the border from tax-raising
states, contrasting our results which indicate positive “cross-price” elasticities of
startup entry to changes in neighboring tax-raising states. Using federal administra-
tive data, DeBacker et al. (2018) examine a large-scale tax reform in Kansas aiming
to enhance new business formation by excluding certain forms of business income
from individual taxation. They report that the policy mainly fueled tax avoidance as
opposed to real supply side responses which highlights the importance of taxation
as an impediment to new business formation. Finally, we show that a 1-day increase
in business registration time in Missouri and Kansas will decrease entrepreneurial
entry by 0.337% and 0.289%, respectively.
Despite this value, some shortcomings remain. Most obviously, these results
are based on hypothetical choices, and recent research suggests that participants
can sometimes respond to surveys inattentively (Malone and Lusk 2018b). To this
point, it is worth reiterating that the discrete choice experiment is not intended to
be a panacea. Rather, this approach is meant to serve as a complement to ongoing
scholarly research in the field. Future studies might be conducted in the lab where
entrepreneurs might make decisions with varying levels of government frameworks.
Furthermore, we asked participants to assume that the two locations were the same
in all things except for the attributes we listed. While it is possible for state govern-
ments to make these changes in the short term without sacrificing services, cutting
income taxes might create long-term consequences in the ability of state govern-
ments to provide entrepreneurial support and infrastructure. Finally, prior studies
have suggested that regulations might affect businesses differently based on their
size (Calcagno and Sobel 2014). We show they also affect startup and location deci-
sions. Future studies might incorporate this article into their empirical design by
asking entrepreneurs about how many additional employees they would hire given a
regulatory or tax change.

Acknowledgements Funding for the study was provided by the Institute for the Study of Free Enterprise
at Oklahoma State University.

References
Acs, Z. J., & Varga, A. (2005). Entrepreneurship, agglomeration, and technological change. Small Busi-
ness Economics, 24(3), 323–334.
Agrawal, D. R., Fox, W. F., & Slemrod, J. (2015). Competition and subnational governments: Tax com-
petition, competition in urban areas, and education competition. National Tax Journal, 68(3),
701–734.

13
190 T. Malone et al.

Aidis, R., Estrin, S., & Mickiewicz, T. M. (2012). Size matters: Entrepreneurial entry and government.
Small Business Economics, 39(1), 119–139.
Arentz, J., Sautet, F., & Storr, V. (2013). Prior-knowledge and opportunity identification. Small Business
Economics, 41(2), 461–478.
Bailey, J. B., & Thomas, D. W. (2017). Regulating away competition: The effect of regulation on entre-
preneurship and employment. Journal of Regulatory Economics, 52(3), 237–254.
Besley, T. (2015). Law, regulation, and the business climate: The nature and influence of the world bank
doing business project. Journal of Economic Perspectives, 29(3), 99–120.
Bastiat, F. (1848). Sophisms of the protective policy. New York: Putnam Publishing.
Brooks, K., & Lusk, J. L. (2010). Stated and revealed preferences for organic and cloned milk: Com-
bining choice experiment and scanner data. American Journal of Agricultural Economics, 92(4),
1229–1241.
Bruce, D., & Mohsin, M. (2006). Tax policy and entrepreneurship: New time series evidence. Small Busi-
ness Economics, 26(5), 409–425.
Bylund, P. L., & McCaffrey, M. (2017). A theory of entrepreneurship and institutional uncertainty. Jour-
nal of Business Venturing, 32(5), 461–475.
Calcagno, P. T., & Sobel, R. S. (2014). Regulatory costs on entrepreneurship and establishment employ-
ment size. Small Business Economics, 42(3), 541–559.
Carlsson, F., & Martinsson, P. (2001). Do hypothetical and actual marginal willingness to pay differ in
choice experiments? Application to the valuation of the environment. Journal of Environmental
Economics and Management, 41(2), 179–192.
Carruthers, B. G., & Lamoreaux, N. R. (2016). Regulatory races: The effects of jurisdictional competi-
tion on regulatory standards. Journal of Economic Literature, 54(1), 52–97.
Cassar, G. (2004). The financing of business start-ups. Journal of Business Venturing, 19(2), 261–283.
Chang, J. B., Lusk, J. L., & Norwood, F. B. (2009). How closely do hypothetical surveys and laboratory
experiments predict field behavior? American Journal of Agricultural Economics, 91(2), 518–534.
Chirinko, R. S., & Wilson, D. J. (2017). Tax competition among U.S. States: Racing to the bottom or rid-
ing on a seesaw? Federal Reserve Bank of San Francisco Working Paper.
Coase, R. H. (1937). The nature of the firm. Economica, 4(16), 386–405.
Coase, R. H. (1992). The institutional structure of production. The American Economic Review, 82(4),
713–719.
Curtis, E. M., & Decker, R. A. (2018). “Entrepreneurship and state taxation”, finance and economics
discussion series 2018–003. Washington: Board of Governors of the Federal Reserve System. https​
://doi.org/10.17016​/FEDS.2018.003.
Da Rin, M., Di Giacomo, M., & Sembenelli, A. (2011). Entrepreneurship, firm entry, and the taxation of
corporate income: Evidence from Europe. Journal of Public Economics, 95(9), 1048–1066.
De Soto, H. (1989). The other path. New York: Harper & Row.
DeBacker, J., Goodman, L., Heim, B. T., Ramnath, S. P., & Ross, J. M. (2018). Pass-through entity
responses to preferential tax rates: Evidence on economic activity and owner compensation in Kan-
sas. National Tax Journal, 71(4), 687–706.
Delgado, M., Porter, M. E., & Stern, S. (2010). Clusters and Entrepreneurship. Journal of Economic
Geography, 10(4), 495–518.
Djankov, S., Ganser, T., McLiesh, C., Ramalho, R., & Shleifer, A. (2010). The effect of corporate taxes
on investment and entrepreneurship. American Economic Journal: Macroeconomics, 2(3), 31–64.
Djankov, S., La Porta, R., Lopez-de-Silanes, F., & Shleifer, A. (2002). The regulation of entry. Quarterly
Journal of Economics, 117(1), 1–37.
Ellison, G., Glaeser, E. L., & Kerr, W. R. (2010). What causes industry agglomeration? Evidence from
coagglomeration patterns. American Economic Review, 100(3), 1195–1213.
Foss, N. J., & Klein, P. G. (2012). Organizing entrepreneurial judgment: A new approach to the firm.
Cambridge, UK: Cambridge University Press.
Gentry, W. M., & Hubbard, R. G. (2005). “Success taxes.” Entrepreneurial entry, and innovation. Innova-
tion Policy and the Economy, 5, 87–108.
Harberger, A. (1962). The incidence of the corporate income tax. Journal of Political Economy, 70(3),
215–240.
Harrison, G. W., & List, J. A. (2004). Field experiments. Journal of Economic Literature, 42(4),
1009–1055.
Haufler, A., Norbäck, P. J., & Persson, L. (2014). Entrepreneurial innovations and taxation. Journal of
Public Economics, 113, 13–31.

13
Entrepreneurial response to interstate regulatory… 191

Henrekson, M., & Sanandaji, T. (2011). Entrepreneurship and the theory of taxation. Small Business Eco-
nomics, 37(2), 167–185.
Hensher, D., Louviere, J., & Swait, J. (1998). Combining sources of preference data. Journal of Econo-
metrics, 89(1), 197–221.
Hensher, D., Rose, J. M., & Greene, W. H. (2015). Applied choice analysis (2nd ed.). Cambridge: Cam-
bridge University Press.
Hsu, D. K., Simmons, S. A., & Wieland, A. M. (2017). Designing entrepreneurship experiments: A
review, typology, and research agenda. Organizational Research Methods, 20(3), 379–412.
Jiang, B., & Capra, C. M. (2016). Are (active) entrepreneurs a different breed? Managerial and Decision
Economics, 39, 613–628.
Kawaguchi, D., Murao, T., & Kambayashi, R. (2014). Incidence of strict quality standards: Protection of
consumers or windfall for professionals? Journal of Law and Economics, 57(1), 195–224.
Klapper, L., Laeven, L., & Rajan, R. (2006). Entry regulation as a barrier to entrepreneurship. Journal of
Financial Economics, 82(3), 591–629.
Kleiner, M. M., & Krueger, A. B. (2010). The prevalence and effects of occupational licensing. British
Journal of Industrial Relations, 48(4), 676–687.
Kleiner, M. M., & Krueger, A. B. (2013). Analyzing the extent and influence of occupational licensing on
the labor market. Journal of Labor Economics, 31(1), S173–S202.
Kleiner, M. M., & Vorotnikov, E. (2017). Analyzing occupational licensing among the states. Journal of
Regulatory Economics, 52(2), 132–158.
Knight, F. H. ([1921] 1985). Risk, uncertainty and profit. Chicago, IL: University of Chicago Press.
Lancaster, K. J. (1966). A new approach to consumer theory. Journal of Political Economy, 74(2),
132–157.
Louviere, J. J., Flynn, T. N., & Carson, R. T. (2010). Discrete choice experiments are not conjoint analy-
sis. Journal of Choice Modelling, 3(3), 57–72.
Louviere, J. J., Hensher, D. A., & Swait, J. D. (2000). Stated choice methods: Analysis and applications.
Cambridge: Cambridge University Press.
Lusk, J. L., & Schroeder, T. C. (2004). Are choice experiments incentive compatible? A test with quality
differentiated beef steaks. American Journal of Agricultural Economics, 86(2), 467–482.
Malone, T., & Lusk, J. L. (2017). Taste trumps health and safety: Incorporating consumer perceptions
into a discrete choice experiment for meat. Journal of Agricultural and Applied Economics, 49(1),
139–157.
Malone, T., & Lusk, J. L. (2018a). An instrumental variable approach to distinguishing perceptions from
preferences for beer brands. Managerial and Decision Economics, 39, 403–417.
Malone, T., & Lusk, J. L. (2018b). Consequences of participant inattention with an application to carbon
taxes for meat products. Ecological Economics, 145, 218–230.
McFadden, D. (1974). Conditional logit analysis of qualitative choice behavior. In P. Zarembka (Ed.),
Frontiers of econometrics (pp. 105–142). New York: Academic Press.
Melo, P. C., Graham, D. J., & Noland, R. B. (2009). A meta-analysis of estimates of urban agglomeration
economies. Regional Science and Urban Economics, 39(3), 332–342.
Michelacci, C., & Silva, O. (2007). Why so many local entrepreneurs? Review of Economics and Statis-
tics, 89(4), 615–633.
Mises, L. V. (1949). Human action: A treatise on economics. New Haven, CN: Yale University Press.
Partridge, M. D., & Olfert, M. R. (2011). The winners’ choice: Sustainable economic strategies for suc-
cessful 21st-century regions. Applied Economic Perspectives and Policy, 33(2), 143–178.
Peterson, B. D., Pandya, S. S., & Leblang, D. (2014). Doctors with borders: Occupational licensing as an
implicit barrier to high skill migration. Public Choice, 160(1–2), 45–63.
Pizzola, B., & Tabarrok, A. (2017). Occupational licensing causes a wage premium: Evidence from a
natural experiment in Colorado’s funeral services industry. International Review of Law and Eco-
nomics, 50, 50–59.
Planet Money Podcast. (2016). Episode 699: Why did the job cross the road? NPR Public Media. https​://
www.npr.org/secti​ons/money​/2016/05/04/47679​9218/episo​de-699-why-did-the-job-cross​-the-road.
Redbird, B. (2017). The new closed shop? The Economic and structural effects of occupational licensure.
American Sociological Review, 82(3), 600–624.
Rohlin, S., Rosenthal, S. S., & Ross, A. (2014). Tax avoidance and business location in a state border
model. Journal of Urban Economics, 83, 34–49.
Sass, T. R. (2015). Licensure and worker quality: A comparison of alternative routes to teaching. Journal
of Law and Economics, 58(1), 1–35.

13
192 T. Malone et al.

Serrato, J. C. S., & Zidar, O. (2016). Who benefits from state corporate tax cuts? A local labor markets
approach with heterogeneous firms. American Economic Review, 106(9), 2582–2624.
Sobel, R. S., Clark, J. R., & Lee, D. R. (2007). Freedom, barriers to entry, entrepreneurship, and eco-
nomic progress. Review of Austrian Economics, 20(4), 221–236.
Stigler, G. J. (1971). The theory of economic regulation. Bell Journal of Economics and Management
Science, 2(1), 3–21.
Teague, M. (2016). Barriers to entry index: A ranking of starting a business difficulties for the United
States. Journal of Entrepreneurship and Public Policy, 5(3), 285–307.
Thornton, R. J., & Timmons, E. J. (2013). Licensing one of the world’s oldest professions: Massage.
Journal of Law and Economics, 56(2), 371–388.
van Rijnsoever, F., & Cerutti, F. (2017). Like a rolling stone? Heterogeneity in location preferences of
early-stage technology based start-ups. Working Paper.
Vaupel, A. (2017). John Oliver uses KC to demonstrate how ‘pointless’ incentives are. Kansas City Busi-
ness Journal. https​://www.bizjo​urnal​s.com/kansa​scity​/news/2017/11/07/last-week-tonig​ht-john-
olive​r-kc-incen​tives​-video​.html.
Williamson, O. E. (1975). Markets and hierarchies, analysis and antitrust implications: A study in the
economics of internal organization. New York: Free Press.
Williamson, O. E. (1985). The economic institutions of capitalism. New York: Free Press.
Williamson, O. E. (1996). The mechanisms of governance. Oxford: Oxford University Press.
Zhao, B. (2013). The effects of state innovation programs on entrepreneurial firms: Three essays. Univer-
sity of Michigan Ph.D. Dissertation Essay.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional affiliations.

13