Barber licensure and the supply of barber shops: evidence from U.S. States.
Hall, Joshua C. ; Pokharel, Shree B.
The Institute of Justice, a public interest law firm in Washington,
D.C., focuses part of its litigation on issues of the ability of
citizens to enter and compete in markets unburdened by unnecessary
regulations. In a study titled License to Work: A National Study of
Burdens from Occupational Licensing (Carpenter et al. 2012), the
Institute reports that "in the early 1950s, only one in 20 U.S.
workers needed the government's permission to pursue their chosen
occupation." In 2008, that number was estimated to be one in three
(Kleiner and Krueger 2013).
Given its pervasiveness, occupational licensing has long been a
subject of debate as to whether it serves to protect the public interest
or the interests of special interest groups by acting as a barrier to
entry. Proponents of occupational licensing argue occupational licensing
enables better quality services to consumers that would otherwise not
have been provided (Arrow 1971). It has also been argued that
occupational licensure encourages prospective entrepreneurs to
accumulate human capital in their occupation of choice (Akerlof 1970,
Shapiro 1986). Opponents, however, argue that occupational licensing
gives rise to regulatory capture (Stigler 1971) and results in barriers
to entry that disproportionately affect the poor and disadvantaged
(Dorsey 1983, Bernstein 1994). Supporting the claim of regulatory
capture is Kleiner (2000), who reports that more often than not members
of licensing boards are chosen from the occupations being licensed.
The literature on occupational licensure has typically focused on
the effects of licensure on wages and safety. A few articles focus on
licensure as a barrier to entry, but those studies largely deal with
high-skilled labor markets. Carpenter and Stephenson (2006), for
instance, find that 150 hours of college course work necessary to sit
for the CPA exam reduces the number of candidates sitting for the CPA
exam by 60 percent.
In this article, we focus on occupational licensure as a barrier to
entry for one relatively low-skilled occupation--barbering. The
barbering profession was one among many professions to be licensed early
in the United States, with Minnesota passing the first barber licensing
law in 1897 (Thornton and Weintraub 1979). Alabama was the last state to
license barbers in 2013 (Burkhalter 2014). Today all states and the
District of Columbia regulate barbering. In 1976, barbering was heavily
regulated with average education and experience requirements of 1,460
hours and a mean apprenticeship period of approximately 18 months
(Thornton and Weintraub 1979). By 2012, average education and experience
requirements were 890 days and average fee requirements were $330
(Carpenter et al. 2012).
While many studies have focused on occupational regulation and
economic outcome variables, such as changes in earnings and employment
(Kleiner 2000) and migration (Mulholland and Young 2016), few studies
have examined the impact of occupational regulatory burdens on
low-income professions such as barbering. Some previous studies have
estimated the relationship between regulatory burdens and the supply of
barbers (Fuchs and Wilburn 1967, Maurizi 1974, Thornton and Weintraub
1979). Thornton and Weintraub (1979) find that average minimum grade
level affects the supply of barbers. Timmons and Thornton (2010) find
that state barber licensure has increased barber earnings by between 11
and 22 percent.
In this study, we estimate the relationship between the state-level
regulatory burden on the practice of barbering and the number of barber
shops in a state. Since many barber shops are one-or two-chair shops,
restrictions on the profession of barbering are restrictions on the
number of barber shops. We hypothesize that states with higher
regulatory burdens on becoming a barber should have fewer barber shops
per capita. Utilizing the one year of regulatory data on barbering from
Carpenter et al. (2012), we find that the number of exams required to
become a barber in a state is negatively related to the number of barber
shops per capita in that year. Conversely, we find that fees, minimum
grade levels, and minimum age requirements do not explain state
variation in the number of barber shops per capita.
Data
We use barber shops per 100,000 inhabitants for all 50 states and
the District of Columbia in 2011 as our measure of entrepreneurial
barber activity. Our data come from the U.S. Census Bureau's
Nonemployer Statistics database, and we use the North American Industry
Classification System (NAICS) code for barber shops (812111) to identify
"establishments known as barber shops or men's hair stylist
shops engaged in cutting, trimming, and styling boys' and
men's hair; and/or shaving and trimming men's beards."
State population in 2011 was obtained from the U.S. Census Bureau. The
dependent variable is the authors' calculation with scores ranging
from 14.0 (Utah) to 92.8 (Alabama) for each state. The score is
calculated by dividing each state's total barber shop
establishments by the state population. For example, Alabama's
score means that, on average, there are approximately 93 barber shops
for every 100,000 residents.
There are three categories of explanatory variables in this study
which might affect the number of barber shops per capita: Measures of
Occupational Regulation, State Controls, and Attributes of
Entrepreneurs. Our major variables of interest fall in the Measures of
Occupational Regulation category and consist of variables representing
governmental burdens imposed by state governments on prospective
barbers. The variables included in this category are average number of
exams, average fees, average minimum grade level, and average minimum
age imposed by states on barbers to acquire a license. These variables
are reported from License to Work: A National Study of Burdens from
Occupational Licensing. While Carpenter et al. (2012) provide regulatory
information for 102 occupations in which the average income is below the
national average, we use only their measures of occupational licensure
for barbering. (1)
All variables in this category are reported in their original form.
Fees are in dollars and represent the payments necessary to achieve an
initial license. Continuing education fees and renewal fees are not
included. Number of Exams represent the number of written and practical
exams required in a state to get a license. Minimum Grade is the minimum
education level necessary to apply for a license. States without a
minimum grade level receive a 0; states with an eighth-grade minimum
receive an 8, with a high school minimum a 12, and so on. For barber
licensure, no state requires more than a 12th grade education. Minimum
Age is the minimum age an individual in the state must be to apply for
barber licensure and varies across states from 0 to 18. Many states,
such as Iowa, have both a minimum grade and age requirement.
In addition to regulatory burdens, state-specific variables related
to the economic or social environment might also influence the decision
to become a barber and open a barber shop. We primarily draw on the
entrepreneurship literature as the motivation for these controls, which
are all measured for 2011. For example, the Unemployment Rate is found
to negatively affect self-employment across OECD countries (Blanchflower
2000). At the level of U.S. states, however, the results are mixed.
Unemployment is found to have an insignificant relationship with new
business starts (Carree 2002) and a negative relationship with latent
entrepreneurship (Gohmann 2012). However, Gohmann and Fernandez (2014)
find that unemployment Granger-causes proprietorships. In addition,
Coomes, Fernandez, and Gohmann (2013) find that the unemployment rate is
positively related to proprietorships at the MSA level. Unemployment
Rate is obtained from the Bureau of Labor Statistics. The role of median
income in influencing entrepreneurship is unclear (Yago, Barth, and
Zeidman 2007), but is generally thought to positively affect the number
of new businesses as individuals seek greater diversity in consumption.
Median Household Income for each state and the District of Columbia was
obtained from the U.S. Census Bureau. Crime has been found to negatively
affect entrepreneurship (Rosenthal and Ross 2010), and we use Property
Crimes from the FBI's Uniform Crime Reports.
In addition to regulatory burden variables and state control
variables that affect entrepreneurship, we also include demographic
controls to capture the attributes of those most likely to start a
business. The variables in this category include percentage of labor
force that is Male, percentage of labor force that is White, and Median
Age of state residents. All demographic data are for 2011 and were
obtained from the Bureau of Labor Statistics and the U.S. Census Bureau.
Kreft and Sobel (2005) and Hall and Sobel (2008) find that the
percentage of labor force that is male and white, as well as the median
age within the state, affect entrepreneurship. Similarly, Langowitz and
Minniti (2007) find that the probability of men being entrepreneurs is
higher than women. Table 1 presents summary statistics for all variables
employed in our article.
The dependent variable, Barber Shops per 100,000 Residents, has a
mean of 38.80. This means that, on average, there were
<2522043H_TB001> approximately 39 barber shops per 100,000
residents per state throughout the United States in 2011. There is
significant variation across states, however, in the number of barber
shops. While Utah had only 14 barber shops per 100,000 residents,
Alabama had approximately 93 barber shops per 100,000 of its residents.
There is a lot of variation in terms of Number of Exams. Alabama
requires no exams while Minnesota and Nevada require four exams. Fees
varied considerably as well from $0 in the District of Columbia to $330
in Kentucky. Minimum Grade requirements also varied across states with
many requiring no educational attainment level while others specifically
require at least a high school or equivalent degree. In terms of Minimum
Age, many states do not have a minimum age requirement to be a barber
while in other states one has to be at least 18 years old.
Unemployment rates vary across states as well. North Dakota had the
lowest unemployment rate of 3.5 percent while Nevada had the highest
unemployment rate of 13.1 percent. Property crime rates across states
also vary notably. Rhode Island had the lowest property crime rate with
approximately 1,395 crimes per 100,000 inhabitants, while Washington,
D.C., had the most property crimes per 100,000 inhabitants. There is
almost a 1.5 times difference between the state with the lowest median
income and the state with the highest median income; Kentucky has median
income of $39,856 while Maryland has a median income of $68,876. Percent
of male population in a state also varies with 47.3 percent in District
of Columbia to 51.9 percent in Alaska. There is a significant difference
within states in terms of racial composition as well. While
approximately 26 percent of Hawaii's population is white, 95.48
percent of Vermont's population is white.
Empirical Approach and Results
Since we are limited in terms of numbers of observations in our
data set, we employ a simple linear OLS regression model for our
empirical analysis. Our model is represented as follows:
BARBSHOPS = [[beta].sub.0] + [beta]REGULATION + [gamma]STATE +
[delta]ENTREPRENEUR + [epsilon]
where [beta], [gamma], and [delta] are row vectors and REGULATION,
STATE, and ENTREPRENEUR are column vectors. BARBSHOPS represents total
barber shops per capita in U.S. states. As mentioned in the previous
section, REGULATION represents barber-specific regulatory variables; it
consists of Number of Exams, Fees, Minimum Grade, and Minimum Age. STATE
represents state controls and therefore includes Unemployment Rate,
Property Crimes, and Income. Attributes of entrepreneurs are represented
by ENTREPRENEUR and consist of Male, White, and Median Age.
Table 2 shows the effect of state-level barber regulations on the
total number of barber shops per capita in 2011. Specification 1
represents a parsimonious specification containing only the primary
variables of interest. While this specification does not explain the
full effect of the explanatory variables on the dependent variable, it
helps to outline the basic relationship between them. The signs of
Minimum Grade, Fees, and Number of Exams are as expected, with Number of
Exams statistically significant at the 1 percent level. The sign on
Minimum Age is positive although not statistically significant.
In Specification 2, we add basic entrepreneur characteristics
controls standard in the literature. We find that Number of Exams is
still significant at the 1 percent level. Fees continue to be negatively
associated with the dependent variable, but is still statistically
insignificant. Male exhibits a strong negative relationship at the 1
percent level on the number of barber shops. White also is negatively
related to the number of barber shops at the 5 percent level. However,
the signs for Male and White exhibit opposite signs than what was
previously found for other measures of entrepreneurship (Kreft and Sobel
2005, Hall and Sobel 2008).
Specification 3 adds basic state controls standard in the
literature--median household income and the unemployment rate. The key
result is drat Number of Exams continues to exhibit a significant
negative effect on the level of barber shops in a state at the 1 percent
level. Income leads to fewer barber shops per capita, although the
economic magnitude is small. The sign on Unemployment Rate is positive
but statistically insignificant (Blanchflower 2000).
Finally, in Specification 4, we add Property Crimes and Median Age.
Number of Exams continues to be negatively related to the number of
barbershops per capita at the 1 percent level. White, Male, and Income
are statistically significant as well as Median Age. Property Crimes are
not significant. This full specification explains 71 percent of the
variation in barber shops per capita in 2011 across U.S. states.
Conclusion
Given the growth in occupational licensure and the importance of
barriers to entry for low-income workers, we analyzed the effect of
barber licensure on the number of barber shops across U.S. states. We
find that the number of required exams is robustly associated in a
negative way with the number of barber shops per capita in a state.
However, we find that other restrictions such as age requirements and
fees have no consistent relationship with the number of barber shops.
This might be the result of our limited data set. We feel that this
exploratory look at the issue of barber licensure opens up future
research in this area, especially research that can establish more of a
causal link.
Further research could focus on the origins of these laws,
especially since historically many of these laws have their roots in
discrimination. Bernstein (1994), for example, details how licensing
laws have historically been used to reduce the number of
African-Americans in certain occupations such as barbering. As Kuznicki
(2009) points out, government power exercised through things like
occupational licensure is never neutral when it comes to race. Our
results also say nothing about the efficacy of the restrictions in terms
of the quality of haircuts received in states with more stringent
regulations. Carpenter (2012) is a good example of the type of applied
research that could be done in this area, as he finds no difference
between licensed and unlicensed florists. Finally, it would be fruitful
to further investigate the vast differences across states in the amount
of regulation of certain industries, such as barbering, from a political
economy perspective, in order to better understand the various special
interests at play.
References
Akerlof, G. (1970) "The Market for Lemons: Qualitative
Uncertainty and the Market Mechanism." Quarterly Journal of
Economics 84 (3): 488-500.
Arrow, K. J. (1971) The Theory of Risk Aversion. In K. J. Arrow
(ed.), Essays in the Theory of Risk-Bearing, 90-120. Chicago: Markham.
Bernstein, D. E. (1994) "Licensing Laws: A Historical Example
of the Use of Government Regulatory Power against African
Americans." San Diego Law Review 31 (1): 89-104.
Blanchflower, D. G. (2000) "Self-Employment in OECD
Countries." Labour Economics 7 (5): 471-505.
Burkhalter, E. (2014) "Barbers Facing New State
Regulations." The Anniston Star (February 5). Available at
www.annistonstar .com/news/barbers-facing-new-state-regulations/article_7ccec4f5 -832a-5848-acb1-4695d14ad821.html.
Carpenter, C. G., and Stephenson, E. F. (2006) "The 150-Hour
Rule as a Barrier to Entering Public Accountancy." Journal of Labor
Research 27 (1): 115-26.
Carpenter, D. M. (2012) "Testing the Utility of Licensing:
Evidence from a Field Experiment on Occupational Regulation."
Journal of Applied Business and Economics 13 (2): 28-41.
Carpenter, D. M.; Knepper, L.; Erickson, A. C.; and Ross, J. K.
(2012) License to Work: A National Study of Burdens from Occupational
Licensing. Washington: Institute for Justice.
--(2015) "Regulating Work: Measuring the Scope and Burden of
Occupational Licensure among Low-and Moderate-Income Occupations in the
United States." Economic Affairs 35 (1): 3-20.
Carree, M. A. (2002) "Does Unemployment Affect the Number of
Establishments? A Regional Analysis for U.S. States." Regional
Studies 36 (4): 389-98.
Coomes, P. A.; Fernandez, J.; and Gohmann, S. F. (2013) "The
Rate of Proprietorship among Metropolitan Areas: The Impact of the Local
Economic Environment and Capital Resources." Entrepreneurship
Theory and Practice 37 (4): 74.5-70.
Dorsey, S. (1983) "Occupational Licensing and
Minorities." Law and Human Behavior 7 (2-3): 171-81.
Fuchs, V. R., and Wilburn, J. A., eds. (1967). Productivity
Differences within the Service Sector. New York: Columbia University
Press.
Gohmann, S. F. (2012) "Institutions, Latent Entrepreneurship,
and Self-Employment: An International Comparison." Entrepreneurship
Theory and Practice 36 (2): 295-321.
Gohmann, S. F., and Fernandez, J. M. (2014) "Proprietorship
and Unemployment in the United States." Journal of Business
Venturing 29 (2): 289-309.
Hall, J. C., and Sobel, R. S. (2008) "Institutions,
Entrepreneurship, and Regional Differences in Economic Growth."
Southern Journal of Entrepreneurship 1 (1): 69-96.
Kleiner, M. M. (2000) "Occupational Licensing." Journal
of Economic Perspectives 14 (4): 189-202.
--(2014) "Life, Limbs, and Licensing: Occupational Regulation,
Wages, and Workplace Safety of Electricians, 1992-2007." Monthly
Labor Review 137: 1-31.
Kleiner, M. M., and Krueger, A. B. (2013). "Analyzing the
Extent and Influence of Occupational Licensing on the Labor
Market." Journal of Labor Economics 31 (2): 173-202.
Kreft, S. F., and Sobel, R. S. (2005) "Public Policy,
Entrepreneurship, and Economic Freedom." Cato Journal 25 (3):
59.5-616,
Kuznicki, J. (2009) "Never Neutral State: American Race
Relations and Government Power." Cato Journal 29 (3): 417-54.
Langowitz, N., and Minniti, M. (2007) "The Entrepreneurial
Propensity of Women." Entrepreneurship Theory and Practice 31 (3):
341-64.
Maurizi, A. (1974) "Occupational Licensing and the Public
Interest." Journal of Political Economy 82 (2): 399-413.
Mulholland, S. E., and Young, A. T. (2016) "Occupational
Licensing and Interstate Migration." Cato Journal 36 (1): 17-31.
Rosenthal, S. S., and Ross, A. (2010) "Violent Crime,
Entrepreneurship, and Cities." Journal of Urban Economics 67 (1):
135-49.
Shapiro, C. (1986) "Investment, Moral Hazard, and Occupational
Licensing." Review of Economic Studies 53 (5): 843-62.
Stigler, G. J. (1971) "The Theory of Economic
Regulation." Bell Journal of Economics and Management Science 2
(1): 3-21.
Thornton, R J., and Weintraub, A. R. (1979) "Licensing in the
Barbering Profession." Industrial and Labor Relations Review 32
(2): 242-49.
Timmons, E. J., and Thornton, R. J. (2010) "The Licensing of
Barbers in the USA." British Journal of Industrial Relations 48
(4): 740-57.
Yago, G.; Barth, J. R.; and Zeidman, B. (2007) Entrepreneurship in
Emerging Domestic Markets: Barriers and Innovation. Springer: New York.
(1) Carpenter et al. (2012) also provide a measure of the days of
education and experience necessary to achieve a license. Doing so
requires a number of assumptions, however, and we prefer to focus on the
directly comparable features of barber regulation listed such as fees
and number of exams. Inclusion of the number of days of education and
experience does not qualitatively affect our empirical results.
Joshua C. Hall is Associate Professor of Economics at West Virginia
University and Shree B. Pokharel is a Graduate Research Assistant.
TABLE 1
SUMMARY STATISTICS
Variable Mean St. Dev. Min. Max.
Barber Shops per 38.80 20.87 14.0 92.8
100,000 Residents
Number of Exams 2.20 0.69 0.0 4.0
Fees 127.63 70.25 0.0 330.0
Minimum Grade 6.78 5.21 0.0 12.0
Minimum Age 13.31 7.08 0.0 18.0
White 79.73 13,33 26.03 95.48
Male 49.33 0.79 47.3 51.9
Income 50,686.47 7,475.12 39,856.0 68,876.0
Unemployment Rate 8.17 1.93 3.5 13.1
Property Crimes 2,863.06 670.07 1,395.2 4,795.5
Median Age 37.73 2.38 29.9 43.5
NOTES: N = 51 (all U.S. states and the District of Columbia).
For sources of data, see discussion in text.
TABLE 2
STATE-LEVEL BARBER REGULATIONS AND NUMBER OF
BARBER SHOPS PER CAPITA
Variable (1) (2) (3) (4)
Number of -13.195 *** -9.079 *** -8.811 *** -8.141 ***
Exams (3.97) (3.14) (2.76) (2.74)
Fees -0.025 -0.003 -0.010 -0.020
(0.04) (0.03) (0.03) (0.03)
Minimum Grade -0.431 0.308 0.308 0.408
(0.56) (0.48) (0.43) (0.44)
Minimum Age 0.622 0.317 -0.093 -0.078
(0.41) (0.32) (0.30) (0.30)
White -0.414 ** -0.394 ** -0.387 **
(0.19) (0.17) (0.18)
Male -13.243 *** -11.255 *** -13.914 ***
(2.82) (2.58) (2.90)
Income -0.001 *** -0.001 ***
(0.00) (0.00)
Unemployment 1.834 1.452
Rate (1.12) (1.14)
Property -0.004
Crimes (0.00)
Median Age -1.713 *
(0.96)
R-squared 0.23 0.57 0.68 0.71
NOTE: Dependent variable is the number of barber shops per
100,000 state residents.
N =51 in all specifications. *, **, and *** denote
statistical significance at the 10, 5, and 1% levels,
respectively. Numbers in parentheses are absolute standard
errors. Constant included but not reported.
