Content uploaded by Troy Blanchard
Author content
All content in this area was uploaded by Troy Blanchard on Mar 18, 2014
Content may be subject to copyright.
The health and wealth of US counties: how the
small business environment impacts alternative
measures of development
Troy C. Blanchard
a
, Charles Tolbert
b
and Carson Mencken
b
a
Department of Sociology, Louisiana State University, 126 Stubbs Hall, Baton Rouge, LA 70803,
USA, troy@lsu.edu
b
Department of Sociology, Baylor University, Waco, TX 76798, USA,
charlie_tolbert@baylor.edu; carson_mencken@baylor.edu
Received on September 1, 2010; accepted on October 19, 2011
In this paper, we evaluate the prospects of small business-driven job creation by assessing
the link between small business and population health, an alternative measure of economic
development. We combine two literatures from the social capital perspective of aggregate
community well-being to model the effects of small-business concentration on aggregate
measures of population health. We argue that entrepreneurial culture facilitates
collective efficacy for a community and provides a problem-solving capacity for addressing
local public health problems. Our analysis demonstrates that communities with a greater
concentration of small businesses, ceteris paribus, have greater levels of population health.
Implications for theory and research are discussed.
Keywords: small business, community well-being, public health, entrepreneurship
JEL Classifications: I, O0
Introduction
The link between economic development and pop-
ulation health for countries is well established.
Researchers have demonstrated that increases in
productivity measured as per capita income result
in improved living conditions in terms of diet, san-
itation and availability of health services (McKeown,
1976;Preston, 1975). In the USA, health insurance is
the key variable that affects access to an increasingly
less affordable health-care system. Moreover, in the
USA, access to health insurance comes primarily
through employer compensation and benefits plans.
Since the late 1970s, the US economy has experi-
enced significant restructuring, and traditional large/
core economy employers have rescaled the organi-
zation of work with practices that have reduced pay,
benefits and job tenure (Edmiston, 2007;Hollister,
2004;Variyam and Kraybill, 1998).
When it comes to economic development, US
policy makers have traditionally pursued a smoke-
stack chasing approach to enhance the welfare of
local residents. Communities sought to lure large
employers from the outside to provide residents
with high-paying jobs (see Cobb, 1982;Isserman,
1994). Sociologists proposed a dual economy the-
ory, a paradigm shift in the way sociologists and
other social scientists thought of the organization of
work (Granovetter and Tilly, 1988;Tolbert et al.,
ÓThe Author 2011. Published by Oxford University Press on behalf of the Cambridge Political Economy Society. All rights reserved.
For permissions, please email: journals.permissions@oup.com
Cambridge Journal of Regions, Economy and Society 2011, 1–14
doi:10.1093/cjres/rsr034
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
1980). The prevailing conclusion from this body of
work was that ‘bigger was better’ as larger firms
offered higher-quality jobs in the core sector of
the economy (Villemez and Bridges, 1988). These
jobs provided high levels of compensation, exten-
sive benefits packages, internal labour markets,
union access and a number of other work structures
that provided employees a lifetime of secure
employment (Kalleberg and Van Buren, 1996).
The periphery economy firms, in contrast, were
either small locally oriented employers who prac-
tised nepotism and provided little opportunity for
individual growth and advancement or firms that
were extremely vulnerable to market competition.
Periphery firms offered lower pay and few, if any,
benefits. Health care was more accessible to those
working in the core sector of the economy. There-
fore, places with a greater concentration of core
sector jobs would have greater levels of population
health.
Globalization has transformed the organization
of work in Western democracies. Traditional US
core economy manufacturing jobs have been
supplanted by technology or migrated to offshore
low-wage markets, putting downward pressure on
a once robust middle class and increasing aggre-
gate levels of inequality (Falk and Lobao, 2004;
Sullivan et al., 2000). Some scholars continue to
demonstrate, however, that large business jobs
have higher wages, higher-quality benefits, due
process in hiring/dismissal and a longer duration
(Edmiston, 2007;Haltiwanger et al., 2009;
Mencken and Winfield, 1998). Others question
whether bigger is still better because larger firms
are more vulnerable to global competition
and have had to restructure in order to survive
(Hollister, 2004). Part of this process included
trimming or eliminating work structures (unions,
internal labour markets, full-time employment)
with alternatives (contract/contingency work,
part-time work, consolidation) that give workers
a more tenuous connection to employment and
benefits.
The existing literature on the firm size compen-
sation effect uniformly concludes that larger firms
pay better and offer more benefits (Amato and
Amato, 2007;Cubbins and Parmer, 2001; Useem,
1988). Hollister (2004), however, finds that be-
tween 1988 and 2003, there has been a significant
drop in firm size-based wage differences (33% in
real dollars) and in firm size-based differences in
access to health insurance. Small firms are not add-
ing greater compensation or benefits, however;
large firms offered less of both in 2003 than in
1985. What restructuring has meant to the US
economy is reduced access to health insurance
and lower (real) wages to purchase health care.
More recently, economic development strategists
have shifted focus away from efforts to attract
external capital into making investments locally
to stimulate the creation of small businesses
(Edmiston, 2007). This approach has been fuelled
by statistics showing that from 1990 to 2003 the
small-business sector was responsible for 79.5%
of US job creation (Edmiston, 2007;U.S. Census
Bureau, Statistics of U.S. Businesses, 1990–2003).
Though the quantity of jobs generated by the small-
business sector is substantial (see Neumark et al.,
2008), the quality of jobs in the small-business sec-
tor has historically been inferior to those in large
enterprises (Variyam and Kraybill, 1998). And
while Hollister (2004) documents a shrinking
wage/benefit gap between large and small firms,
there is still a persistent gap. Small firms do not
have the advantages of scale to offer high wages
and extensive benefits packages, in particular health
insurance (Edmiston, 2007;Keene et al., 2010).
This transition in economic development policy
begs an important question: will small business-
based economic development efforts have a nega-
tive impact on the health and well-being of
community residents? The public health literature
suggests two competing outcomes. On the one
hand, traditional income-based explanations of
population health portend that the prospects for
community health are bleak because jobs in small
businesses usually offer lower wages and are less
likely to provide health insurance coverage. More-
over, economic restructuring has ‘created’ a class
of independent contractors, with large businesses
furloughing employees from full-time benefits
eligible positions and rehiring them as short-term
Blanchard, Tolbert and Mencken
2of14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
contract workers with no benefits (see Arum and
Mueller, 2004). Second, for some workers, self-
employment is a survival adaptation to changing
economic circumstances. It becomes an escape
from the roller coaster cycles of furloughs and call-
backs that have characterized life with big firms
during the last decade. Without health insurance
coverage, access to medical care, both therapeutic
and preventative, will be limited (see Gortmaker
and Wise, 1997;Olafsdottir, 2007). Therefore, we
should see a positive correlation between small-
business concentration and aggregate measures
of poor health (higher rates of mortality, obesity
and diabetes).
On the other hand, a well-established literature in
public health research, which focuses on the role of
social capital and collective efficacy, indicates that
there is a positive relationship between collective ef-
ficacy and community health (see Islam et al., 2006;
Macinko and Starfield, 2001). Collective efficacy
focuses on the connectedness of community mem-
bers that occurs through organizational activity di-
rected at problem solving (Dominguez and Arford,
2010;Gutierrez and Lewis, 1999). We maintain that
an important implication of the small-business sector
for public health is that small locally oriented busi-
nesses contribute to the collective efficacy of a com-
munity. In the literature review and analysis that
follows, we attempt to connect these two literatures
on the role of collective efficacy and community
health with that pertaining to the importance of the
small-business sector in creating a climate of local
orientation and local civic engagement.
The small-business environment, civic
community and collective efficacy
Over the past decade, researchers have paid increas-
ing attention to the role of small business in eco-
nomic development (Lyson et al., 2001;Tolbert
et al., 1998,2002). In this line of study, communi-
ties with a strong entrepreneurial culture experience
improved economic outcomes, such as median
household income, poverty and income inequality.
These authors argue that, unlike large manufactur-
ing plants and chain retail big box stores, small
businesses are embedded in the community and
have a greater investment in the well-being of local
residents (see also Mitchell, 2006).
An important aspect of the entrepreneurial cul-
ture is the capacity for a community to self-develop.
Self-development refers to an economic develop-
ment activity that requires investment of local
resources to stimulate the formation of locally
owned business enterprises (Green et al., 1990).
For rural communities, this approach has been uti-
lized to create economic development programmes
(for an example, see Korsching and Allen, 2004).
With respect to the well-being of communities
and public health, the notion of self-development
implies that community residents and leaders have
a higher level of collective efficacy or the capacity
and willingness of community members to take re-
sponsibility for solving local problems (Sampson
et al. 1997,1999). Collective efficacy requires
a milieu of trust, cooperation and local orientation
(Araya et al., 2006). The presence of small-business
owners enhances collective efficacy because these
owners are agents in local economic development.
Many small businesses are operated by entrepre-
neurs who are important fixtures in the local com-
munity and who have accumulated substantial
human and social capital (Dunn and Holtz-Eakin,
2000;Hout and Rosen, 2000). Because they are
investing in the local community, the local entre-
preneur classes are seeking significant returns on
their investments. It is in their financial interest to
use their resources to maximize the efficacy of the
community. In contrast, communities seeking to
develop through outside investment become depen-
dent on the actions of external entities to address
problems.
The value of the entrepreneurial spirit and com-
munity self-development was described decades
ago by Mills and Ulmer (1946). In a comparative
analysis of communities with and without a strong
small-business sector, the authors found that small
businesses are central to local problem solving
through two mechanisms. First, small-business
owners were motivated to help solve local problems
because community improvement enhanced their
quality of life and the profitability of their business.
Health and wealth of US counties
3of14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
In contrast, communities lacking a strong small-
business sector demonstrated a diminished capacity
to address local problems.
A second point raised by Mills and Ulmer is the
concentration of community power. A strong small-
business environment creates a diffusion of eco-
nomic activity across a larger number of business
owners. This results in a pluralistic power structure
where a large number of business leaders compete
for prestige and influence in local decision-making.
Communities with a higher concentration of eco-
nomic activity through the presence of large
employers tended to be dominated by the interests
of the corporation (Mills and Ulmer, 1946). When
community interests diverge from the interests of
the corporation, a large employer may threaten to
relocate or withhold support.
More recently, the multilevel association between
small-business environments and collective efficacy
has been documented by Blanchard and Matthews
(2006). These authors find that the prevalence of large
employers within a community reduces participation
in community affairs because residents are alienated
from the decision-making process as corporate goals
take priority over the needs of local residents. Similar
to the findings of Mills and Ulmer (1946) regarding
small-business communities, residents in communi-
ties with smaller competitive business environments
are far more active in local problem solving and dem-
onstrate a higher level of community efficacy.
The small-business environment and
local population health
We hypothesize that communities with a strong en-
trepreneurial culture will demonstrate higher levels
of population health than communities lacking
a strong small-business sector. This hypothesis is
based on the self-development orientation of com-
munities possessing a strong small-business sector.
Within the context of public health, collective
efficacy is conceptualized as a community-level
form of social capital that has been linked to a vari-
ety of health outcomes (Dominguez and Arford,
2010). Collective efficacy focuses on the connect-
edness of community members that occurs through
organizational activity directed at problem solving
(Gutierrez and Lewis, 1999). An important impli-
cation of the entrepreneurial culture for public
health is that the small-business sector contributes
to the collective efficacy of a community.
The literature suggests several pathways through
which collective efficacy may influence health
outcomes (Islam et al., 2006). One mechanism for
linking collective efficacy to population health is
through a community’s willingness to invest in lo-
cal health infrastructure. Prior studies of population
health find that investment in the local health-care
system is directly associated with improved popu-
lation health (Daly et al., 1998,Lynch et al., 2000).
Health-care infrastructure may include hospitals
and other medical facilities, physicians and other
medical personnel.
A second and closely related mechanism is
investments in other aspects of health, such as
educational programming, environmental safety,
affordable housing and recreational facilities
(Lynch et al., 2004). Likewise, community health
promotion programmes are vital instruments
through which local populations are educated about
their health status and learn intervention informa-
tion to avoid or manage chronic illnesses (Hawe
and Shiell, 2000). Community health programmes
depend on vertical integration among national,
state, regional and local health authorities and com-
munity stakeholders in order to be successful. Yet
one programme does not fit all communities or
health issues. As a result, no one agency has suffi-
cient resources to identify community needs and to
design the right programme for a specific commu-
nity. What is required is significant input and co-
ordination from stakeholders, local government
entities, private businesses, faith-based organiza-
tions and the like. Green et al. (2001) argue that if
trust is not established among all constituents, then
the programme will not be successful. Moreover,
this trust is also necessary to institute local legisla-
tion, such as banning smoking in local restaurants
and bars, in order to minimize community division
and backlash to the legislation (see Stebbins, 1997).
Recent research by Payne and Williams (2008)
serves as an example. They document the role of
Blanchard, Tolbert and Mencken
4of14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
collective efficacy in Riverside, California. In this
case, a variety of local organizations, such as
churches and civic groups, and local government
agencies organized to develop a community centre
in an impoverished neighbourhood to provide in-
formation and health services for local residents.
With respect to obesity, Cohen et al. (2006) argue
that high levels of collective efficacy help to main-
tain the built environment in terms of safety and
walkability of the community. According to these
authors, safety and walkability influence levels of
physical activity and body mass index (Ross and
Mirowsky, 2001). Another mechanism through
which the entrepreneurial culture relates to health
is the social control of health behaviours. In a social
environment with high collective efficacy, commu-
nity members are better able to enforce formal and
informal norms regarding public behaviours, such
as smoking, violence, alcohol consumption or poor
diet. In addition, because collective efficacy
involves community participation and the protec-
tion of the well-being of others, collective efficacy
may lead individuals to avoid unhealthy behaviours
by enhancing self-accountability (Ewing et al.,
2003;Folland, 2007). Cohen et al. (2006) identify
two pathways through which collective efficacy
influences health outcomes via social control. First,
residents of communities with low collective effi-
cacy experience higher levels of stress due to the
lack of social ties and social support. A second
pathway involves the level of care shown by
adults in high collective efficacy neighbourhoods
towards the behaviour of children in the commu-
nity. A high level of collective efficacy results in
adult community members being involved in local
sports leagues and encouraging healthier food
options for children in addition to expressing
disapproval towards sedentary activity.
We expect that a thriving small-business sector
will yield salutary health benefits to communities.
The entrepreneurial culture generated by the small-
business sector will create local orientation among
community leaders from the small-business class.
This, in turn, will enhance the level of collective
efficacy in communities. In turn, the higher level of
collective efficacy will yield higher level of public
health by providing residents with health infrastruc-
ture, the trust necessary to implement community
health programmes and a greater capacity for social
control of negative health behaviours.
The literature establishes two competing hypoth-
eses. First, the income model provides that a higher
concentration of small businesses represents dimin-
ished access to health care because the small busi-
nesses are less likely to offer health benefits to
employees. Therefore, counties with a greater con-
centration of small businesses should have higher
rates of poor health. The alternative is the entrepre-
neurial culture argument. The small-business sector
is part of the locally oriented, civically engaged in-
dependent middle class. We argue that this creates
greater collective efficacy, which in turn leads to
positive health outcomes, for reasons mentioned
above. From the entrepreneurial culture argument,
we expect to find that counties with a greater pro-
portion of small businesses will have lower rates of
poor health.
Data and methods
We test our hypotheses using the 2000 Census of
Population and Housing, Summary Files 3, the
2002 County Business Patterns, 2002 Nonemployer
Statistics, the 2007 Centers for Disease Control
Obesity and Diabetes Estimates, and the National
Center for Health Statistics Compressed Mortality
records from 1994 to 2006. The units of analysis for
our study are 3060 counties in the contiguous USA.
In the event that a county contained an independent
city, data for the independent city and county were
combined to a single-county unit.
Dependent variables
The dependent variables in our analysis include the
age-standardized mortality rate from 2000 to 2006,
the percent of the population that is classified as
obese and the percent of the population diagnosed
with diabetes. For the mortality rate, we examine
age-standardized rates using the 2000 US popula-
tion as our standard. Standardization purges the
mortality rate of differences due to unevenness in
the distribution of age across counties.
Health and wealth of US counties
5of14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
Our direct standardization of the mortality rate by
age takes the following form:
m=PmaPa
P
3100;000;ð1Þ
where mrefers to the age-standardized mortality
rate per 100,000 persons, m
a
is the age-specific
mortality rate in a given county, P
a
is the number
of persons in the standard population (2000 US
population) for a given age group and Pis the total
2000 US population. Age-specific mortality rates
(m
a
) are obtained by dividing the number of deaths
in a county for a given age, sex and race group by
the number of persons in the county that are in that
group. We then obtained the average standardized
rate across the 7-year period.
Independent variables
The independent variables in our analysis include
measures of the business structure in the commu-
nity and control variables. Our measures of the
business structure draw on data from the 2002
County Business Patterns and the 2002 Nonem-
ployer Statistics. We operationalize entrepreneurial
culture as the number of small businesses per
100,000 people. We classify small businesses as
those business establishments with zero to four
employees. We also include measures of the num-
ber of large manufacturing and the number of large
retail establishments per 100,000 people. Large
manufacturing establishments are defined as
manufacturing establishments with 500 or more
employees. Large retailers include those retailers
with 100 or more employees.
Recent work on the importance of the cultural
economy in regional economic development sug-
gests that a higher concentration of creative class
occupations, such as education, law, health care,
arts, culture, entertainment and science/engineering
(to name a few), are vital to regional economic
growth and development (McGranahan et al.,
2011). This creative class is composed of highly
skilled individuals with extensive entrepreneurial
capabilities (see Florida et al., 2008, 616). The cre-
ative class consumes locally produced goods and
services and generates products and services that
have export capacity (see Markusen and Schrock,
2006). In addition to being creative and innovative,
this class is a comparative advantage for locales. It
signals to other highly skilled talent that a particular
place is a good place to live and work.
We control for the percent of the workforce in
each county in 2000 employed in the creative class
and the bohemian class (arts) occupations. Creative
class occupations include workers employed as
top executives, managers (advertising, marketing,
promotions, public relations and sales), financial
managers, operations specialities managers (except
financial managers), other management occupations
(except farmers and farm managers), business
operations specialists, other financial specialists,
computer specialists, mathematical science occupa-
tions, architects, surveyors, cartographers, engineers,
drafters, technicians (engineering and mapping), life
and physical scientists, social scientists and related
workers, postsecondary teachers, librarians, curators,
archivists, media and communications workers, sales
representatives (services, wholesale and manufactur-
ing), other sales and related occupations (including
supervisors). Bohemian occupations include art/
design workers and entertainers (performers, sports
and related workers).
These occupational classifications are provided
by the Economic Research Service (ERS), U.S.
Department of Agriculture (www.ers.usda.gov/
data/creativeclasscodes/, see also McGranahan
and Wojan, 2007). Given that creative class mem-
bers are entrepreneurial, highly skilled and innova-
tive, we need to make sure that our small-business
measure of entrepreneurial culture is not in actuality
capturing the effects of the local creative class.
Moreover, it is sound theoretical reasoning to ex-
pect that these two measures have the same effects
on health outcomes. Perhaps, the creative class is
composed of healthier people. It may also be that
the creative class helps to create the type of collec-
tive efficacy that will lead to healthier places.
With the exception of health insurance coverage
and creative class measures, our choice of variables
is based on prior research on mortality rates for
counties (see McLaughlin et al., 2001). We measure
the level of health insurance coverage in the
Blanchard, Tolbert and Mencken
6of14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
population by including a control for the percent of
the population without health insurance. This mea-
sure comes from the U.S. Census Bureau Small
Area Health Insurance Estimates Program. The
models also include control variables for health
insurance coverage, per capita income, income in-
equality, minority concentration, health infrastruc-
ture, metropolitan status and population size.
Per capita income is measured from the 2000
Census of Population and Housing using total
household income per person. We also control for
the percentage of the population 65 years and older.
Following prior research, we measure income in-
equality using the Gini coefficient of inequality
(McLaughlin et al., 2001). We include a measure
of the percent of the county population that is
non-Hispanic Black and the percent Hispanic to
account for the effects of minority concentration
noted in prior sociological analyses of mortality
(LaVeist, 1992;McLaughlin and Stokes, 2002).
The local health infrastructure is operationalized
as the number of physicians per 1000 residents in
2002. Metropolitan status is classified using
the 2003 Urban Influence Codes (ERS 2003:
http://www.ers.usda.gov/data/urbaninfluencecodes/
2003/). We control for metropolitan status because
the availability of health-care facilities and profes-
sionals varies substantially across rural and urban
localities (Pathman et al., 2004). We also include
a control for the natural logarithm of the county
population size. Descriptive statistics for the varia-
bles in the analysis are reported in Table 1.
Analytical technique
To test our hypothesis regarding the effect of the
entrepreneurial culture on population health out-
comes, we estimate fixed-effect weighted least
squares (WLS) regression models that include
a time-lagged measure of the level of population
health in the county. We employ a WLS technique
where models are weighted by the population size.
We do so because ordinary least squares regression
Table 1. Descriptive statistics
Mean SD
Entrepreneurial culture (establishments per 100,000 people)
Number of businesses with 0–4 employees 7603.560 2364.000
Other businesses (establishments per 100,000 people)
Number of large manufacturers (500+employees) 1.587 2.913
Number of large retailers (100+employees) 4.586 4.845
Creative occupation measures
Percent of labour force employed in creative class occupations 17.300 5.900
Percent of labour force employed in Bohemian class occupations 0.700 0.400
Controls
Percent of population uninsured 17.754 7.320
Income inequality (Gini coefficient) 42.416 3.488
Per capita income 17451.820 3883.940
Percent of population non-Hispanic Black 8.593 14.402
Percent of population Hispanic 6.156 12.135
Number of physicians per 100,000 persons 151.065 319.587
Population size, natural log 10.242 1.407
Percent of population aged 65 and older (2000) 14.750 4.230
County part of metropolitan area (1 =yes, 0 =no) 0.344 0.475
Dependent variables
Age-adjusted mortality rate (deaths per 100,000), 2000–2006 876.370 130.939
Age-adjusted mortality rate (deaths per 100,000), 1994–1998 922.687 128.182
Percent obese, 2007 28.287 3.623
Percent with diabetes, 2007 9.645 2.009
Health and wealth of US counties
7of14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
weights all counties equally such that a county with
1,000,000 residents exerts the same influence in the
model as a county with a population of 5000. In
supplementary analyses, we estimated models with
and without weights and obtained similar findings.
Our fixed-effect models include binary variables
for each state. We utilize a fixed-effect approach be-
cause our exploratory analyses suggested that average
levels of population health varied substantially across
states. For example, Louisiana, Mississippi and West
Virginia contain large clusters of counties with excep-
tionally high rates of mortality, obesity and diabetes.
To account for possible endogeneity in our anal-
yses, we include the 1994–1998 age-adjusted
mortality rate in all models. In models predicting
the percent obese and the percent with diabetes,
time-lagged measures (that is, measurements prior
to 2000) are not available. Including the time-
lagged measure of mortality provides a more
robust test of our hypothesis because our findings
are less likely to be a function of reverse causality
where healthier communities may select specific
population or businesses.
We also examined our regression models for
evidence of multicollinearity. Neter et al. (1989)
note that a variance inflation factor (VIF) value of
10 is evidence of problematic multicollinearity. No
coefficients related to our hypothesis tests had VIF
values that indicated collinearity. Three of the con-
trol variables in the model exhibited VIF values that
exceeded this threshold. These occurred for the var-
iables percent uninsured (10.0), per capita income
(10.5) and the percent employed in creative class
occupations (13.6). We evaluated a number of model
specifications where each of the three variables were
omitted to assess the consistency in the model coef-
ficients and standard errors. In all models, the central
findings of the analysis were consistent with those
presented here.
Results
In our analyses, we test two hypotheses about
county-level health measures. We proposed that
the presence of a strong entrepreneurial culture,
measured as the number of small businesses with
zero to four employees, would be associated with
healthier communities. We find support for this ar-
gument. The concentration of small businesses is
associated with lower rates of mortality, obesity
and diabetes (see Table 2). We obtain this finding
after accounting for between-state variations in the
baseline level of the dependent variables in addition
to the control variables in the model. Moreover, this
finding holds after adjusting for the lagged age-
adjusted mortality rate from 1994 to 1998.
In addition to statistical significance, the coeffi-
cient for our small-business measure has a large
effect in each of the models. In the model predicting
the mortality rate, the standardized coefficient for
our small-business measure is exceeded only by the
standardized coefficients for the race distribution
(percent non-Hispanic Black and percent Hispanic),
the percent of the labour force employed in creative
class occupations and the lagged value of the
dependent variable. For the obesity model, our
small-business variable has the largest standardized
coefficient in the model. This is an important find-
ing given the salience of obesity as a persistent
social problem in the USA. In the model predicting
diabetes, the standardized coefficient for the small-
business measure is exceeded by the standardized
coefficients for the creative class, percent unin-
sured, per capita income, percent non-Hispanic
Black, percent Hispanic and the percent of the
population aged 65 and older.
These findings provide robust support for our hy-
pothesis that a small-business sector is important for
local development and that these effects extend be-
yond job growth. Moreover, these effects are signif-
icant, controlling for two measures of creative class
concentration. Both of these measures (proportion
employed in creative class occupations, proportion
employed in bohemian occupations) have significant
effects: the greater the concentration of both, the
lower the level of obesity, death rate and diabetes.
Community health is not just the types of jobs that
are located there but also the structure of the business
sector. A place with a greater proportion of small
businesses will have a healthier population.
The data also show some support for the income
model of community health. The percent of the
Blanchard, Tolbert and Mencken
8of14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
Table 2. Fixed-effect regression estimates predicting age-adjusted mortality rate (2000–2006), percent of adults obese (2007) and percent of adults with diabetes (2007)
Age-adjusted mortality Rate percent obese Percent diabetic
bs.e. bbs.e. bbs.e. b
Entrepreneurial culture (establishments
per 100,000 people)
Number of businesses with 0–4 employees 0.004 *** 0.00060 0.074 0.0005 *** 0.00002 0.229 0.0001 *** 0.00001 0.126
Other businesses (establishments per 100,000
people)
Number of large manufacturers (500+
employees)
0.665 0.39990 0.011 0.0055 0.01790 0.002 0.0105 0.00800 0.010
Number of large retailers (100+employees) 1.647 *** 0.22600 0.061 0.0872 *** 0.01020 0.086 0.0036 0.00400 0.008
Creative class occupations
Proportion creative class 2.344 *** 0.30300 0.157 0.112 *** 0.01300 0.203 0.073 *** 0.00600 0.308
Proportion Bohemian class 10.17 *** 1.88000 0.058 0.249 * 0.08500 0.039 0.089 * 0.03800 0.032
Controls
Percent uninsured 0.226 0.28600 0.014 0.107 *** 0.01290 0.176 0.0586 *** 0.00500 0.222
Income inequality (Gini coefficient) 2.122 *** 0.36300 0.067 0.1378 *** 0.01630 0.117 0.022 *** 0.00700 0.043
Per capita income 0.00004 0.00100 0.002 0.0003 *** 0.00000 0.051 0.00003 * 0.00001 0.187
Percent non-Hispanic Black 0.998 *** 0.09200 0.118 0.0392 *** 0.00410 0.125 0.0245 *** 0.00100 0.181
Percent Hispanic 1.395 *** 0.11100 0.194 0.0488 *** 0.00500 0.183 0.0241 *** 0.00100 0.209
Number of physicians per 100,000 persons 0.0024 0.00410 0.005 0.0007 *** 0.00010 0.039 0.0002 *** 0.00008 0.031
Ln population size 0.2818 0.83800 0.004 0.289 *** 0.03700 0.117 0.0728 *** 0.01700 0.068
County part of metropolitan area 1.872 2.36100 0.007 0.151 0.10600 0.014 0.071 0.04700 0.015
Age-adjusted mortality rate, 1994–1998 0.8122 *** 0.01120 0.742 0.004 *** 0.00050 0.108 0.0019 *** 0.00020 0.111
Percent 65+1.641 *** 0.30000 0.053 0.0531 *** 0.01360 0.047 0.1181 *** 0.00600 0.240
Intercept
a
167.290 *** 21.80000 — 35.797 *** 0.98200 — 4.904 *** 0.51000 —
N3059 3060 3060
R
2
0.91 0.87 0.85
a
Model includes binary variables for each state.
*p<0.05, **p<0.01, ***p<0.001.
Health and wealth of US counties
9of14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
population that is not insured is associated with
higher levels of obesity and diabetes. Increases in
per capita income are associated with improved
population health via lower levels of obesity and
diabetes. We are not arguing that the income model
is irrelevant. However, beyond the explanatory
power of variables like health insurance and in-
come, the nature of the business sector does matter.
Our models also yield a number of important find-
ings regarding the association between the presence
of large establishments and population health. First,
the number of large manufacturing establishments
per 100,000 persons is not significantly associated
with any of our measures of population health. This
finding is interesting because to our knowledge
prior studies have not explored the link between
the presence of large enterprise and population
health outcomes.
A second important finding in our analyses is that
the presence of large retailers has a detrimental ef-
fect on two population health measures. In terms of
the age-adjusted rate of mortality and the percent of
adults obese, the presence of large retailers is
related to poorer health outcomes. Large big box
retail operations are not conducive to improving
community health, net of other important business
sector and demographic controls.
With few exceptions, our control variables gen-
erally conform to findings from prior studies. Coun-
ties with a large presence of physicians per 100,000
persons enjoy less obesity and diabetes. We also
find that the percent of the population that is
Hispanic is negatively associated with mortality,
obesity and diabetes rates. This finding is consistent
with prior studies that find a Latino health paradox
where despite lower levels of socio-economic well-
being, health among Hispanics is similar to that of
non-Hispanic whites (for a review, see Franzini
et al., 2001).
However, the percent non-Hispanic Black yields
inconsistent findings. Prior studies have observed
that the percent non-Hispanic Black is associated
with lower levels of population health. Model
results for the percent obese and percent with di-
abetes largely conform to this trend. However,
counties with a high percent non-Hispanic Blacks
have lower rates of age-adjusted rates of mortality.
In supplementary analyses, we observed that the
direction of this coefficient changed from positive
to negative after controlling for the time-lagged
value of the mortality rate. This finding suggests
that after adjusting for prior levels of mortality,
the percent non-Hispanic Black is actually related
to improved population health outcomes. Though
not a common finding, we identified one study
(Blanchard et al., 2004) that found a similar finding
for the rural South. Moreover, prior studies have not
included time-lagged values of the dependent
variable to account for possible endogeneity.
A second inconsistent finding in our models is
that income inequality, as measured by the Gini
coefficient, is associated with lower rates of obesity
and diabetes but higher rates of age-adjusted mor-
tality. We found similar results using alternative
model specifications that did not include fixed
effects or the time-lagged measure of mortality.
We also estimated models with categorical meas-
ures of the Gini coefficient to test for nonlinearities
in the effect and obtained the same results.
Conclusions
In this study, we draw on the civic community
tradition to develop an alternative framework for
understanding the linkage between economic
development and health. We argue that economic
development strategies aimed at growing the small-
business sector have a latent beneficial effect on
health outcomes. Our findings support a small
businesses perspective. We find that counties with
a vibrant small-business sector have lower rates of
mortality and a lower prevalence of obesity and
diabetes. Small-business owners produce important
noneconomic rewards for communities, such as
enhanced stocks of social capital and collective
efficacy. In this way, the small-business sector
may produce salutary rather than unfavourable
community health outcomes.
In the past, communities largely sought to de-
velop using external investment. More recently,
community development experts have shifted focus
to investments in local infrastructure that foster
Blanchard, Tolbert and Mencken
10 of 14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
small-business growth (Edmiston, 2007). With re-
spect to health, this approach is counter intuitive
because small businesses generate lower-paying
jobs with fewer benefits. Theories of modernization
and development largely posit that growth in pro-
ductivity yields improved health conditions through
expansion of payrolls, which also provides access
to health insurance, the gatekeeper for high-quality
medical care in the USA. We are not arguing that
the income model is without validity. Our own
shows that income-related measures (per capita in-
come, percent uninsured, income inequality) do
have predicted effects on some health indicators.
However, the small-business effects in our models
point out some potential limitations in the income
model approach.
We posit an alternative theoretical approach
that explains how small-business development
improves population health by enhancing the en-
trepreneurial culture of the community. Entrepre-
neurial culture provides a local orientation that
allows for greater levels of interaction and trust
among community members. This, in turn, helps
to create collective efficacy, which has positive
effects on community health in a number of ways
discussed previously in the paper. In the publically
available county-level data, valid measures of col-
lective efficacy are non-existent. Therefore, we
have to assume that the process of entrepreneurial
culture works as we described it in this paper.
However, we show that effects of small-business
concentration are robust in each model and are net
of important measures from income-based models
(percent uninsured, per capita income, concentra-
tion of large employers). Small businesses matter,
and we argue that they matter in the way proposed
in the hypothesis. However, future research with
intervening measures of collective efficacy is nec-
essary to confirm our conclusions. Furthermore,
while we control for percent uninsured, a better
measure would be the types of health insurance
(if any) the small businesses provide. These data
are not available from government sources.
Perhaps, future studies can do community case
studies that include this potentially important
measure.
Another key contribution of this paper is to
highlight the role of health as an indicator of
development. Recently, researchers have begun to
examine the reciprocal relationship between health
and economic development (Bloom et al., 2004).
Studies suggest that health may be not only a good
indicator of community well-being but also an en-
gine to drive productivity. Though not addressed in
our analyses, future analyses should consider the de-
gree to which a healthy population is better able to
produce nascent entrepreneurs.
More importantly, our analysis demonstrates that
investment in locally grown enterprise has the
potential to yield large returns for communities.
Fostering an entrepreneurial culture in a community
may stimulate self-determinism and proactivity by
local residents to manage local affairs and address
problems. This model of social organization stands
in stark contrast to the orientation of communities
dominated by a handful of large employers. In these
communities, residents are less likely to be in-
volved in community affairs such as voting, partici-
pating in rallies, signing petitions and the like
(Blanchard and Matthews, 2006). In addition to
health, we expect that our entrepreneurial culture
approach could be applied to a variety of indicators
of well-being, such as crime, suicide, population
growth and school performance. Moreover,
researchers also should explore the manner in
which the entrepreneurial culture may interact with
religious institutions that provide a theological
underpinning for entrepreneurship.
We also note that this paper has treated the
effects of small businesses as homogenous. Profes-
sional small businesses are qualitatively different
from non-professional small businesses. The for-
mer are often formed by local entrepreneurs seeking
greater earnings that would be typically offered
through a corporation (that is, accountant, legal
services). The latter is often characterized by indi-
viduals starting their own business because they are
having a difficult time locating stable employment
with larger businesses (Boden, 1999;Budig, 2006).
In addition, small non-professional businesses
owned by women may be significantly different
(and less profitable) than those operated by men
Health and wealth of US counties
11 of 14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
(that is, day-care centre versus appliance repair).
Unfortunately, there are disclosure issues at the
county level with many of the data we used in this
study. We are unable to tell which industries the
small employers are from. Therefore, future studies
should, when data permit, recognize the diversity of
operations within the small-business category.
While our study focused on the effects of the
small-business sector on community health out-
comes, the data from the creative class measures
show opportunities to integrate these two theoreti-
cal perspectives (civil society/entrepreneurial cul-
ture and creative class). Members of the creative
class may constitute part of the small-business class
in a local community; however, our analysis shows
that all three measures have expected and signifi-
cant effects on community health measures. This
suggests that small business and creative class mea-
sure different aspects of the community that all lead
to the same expected outcome. More exploration of
this interconnection is warranted.
Finally, we note that the results are based on
counties in the USA, a nation without a comprehen-
sive state health system. One might expect that sim-
ilar results do not apply in nations that have such
state-sponsored policies. However, these results
should be of interest to international scholars. First,
the results are more about collective efficacy and
not how one gets health care. A small-business
sector could lead to other prosocial outcomes
(for example, less delinquency) via collective effi-
cacy. Second, the small-business sector can be
helpful in promoting community health education
programmes, which lead to higher levels of health.
Finally, in a more draconian sense, the current fiscal
crisis that is gripping many nations, particularly in
the European Union, may lead to a reduction in
social benefits, including health care.
Acknowledgements
This paper is based on work supported by the U. S.
Department of Agriculture, National Research Initiative
under award number 2006-35401-18859, and by the So-
cial and Economic Sciences Division, National Science
Foundation under award number 0757201.
References
Amato, L. H., Amato, C. H. (2007) The effects of firm
size and industry on corporate giving. Journal of Busi-
ness Ethics,72: 229–241.
Araya, R., Dunstan, F., Playle, R., et al. (2006) Percep-
tions of social capital and the built environment and
mental health. Social Science and Medicine,62: 3072–
3083.
Arum, R., Mueller, W. (2004) The Reemergence of
Self-Employment: A Comparative Study of Self-
Employment Dynamics and Social Inequality. Prince-
ton, NJ: Princeton University Press.
Blanchard, T., Cossman, J., Levin, M. (2004) Multiple
meanings of minority concentration: incorporating con-
textual explanations into the analysis of individual
level US black mortality outcomes. Population Re-
search and Policy Review,23: 309–326.
Blanchard, T., Matthews, T. (2006) The configuration
of local economic power and civic participation
in the global economy. Social Forces,84:
2241–2257.
Bloom, D., Canning, D., Jamison, D. (2004) Health,
wealth, and welfare. Finance and Development,24:
10–15.
Boden, R. J. (1999) Flexible working hours, family
responsibilities and female self-employment:
gender differences in self-employment selection.
American Journal of Economics and Sociology,
58: 71–83.
Budig, M. (2006) Intersections on the road to self-
employment: gender, family, and occupational class.
Social Forces,84: 2223–2239.
Cobb, J. (1982) The Selling of the South: The Southern
Crusade for Industrial Development. Champaign, IL:
University of Illinois Press.
Cohen, D. A., Finch, B. K., Bower, A., et al. (2006)
Collective efficacy and obesity: the potential influence
of social factors on health. Social Science and Medi-
cine,62: 769–778.
Cubbins, L. A., Parmer, P. (2001) Economic change and
health benefits: structural trends in employer-based
health insurance. Journal of Health and Social Behav-
ior,42: 45–63.
Daly, M. C., Duncan, G. J., Kaplan, G. A., et al. (1998)
Macro-to-micro links in the relation between income
inequality and mortality. The Milbank Quarterly,76:
315–339.
Dominguez, S., Arford, T. (2010) It’s all about who
you know: social capital and health in low-income
communities. Health Sociology Review,19: 114–129.
Dunn, T., Holtz-Eakin, D. (2000) Financial capital, hu-
man capital, and the transition to self-employment: ev-
idence from intergenerational links. Journal of Labor
Economics,18: 282–305.
Blanchard, Tolbert and Mencken
12 of 14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
Edmiston, K. (2007) The role of small and large busi-
nesses in economic development. Economic Review
Federal Reserve Bank of Kansas City,92: 73–97.
Ewing, R., Schmid, T., Killingsworth, R., et al. (2003)
Relationship between urban sprawl and physical activ-
ity, obesity, and morbidity. American Journal of
Health Promotion,18: 47–57.
Falk, W. W., Lobao, L. M. (2004) Who benefits from
restructuring? In Brown, D., Swanson, L. (eds). Chal-
lenges for Rural America in the Twenty-First Century,
pp. 152–165. State College, PA: The Pennsylvania
State University Press
Florida, R., Mellander, C., Stolarick, K. (2008) Inside the
black box of regional development: human capital, the
creative class, and tolerance. Journal of Economic Ge-
ography,8: 615–649.
Folland, S. (2007) Does ‘‘community social capital’’
contribute to population health. Social Science and
Medicine,64: 2342–2354.
Franzini, L., Ribble, J. C., Keddie, A. M. (2001) Under-
standing the Hispanic paradox. Ethnicity and Disease,
11: 496–518.
Gortmaker, S. L., Wise, P. H. (1997) The first injustice:
socioeconomic disparities, health services technology,
and infant mortality. Annual Review of Sociology,23:
147–170.
Granovetter, M. S., Tilly, C. (1988) Inequality and labor
processes. In Smelser, N. (ed.) Handbook of Sociology,
pp. 175–221. Newbury Park, CA: Sage Publications.
Green, G., Flora, J., Flora, C., et al. (1990) Local self-
development strategies: national survey results. Jour-
nal of the Community Development Society,21: 55–73.
Green, L., Daniel, M., Novick, L. (2001) Partnerships and
coalitions for community-based research. Public
Health Reports,116 (Suppl 1): 20–31. Centers for Dis-
ease Control.
Gutierrez, L. M., Lewis, E. A. (1999) Empowering Women
of Color. New York: Columbia University.
Haltiwanger, J., Davis, S., Krizan, C. J. et al. (2009)
Measuring the dynamics of young and small business:
integrating the employer and non-employer businesses.
In Dunne, T., Jensen, J., Roberts, M. (eds.) Producer
Dynamics: New Evidence from Micro Data, NBER/
University of Chicago Press.
Hawe, P., Shiell, A. (2000) Social capital and health pro-
motion: a review. Social Science and Medicine,51:
871–885.
Hollister, M. N. (2004) Does firm size matter anymore?
The new economy and firm size wage effects. Ameri-
can Sociological Review,69: 659–676.
Hout, M., Rosen, H. (2000) Self-employment, family
background, and race. Journal of Human Resources,
35: 670–692.
Islam, K. M., Merlo, J., Kawachi, I., et al. (2006) Social
capital and health: does egalitarianism matter? A liter-
ature review. International Journal for Equity in
Health,5: 1–28.
Isserman, A. (1994) State economic development policy
and practice in the United States: a survey article. In-
ternational Regional Science Review,16: 49–100.
Kalleberg, A., Van Buren, M. E. (1996) Is bigger better?
Explaining the relationship between organizational size
and job rewards. American Sociological Review,61:
47–66.
Keene, J. R., Prokos, A. H. (2010) Gendered disparities in
take-ups of employee health benefits. Sociological Per-
spectives,53: 503–526.
Korsching, P., Allen, J. (2004) Locality based entrepre-
neurship: a strategy for community economic vitality.
Community Development Journal,39: 385–400.
LaVeist, T. A. (1992) The political empowerment and
health status of African-Americans: mapping a new
territory. American Journal of Sociology,97: 1080–
1095.
Lynch, J., Smith, G. D., Harper, S., et al. (2004) Is income
inequality a determinant of population health? Part 1. A
systematic review. The Milbank Quarterly,82: 5–99.
Lynch, J. W., Smith, G. D., Kaplan, G. A., et al. (2000)
Income inequality and mortality: importance to health
of individual income, psychosocial environment, or
material conditions. British Medical Journal,320:
1200–1204.
Lyson, T., Torres, R., Welsh, R. (2001) Scale of agricul-
tural production, civic engagement, and community
welfare. Social Forces,80: 311–327.
Macinko, J., Starfield, B. (2001) The utility of social
capital in research of health determinants. The Milbank
Quarterly,79: 387–427.
Markusen, A., Schrock, G. (2006) The artistic dividend:
urban artistic specialisation and economic development
implications. Urban Studies,43: 1661–1686.
McGranahan, D., Wojan, T. R. (2007) Recasting the cre-
ative class to examine growth processes in rural and
urban counties. Regional Studies,41: 197–216.
McGranahan, D., Wojan, T. R., Lambert, D. M. (2011)
The rural trifecta: outdoor amenities, creative class, and
entrepreneurial context. Journal of Economic Geogra-
phy,11: 529–557.
McKeown, T. (1976) The Modern Rise of Population.
London: Edward Arnold.
McLaughlin, D., Stokes, S., Nonoyama, A. (2001) Res-
idence and income inequality: effect on mortality
among U.S. counties. Rural Sociology,66: 579–598.
McLaughlin, D. K., Stokes, C. S. (2002) Income inequal-
ity and mortality in US counties: does minority racial
concentration matter? American Journal of Public
Health,92: 99–104.
Mencken, F. C., Winfield, I. (1998) In search of the
‘‘right stuff’’: the advantages and disadvantages of in-
formal and formal recruiting practices in external labor
Health and wealth of US counties
13 of 14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
markets. American Journal of Economics and
Sociology,57: 135–153.
Mills, C. W., Ulmer, M. (1946) Small Business and Civic
Welfare. Report of the Smaller War Plants Corporation
to the Special Committee to Study Problems of Amer-
ican Small Business. Document 135. U.S. Senate, 79th
Congress, 2nd Session, February 13. Washington, DC:
U.S. Government Printing Office.
Mitchell, S. (2006) Big-Box Swindle: The True Cost of
Mega-Retailers and the Fight for America’s Indepen-
dent Businesses. Boston, MA: Beacon Press.
Neter, J., Wasserman, W., Kutner, M. (1989) Applied
Linear Regression Models. Homewood, IL: Irwin.
Neumark, D., Wall, B., Zhang, J. (2008) Do Small Busi-
nesses Create More Jobs? New Evidence from the Na-
tional Establishment Time Series. NBER Working
Paper No. 13818. Cambridge, MA: National Bureau
of Economic Research.
Olafsdottir, S. (2007) Fundamental causes of health dis-
parities: stratification, the welfare state, and health in
the United States and Iceland. Journal of Health and
Social Behavior,48: 239–253.
Pathman, D. E., Konrad, T. R., Dann, R., et al. (2004)
Retention of primary care physicians in rural health
professional shortage areas. American Journal of Pub-
lic Health,94: 1723–1729.
Payne, P., Williams, K. R. (2008) Building social capital
through neighborhood mobilization: challenges and
lessons learned. Journal of Preventive Medicine,34:
S42–S47.
Preston, S. (1975) The changing relationship between
mortality and level of economic development. Popula-
tion Studies,29: 48–55.
Ross, C., Mirowsky, J. (2001) Neighborhood disadvan-
tage, disorder, and health. Journal of Health and Social
Behavior,42: 258–276.
Sampson, R. J., Morenoff, J. D., Earls, F. (1999) Beyond
social capital: spatial dynamics of collective efficacy for
children. American Sociological Review,64: 633–660.
Sampson, R. J., Raudenbush, S. W., Earls, F. (1997)
Neighborhoods and violent crime: a multilevel study
of collective efficacy. Science,277: 918–924.
Stebbins, K. (1997) Clearing the air: challenges to intro-
ducing smoking restrictions in West Virginia. Social
Science and Medicine,44: 1393–1401.
Sullivan, T. A., Warren, E., Westbrook, J. L. (2000) The
Fragile Middle Class: Americans in Debt. New haven,
CT: Yale University Press.
Tolbert, C., Irwin, M., Lyson, T., et al. (2002) Civic
community in small town America: how civic welfare
is influenced by local capitalism and civic engagement.
Rural Sociology,67: 90–113.
Tolbert, C., Lyson, T., Irwin, M. (1998) Local capitalism,
civic engagement, and socioeconomic well-being. So-
cial Forces,77: 401–427.
Tolbert, C. M., Horan, P. M., Beck, E. M. (1980) The
structure of economic segmentation: a dual economy ap-
proach. American Journal of Sociology,85: 1095–1116.
U.S. Census Bureau, Statistics of U.S. Businesses
(SUSB). (1990–2003) Employer Firm Births and
Deaths by Employment Size of Firm. Available
online at: http://www.sba.gov/advo/research/dyn_b_
d8906.pdf [Accessed 30 August 2010].
Useem, M. (1988) Market and institutional factors in cor-
porate contributions. California Management Review,
30: 77–88.
Variyam, J. N., Kraybill, D. S. (1998) Fringe benefits
provision by rural small businesses. American Journal
of Agricultural Economics,80: 360–368.
Villemez, W. J., Bridges, W. P. (1988) When bigger is
better: differences in the individual effect of firm and
establishment size. American Sociological Review,53:
237–255.
Blanchard, Tolbert and Mencken
14 of 14
at Baylor University on January 18, 2012http://cjres.oxfordjournals.org/Downloaded from
