The End is Nigh:
Limits to Growth of the Nonprofit Sector
DRAFT – PLEASE DO NOT CITE
JESSE D. LECY
Assistant Professor of Public Policy
Georgia State University
ERIC J. CHISHOLM
PhD Student in Public Policy
Georgia State University
OCTOBER, 2013
ABSTRACT: The nonprofit sector has experienced exponential growth over the past three decades
with nearly 50,000 new nonprofits created last year. Past examples of industry growth suggest that
this rate of growth is not sustainable. Empirical population ecology studies of nascent industries
show a period of rapid growth followed by market saturation, then consolidation of organizations
and market share resulting in increased competition for small and new organizations. We use
historical nonprofit data from the NCCS and apply ecological models to show that the nonprofit
sector may be fast approaching growth limits. Market saturation varies by metropolitan area and
nonprofit subsector.
INTRODUCTION
The nonprofit sector has sustained tremendous expansion over the past half a century with growth
rates exceeding both business and government sectors (Roeger, Blackwood et al. 2012). This
expansion has been important as nonprofits have been a major source of economic and
employment growth (Salamon, Sokolowski et al. 2012). We argue here that these high growth
rates in the nonprofit sector will soon cease. Using data over a twenty-year time frame we show
that market exit rates are converging with entry rates, which will result in slow and sometimes
negative industry growth. The nonprofit sector is approaching a period of saturated markets where
resources are spread thin and new organizations struggle for survival.
Many notable studies of nonprofit markets have focused on entry rates of new nonprofits (Corbin
1999, Saxton and Benson 2005, Seokki 2010, Stretesky, Huss et al. 2011, Harrison and Thornton
2012, Lecy and Van Slyke 2013), factors that predict organizational vulnerability and demise
(Galaskiewicz and Bielefeld 1998, Hager 2001, Hager, Galaskiewicz et al. 2004, Duckies, Hager et
al. 2005, Dougherty, Maier et al. 2008, Fernandez 2008, Carroll and Stater 2009, Pandey, Sjoquist
et al. 2009, Wollebaek 2009, Maier 2010, Vance 2010, Walker and McCarthy 2010, Burger and
Owens 2013), and overall growth of the sector (Grønbjerg and Paarlberg 2001, Matsunaga and
Yamauchi 2004, Harrison and Laincz 2008, Luksetich 2008, Armsworth, Fishburn et al. 2012).
These studies have by and large treated entry, exit and growth as stable or equilibrium processes,
not dynamic market processes. Research on the aging process of industries shows us that these
rates should naturally change over time, often in predictable ways (Carroll and Hannan 1995). We
are the first to track the convergence of the entry and exit rates in the nonprofit sector to examine
growth as a dynamic process over time that will be significantly slowed by markets saturation.1
Figure 1: Rates of entry, exit and growth for nonprofits between 1989 and 2008. Source: NCCS
Core Trend datafiles.
1
Harrison and Laincz (2008) reported on trends in rising exit rates but their analysis concluded that failure rates of
nonprofits were low compared to for-profit counterparts and they did not discuss long-term implications of market
saturation.
We will draw upon several literatures on aging industries and market dynamics to consider
potential scenarios that might emerge as a result of nonprofit market saturation. By market
saturation we mean specifically that a market cannot support more nonprofits, i.e. that entry rates
equal exit rates. Market saturation does not imply that demand for nonprofit services has been met.
The analysis is approached from a supply-side perspective since carrying capacity is tied more to
funding sources for nonprofits than it is to demand for services (Lecy and Van Slyke 2013; Mirae
Kim; Pevcin 2012). There is also an important distinction between market capacity (the number of
nonprofits that a market can support), and sector capacity (the total services provided, often
proxied by the total spending on programs). It is feasible for a market to reach capacity but sector
capacity to continue expansion because of growth of existing nonprofits.
We contribute to the understanding of nonprofit market forces by demonstrating a clear trend
towards market saturation across all major metropolitan areas. There is, however, heterogeneity in
convergence rates across metro areas. Some cities have already entered periods of slow or negative
growth; others are still expanding fairly rapidly. We devote the second part of the paper to
exploring factors associated with stable or declining sector growth rates across metropolitan areas.
In the last part of the paper we consider the implications of this increased competition on
nonprofits. We track the overall decrease in HHI measures over time, which suggests that
nonprofit markets are becoming more competitive at the same time they become more saturated.
This is good in the sense that we do not see evidence that might suggest monopolistic behavior. It
does, however, raise questions about the changing nature of competition in the sector. We examine
the relationship between nonprofit size and exit rates to demonstrate that large nonprofits
experience less failure than small, new nonprofits.
OVERVIEW OF GROWTH OF THE NONPROFIT SECTOR
The nonprofit sector is sizable: there are currently an estimated 2.3 million nonprofits operating in
the US and 1.3 million registered with the IRS. There are approximately 64,000 nonprofits with
revenues above $10 million each year, and sector revenues accounted for $1.51 trillion dollars in
economic activity in 2012. Collectively nonprofits control $2.71 trillion in assets (Roeger,
Blackwood et al. 2012).
Mass mobilization during war periods have always had a stimulating effect on civil society
(Skocpol 2013), and World War II was no different. Almost two decades later the major civil
society transition from membership organizations and social clubs to professionalized and
advocacy-focused organizations began with significant social movements in the 1960’s including
women’s rights, civil rights, and later on the environmental movement (Skocpol 2013). The birth
of the modern safety net following WWII and the War on Poverty created demand for new social
service organizations. The inter-dependence between nonprofits and government programs
deepened in the 1980’s with the hollowing out of the state which shifted government capacity from
direct service provision towards contracts with private service-providers, many of which were
nonprofits. All of these trends had significant impacts on the growth and expansion of the
nonprofit sector so that by 2010 there were over 50,000 new nonprofit 501(c)(3) statuses being
granted by the IRS each year.
Figure X: Number of new nonprofits granted 501(c)(3) status each year. Source: The 2011 NCCS
Business Master File.
The expansion of the nonprofit sector over the past fifty years has varied by subsector.
Environmental nonprofits are fairly new to the game, for example, so that subsector has been
expanding faster than other subsectors over the past two decades (Straughan 2008). Although rates
vary a great deal, growth over all subsectors in the past twenty years has been significant. It is for
this reason that nonprofits have added employment at a faster rate that the government or private
sectors, even during periods of economic downturn (Salamon, Sokolowski et al. 2012).
Figure X: Sector growth over a twenty-year period. Source: NCCS 2012 Core Trend file.
Based upon this impressive performance and the apparent upward trajectory of sector size it is easy
to see why scholars have been optimistic about persistent nonprofit growth. Total sector size can
be deceiving, though, since by the peak of any industry the process of decline is already underway.
Organizational ecology shows us that past growth is not a good predictor of future growth of
markets, and the dynamics of entry, exit, and market concentration are paramount to understanding
expansion or contraction of an industry. Some literature on market dynamics in aging industries is
discussed next.
THEORIES OF MARKET CONCENTRATION AND INDUSTRY SHAKEOUT
Our analysis of industry dynamics builds upon the ecological models of organizations developed
by Hannan, Freeman, Carroll, and collaborators (Aldrich and Pfeffer 1976, Hannan and Freeman
1977, Carroll 1984, Hannan and Freeman 1993, Carroll and Hannan 1995). In this work they ask
the question, what do industries look like as they are created, mature, and age and how do these
industry environments affect the organizations operating within them? Organizational ecology was
developed partly to challenge the dominant orthodoxy at the time that management practices were
largely responsible for the behavior and performance of firms. The ecological literature shows that
environments (i.e. markets) can have a large influence on performance of individual firms, and
also that industries evolve over time in large part through selection processes on populations of
organizations, which includes the introduction of innovation through new firms and the closure of
old firms, versus change occurring merely through adaptation of existing organizations. These
ecological models are complimentary to life-cycle studies of industry in economics and theories of
creative destruction and entrepreneurism (Schumpeter 1942, Nelson 1982, Agarwal and Gort
1996), but they tend to emphasize different mechanisms such as the legitimization process
(organizational ecology) versus technological transformation of production processes and cost
structures or the emergence of economies of scale (industrial economics), or a focus on
entrepreneurial processes and market control (business strategy).
Unlike equilibrium models of markets the emphasis of ecological models is the evolution of
industries over time. New industries are created through the emergence of new technologies and
demands in society. Nascent industries always face challenges in early phases as consumers are
unfamiliar with the product and the organizational form has yet to be legitimized. If the new
industry provides enough value to the customers and production is profitable then the industry
begins to expand. New markets can be extremely profitable, especially for equity investors as
firms grow, so early entrants can experience large returns. This fuels the interest of potential
competitors, entry rates rise, and the total size of the industry, measured by the number of firms,
grows rapidly. As new firms flood the market, though, the competition for existing and future
customers becomes fierce. Firms that are not able to establish a stable market niche or demonstrate
adequate returns on capital over time are forced to exit. A tipping point is reached when firm exits
exceed entrants, thus causing a decline in the number of firms in the market (Carroll and Hannan
1995, Agarwal and Gort 1996).
The decline in the number of firms should not be confused with industry collapse; the overall
industry capacity may still expand past the tipping point through growth of surviving firms and
industry concentration. For example, the number of computer manufacturing firms declined
rapidly following industry shakeout, but the overall production of computers continued to rise as
several firms expanded production and increased their market share (McClellan 1984). Similar
processes are observed in the railroad industry (firms consolidated while total miles of track
expanded), labor unions (membership rising despite fall in unions), and automobile manufacturing
after industry consolidation (Carroll and Hannan 1995).
The process described above has been called an industry shakeout (Jovanovic and MacDonald
1994, Klepper 1997), a process by which the total number of firms declines rapidly at a specific
point in time. In most instances it is related primarily to population dynamics and the industry lifecycle, but in specific instances it can be induced by an external shock as was the case of the
banking industry shakeout during the Great Depression (Walter 2005). These processes will vary
by industry as a result in differences in market size, production technologies (especially whether
economies of scale can be achieved), market concentration, and economic costs or barriers to
market entry (Bartelsman, Haltiwanger et al. 2004, Bartelsman, Scarpetta et al. 2005). Empirical
examples of market shakeout include the US tire industry (Carree and Thurik 2000), digital
markets (Day and Fein 2003), the German laser industry (Buenstorf 2007), labor unions (Hannan
and Freeman 1987), beer brewing (Horvath, Schivardi et al. 2001, Tremblay, Iwasaki et al. 2005),
and newspaper industries (Van Kranenburg, Palm et al. 1998, Van Kranenburg, Palm et al. 2002)
to name just a few. The cause of a shakeout process is a function both of declining entry rates as
potential firms recognize that the market is crowded or perhaps cannot attract enough investment
to enter, and failure rates increase as resources and customers become scarce.
Bonaccorsi (2000) points out that most of the work on industrial dynamics has focused on cases
where shakeout has occurred, but industries that grow with a non-shakeout pattern, which are the
majority of manufacturing industries, have not received as much scrutiny. There are examples of
steady industry expansion with a plateau of industry participants without the subsequent crash
including a study of the synthetic dye industry (Murmann and Homburg 2001) and the turboprop
engine industry (Bonaccorsi and Giuri 2000), petrochemicals, disposable diapers, and zippers
(Bonaccorsi and Giuri 2000). In many of these cases, specialization of process engineering and
marketing firms eroded the advantages of plant size and R&D capacity of incumbants, thus
preventing the high market concentration and subsequent shakeout that occurred in other industries
(Bonaccorsi and Giuri 2000).
As a result of this gap in the literature we do not have as well-formed models of slow growth as we
do of industry shakeout, but we can hypothesize what such a market should look like. Instead of a
rapid fall in entry rates followed by a rise in exit rates leading to an overall negative growth rate,
entry and exit rates could converge to a stable long-term growth rate (one that tracks the expansion
of the population or the economy, perhaps). This would result in a stable number of firms when the
number of new firms is equal to the number of exiting firms each year, or a steady expansion if the
growth rate is positive.
Figure XX, Theoretical vital rates in a steady-growth industry that does not experience shakeout.
Both scenarios above are theoretically interesting for the nonprofit sector. A shakeout period
would result in a decline in the number of nonprofits within communities with saturated markets.
In order to maintain sector capacity existing nonprofits would have to expand to fill gaps in
services, resulting in higher market concentration. This scenario is possible but not necessarily the
likely scenario. A stable population scenario could occur if the downward trend in entry and the
upward trend in exit both flatten out, resulting in a flat or slowly expanding population of
nonprofits. Both cases would have significant implications for donor policies, service capacity
within communities, and innovation.
MARKET CONCENTRATION
The idea of market concentration plays a central role in competition theory. The theory predicts
that as a market matures specific firms will gain competitive advantage through more efficient
production processes, economies of scale, improved quality, superior marketing, patents and
trademarks, or other strategic means. As this occurs they take market share from less competitive
firms, leading to high levels of market concentration. Concentration is usually measured by how
much of the total market is controlled by a small number of firms, the market often defined as the
total dollar amount of goods and services purchased. The HHI Index was developed as a way to
quantify market concentration through a simple scale that ranges from zero to one. Various studies
lead to useful heuristics on market competitiveness: HHI levels from 0.0 to 0.2 are considered to
be competitive, healthy markets. Levels from 0.2 to 0.4 are considered to be markets with high
concentration and potential for non-competitive behaviors that can interfere with market
efficiency. And levels over 0.4 are interpreted as quasi-monopolistic markets. Lower levels of
concentration are favorable, and are assumed to coincide with healthy market characteristics such
as low barriers to entry, efficient allocation of capital, implicit safeguards against rent-seeking
behavior, and ultimately conditions for the welfare-maximizing behavior of markets. Studies of
market concentration predict that higher concentration results in less competitive markets where a
few producers yield a high amount of power and generally have competitive advantages over
smaller firms. As a result we would expect higher concentrations to coincide with stagnant
markets, higher firm exit rates, and slower industry growth (in number of firms).
Theories of the industry life-cycle and market concentration are related in important ways.
Shakeout and market concentration are somewhat synonymous processes. The rapid drop in firms
occurs during shakeout because of market concentration – specific firms have control over an
industry and their competitive advantage prevents new firms from entering the market, thus
lowering entry rates, and also erodes return for existing firms, incentivizing exit. In some cases,
dominant firms can also gain market share through buyouts and takeover, another form of firm
exit. Stated succinctly, the causes of shakeout and market concentration are the same – market
consolidation that occurs because of some firms gaining dominance through competitive
advantage.
But industry maturation does not always lead to a market shakeout or increased market
concentration. Ecological models of markets offer an alternative explanation of changes in vital
rates. Just as a finite tract of land has a specifying carrying capacity for an animal population, an
industry embedded within an economy can only grow to a specific finite size before demand is
satiated. When a market becomes saturated customer acquisition becomes more challenging and
costly, therefore firm sustainability is more challenging and exit rates increase. This process will
happen in a perfectly competitive market, and is thus independent of the level of concentration.
Market size relative to total carrying capacity will predict exit rates above and beyond
concentration.
COMPETING HYPOTHESES
These two perspectives on competition, saturation versus concentration, lead to different testable
hypotheses:
H1: Increased market concentration should lead to less competitive markets,
competitiveness measured by lower entry rates and higher exit rates.
H2: Increased market size will result in higher levels of market saturation, thus leading to
lower entry rates and higher exit rates.
Market carrying capacity is not always known and is perhaps not even static (changing preference
or complimentary goods and services can increase demand). But we assume here that market
saturation will increase over time as the current market expands. This will be true as long as
carrying capacity (demand for nonprofits) is not expanding at a faster rate than markets size (total
supply of services), which is a reasonable assumption in most circumstances. As a result, market
age can be used as a proxy for market saturation with the added assumption that markets approach
their carrying capacity at more or less the same pace. Since market concentration will also increase
over time (empirically this has been the case in all of the shakeout examples that have been
mentioned), then market age can also be correlated with concentration. Stating the hypotheses
differently then:
H1: Increased market concentration will lead to lower entry and higher exit rates, even
when controlling for market age.
H2: Older markets will lead to lower entry and higher exit rates, even while controlling for
market concentration.
H3: Population growth will increase overall market capacity.
The remainder of this paper will focus on these three hypotheses. The next section introduces the
data and models used for the analysis, and we conclude with some implications of the results.
DATA AND METHODS
We are interested in the variation in nonprofit sector growth rates across metro areas. In this
section we present a basic, descriptive model that helps us understand the factors related to
nonprofit sector vital rates (entry, exit, and growth). We examine three OLS models, one of each
vital rate regressed on a set of factors related to nonprofit density. Specific attention will be paid to
population growth, sector age within a metro area, and market concentration as these related to the
stated hypotheses.
The Metropolitan Statistical Area (MSA) is used as the unit of analysis for all variables. MSAs are
geographic regions related to cities and their surrounding areas. They are not a legal administrative
division, but rather defined by the U.S. Office of Management and Budget (OMB) to describe a
region.
This analysis employs dependent variables measuring entry (births), exit (death), and growth rates
(entry minus exit) for the nonprofit sector in each MSA. Nonprofit data comes from the National
Center for Charitable Statistics (NCCS) Core Trend files, which contains financial data for all
nonprofits from 1989-2009. Births were calculated by capturing the earliest year each nonprofit
appears in the data; conversely, deaths were calculated by taking the last appearance for each
nonprofit. In order to account for the size of the sector, we then divided deaths and births by the
number of nonprofits in the MSA to calculate the rate of both. The growth rate, our third
dependent variable, is simply the death rate subtracted from the birth rate in order to measure how
close each variable is in any given year.
In addition to the Age, HHI, and population growth variables needed to test the stated hypotheses,
additional covariates have been included in the model as control variables since they have all been
identified as relevant factors in studies of nonprofit density (Grønbjerg and Paarlberg 2001, Saxton
and Benson 2005, Lecy and Van Slyke 2013). We also included dummy variables for ten regions
in the country as there seem to be important geographic differences in the age and size of the
nonprofit sector.
There is a high amount of volatility within an MSA from year to year, so we use a three-year
average from 2004 to 2006 to calculate a smoothed rate for 2005. The NCCS Core Trend file has
county as its highest geographic unit. In order to analyze trends for MSAs we used a crosswalk
from the Missouri Census Data Center to aggregate the figures to the MSA level.
Population, median income, percent with a college education, and unemployment rates come from
the 2005 American Community Survey (ACS), available from the Census website. In addition, we
gathered the Gini Coefficient of inequality from the Census website and the 2006 ACS; 2006 was
the earliest year for which data was available. Population, income, education, unemployment, and
inequality each came prepared at the MSA level.
Population growth measures the percentage change in population from 2000 to 2010. 2000 and
2010 data comes from the Longitudinal Tract Data Base (LTDB), a data set built and maintained
by researchers at the US2010 project housed at Brown University. The LTDB is publicly available
census data for every year since 1970 that the researchers have reconfigured into 2010 geographic
units. Data comes prepared at the census-tract level, so we used a second crosswalk from the
Missouri Census Data Center to aggregate the numbers for MSAs.
We capture political ideology with the percentage vote for the Republican Party in the 2008
presidential election. The CQ Press gathers national election data and makes it available through
the Census’ USA Counties Database. We aggregated the total number of votes and number of
registered voters to the MSA level and then calculated the percentage vote for the Democratic and
Republican candidates.
The composition of the nonprofit sector is captured by three variables. Philanthropic dollars comes
from the NCCS Core Trend file, representing the total assets held by philanthropic organizations at
the end of the year 2005. Revenue Mix represents what portion of each nonprofit’s revenue comes
from public support, such as contributions and government grants. Found in the NCCS Core Trend
file, we calculated the revenue mix by finding the percentage of total revenue in 2005 from public
support. In the analysis, we aggregated the total revenue and contributions for the MSA as a whole
prior to calculating the proportion from public support, so the variable measures the revenue mix
for the region, not each nonprofit. We calculated the age of the nonprofit sector in each MSA by
combining two data files from the NCCS, the Core Trend file previously and the Business Master
File (BMF). The BMF contains a variable for the official beginning date of each nonprofit, which
allowed us to capture the average age in 2005 of all organizations in the MSA.
Government size has been identified as an important variable in predicting nonprofit density since
government services and nonprofit services can often be substitutes (when local governments
choose to contract out services they often turn to nonprofits). We account for the size of the
government sector in each MSA with four variables from the Census’ USA Counties Database.
Government earnings and employment measures the total amount of wages and number of
government employees as categorized within the North American Industry Classification System
(NAICS). The Bureau of Economic Analysis within the Department of Commerce prepares the
data for both variables. Additionally, data on the total number of grants and direct payments from
the federal government were obtained from the Census Bureau’s Census of Governments. Because
each of these variables are highly correlated and thus including them together in the model will
induce variance inflation on these coefficients we instead create an index of government size using
principle component analysis. The variables were centered and components extracted. All of the
variables loaded on the first component, which explains 95% of the total variance. A weighted
index was created from the separate government employee earnings, direct payments, government
grants, and number of government employee variables. This index is interpreted as a single
measure of government size or capacity in the metro area.
Table 1: Descriptive Statistics
Variable
Mean
St. Dev
Minimum
Maximum
Growth Rate
0.03
0.02
-0.03
0.09
Entry Rate
0.08
0.02
0.04
0.14
Exit Rate
0.05
0.01
0.02
0.11
674,369
1,450,780
68,203
18,351,099
Population Growth
9.0%
9.0%
-16.0%
38.0%
Ave. Age of Nonprofits
18.21
2.39
11.61
24.49
Gini Coefficient
0.44
0.03
0.37
0.54
Unemployment
6.94
1.88
2.5
16.5
Republican Vote
0.5
0.11
0.2
0.78
College Graduation Rate
16.04
4.33
6.8
33.9
% Revenue From Contr.s
0.79
0.13
0.25
0.96
$43,936
$7,544
$24,501
$76,478
Foundation Spending
$150,830,116
$446,435,647
$4,149
$4,604,307,314
Government Earnings
$73,067,724
$2,452,933
$5,362,031
$154,636
55,084
105,498
4,689
1,361,785
Direct Payments
$40,440,076
$1,596,260
$3,113,806
$154,597
Government Grants
$36,787,529
$1,037,020
$2,541,465
$51,361
Population
Median Income
Government Employees
Table 2: Select Descriptive Statistics by Region
Region
Pacific North
Southwest
Mountain West
Southeast
Mid-South
Mid-Atlantic
Empire
Northeast
Midwest
Plains
Population
467,398
1,052,658
379,823
594,978
648,542
395,487
1,773,281
851,683
648,448
356,703
Population
Growth
0.148
0.134
0.146
0.131
0.105
0.061
0.023
0.031
0.033
0.081
Nonprofit
Age
16.3
16.8
16.9
17.0
17.4
19.0
19.6
19.9
20.1
20.1
HHI
0.163
0.139
0.217
0.193
0.221
0.195
0.181
0.120
0.221
0.167
Entry
Rates
0.088
0.089
0.087
0.089
0.080
0.070
0.069
0.070
0.068
0.074
Exit
Rates
0.050
0.058
0.058
0.055
0.050
0.046
0.043
0.046
0.046
0.052
Growth
Rates
0.038
0.032
0.029
0.034
0.030
0.024
0.027
0.024
0.022
0.022
RESULTS
We estimate an OLS regression model using 2005 data to examine the relationship between
metropolitan characteristics, nonprofit market characteristics, and vital rates. Since it is a crosssectional model the results should be interpreted as primarily descriptive; cities with higher/lower
levels of the independent variable tend to have higher/lower vital rates. The model is meant to help
identify which variables are salient when looking at nonprofit sector growth.
The full regression models can be found in the appendix. They include all covariates in the model,
but the tables here report only primary policy variables (population growth, average age of
nonprofits, and market concentration). The majority of the control variables included in the model
do not yield statistically significant results. Additional discussion on these variables can be found
in the appendix.
Table 3: Selected results from OLS regression of nonprofit sector vital rates in 313 metro areas.
Model 1
Model 2
Model 3
Model 4
0.034**
(0.011)
-0.002***
(0.000)
0.055***
(0.011)
0.033**
(0.011)
-0.002***
(0.000)
0.002
(0.001)
0.033**
(0.011)
-0.002*
(0.001)
-0.006
(0.007)
0.000
(0.000)
-0.014
(0.010)
-0.001*
(0.000)
0.000
(0.001)
-0.014
(0.010)
0.001
(0.001)
-0.019**
(0.006)
0.001**
(0.000)
0.047***
(0.012)
-0.002**
(0.000)
0.001
(0.001)
0.047***
(0.012)
-0.003**
(0.001)
0.013
(0.008)
-0.001
(0.000)
ENTRY RATES
Pop. Growth 2000-2010
Average Nonprofit Age
HHI (Revenue Concentration)
0.002
(0.001)
Age*HHI
EXIT RATES
Pop. Growth 2000-2010
Average Nonprofit Age
-0.014
(0.010)
-0.001*
(0.000)
HHI (Revenue Concentration)
-0.006
(0.009)
0.000
(0.001)
Age*HHI
GROWTH RATES
Pop. Growth 2000-2010
Average Nonprofit Age
0.048***
(0.012)
-0.002**
(0.000)
HHI (Revenue Concentration)
0.061***
(0.012)
0.001
(0.001)
Age*HHI
N
313
313
313
313
Several interesting findings emerge from the analysis. First of all, as predicted population growth
is positively related to nonprofit sector growth. The cities that have been growing also have more
nonprofit entries and higher overall growth as indicated by the positive and significant coefficients.
We assume that this stems from larger sector capacity as a result of larger populations to serve,
donor pools, etc. Surprisingly, though, an expanding population is not associated with lower exit
rates. Hypothesis 3 is supported with the caveat that increased market capacity will not necessarily
impact exit rates, which appear to be more strongly tied to market age and concentration instead of
size.
Second, sector age is a much better predictor of vital rates than market concentration. In all of the
models we see that average age of nonprofits within a sector (which we interpret as a proxy for the
age of the sector itself – when the sector began to develop relative to other metro areas) is
statistically significant in each of the model, even when controlling for market concentration in
Model 3. Conversely, market concentration is not significant in any of the models. Thus we find
strong support for Hypothesis 2, that carrying capacity matters, and weak support for Hypothesis 1,
that market concentration has a big impact on vital rates. The results favor an ecological view of
markets that describe market size expanding until it reaches capacity, then declining. The negative
sign on the coefficients for Age imply that older markets have lower rates of entry and grow at
slower rates, which is predicted by ecological literatures (Carroll and Hannan 1995).
The one caveat to these findings is that the negative coefficient on age in the model using exit rates
as the dependent variable suggests that older markets have lower exit rates, which is opposite of
what the ecological literature would predict. This result makes more sense when we examine the
coefficients in Model 4, which include an interaction term for sector age and market concentration.
In this case we see that concentration now matters, and the interaction is also significant, even
though market age is no longer significant. The interpretation of this finding is that market
concentration matters only in older markets. Younger markets may not yet be close to saturation,
so market concentration may not have as much of a binding effect. In older markets, though,
competition may be experienced more as resources are scarcer in crowded markets. In this
scenario, market concentration may have a larger effect.
In general we find that growth of a city leads to opportunity for nonprofit sector expansion.
Population growth appears to be more important than the size of the government or philanthropic
capital within a city. The age of the market, measured by the average age of nonprofits operating
within the market, has a negative effect on entrance and growth rates – new nonprofits are less
inclined to enter crowded markets. Exit rates, however, are not directly affected by population
expansion (market expansion), nor by the average age of the sector. They appear to be more
influenced by the interaction between sector age within a city and market concentration – older
markets that are highly concentrated have higher organizational failure rates.
DISCUSSION ON COMPETITION IN THE NONPROFIT SECTOR
We find ourselves in an interesting historical period in the nonprofit sector as the industry has been
expanding rapidly over the past few decades, but that expansion appears to be tapering with falling
market entry rates and rising exit rates. If these rates converge then the market size will plateau. If
entry rates fall below exit rates then there will be industry shake-out characterized by a decline in
the number of nonprofits (though capacity can continue to expand) and a rise in market
concentration. It is difficult to predict at this time which scenario will occur, but it is possible to
examine long-term trends in market concentration. We can see from the graphic below that total
market concentration has been falling over time. If the shakeout scenario was more salient we
might expect to see market concentration on the rise as firms begin to leave crowded markets and
the remaining firms expand to take their place.
Economies of scale and improved organizational efficiencies of veteran organizations are
important contributors of this process, but there are some reasons why we might not expect to see
these in nonprofit markets. Services to vulnerable populations are often labor-intensive, and thus
are difficult to scale. And because nonprofits are often embedded in two-sided markets where
service recipients do not always pay for services, rather donors cover costs, there are not always
clear incentives to develop new production technologies as they might not help gain market share.
As a result, when markets become saturated then steady but slow or plateaued growth is a feasible
scenario. Just based upon linear trend forecasts of the decline in growth rates, market saturation
may occur as early as 2020.
The idea of churn is very important for innovation and dynamism to occur in markets. For churn to
occur new organizations have to grow to replace older and more established organizations. This is
the process of creative destruction so famously described by Schumpeter (1942), a process that
helps innovation to drive economic growth. Churn is not guaranteed within the nonprofit sector,
though. When examining the nature of competition within the nonprofit sector it is important to
understand how different types of organizations experience competition. If we reach a steady-state
where entry rates equal exit rates for a particular sector, for example, which organizations are
exiting the market? Is it only small ones, or is it a mix of large and small organizations? If large
organizations are unaffected by competition then only the small ones will be affected by market
saturation. This has serious implications for the nonprofit sector as it may slow innovation, change,
and adaptability. Having a better understanding of the large market forces at work, as well as the
cost structures and economic processes that drive entry and exit, will help us better understand the
changes in the sector that we can expect over the next decade.
Appendix A:
OLS MODEL: ENTRY RATES
Age
Pop. Growth 2000-2010
0.034**
0.055*** 0.033**
'(0.011)
'(0.011)
'(0.011)
-0.002***
-0.002***
'(0.000)
'(0.000)
0.002
0.002
'(0.001)
'(0.001)
0.033**
'(0.011)
-0.002*
'(0.001)
-0.006
'(0.007)
0.000
'(0.000)
-0.065
(0.078)
0.004
'(0.003)
-0.001
'(0.000)
-0.013*
'(0.006)
0.003
'(0.008)
-0.026
'(0.034)
0.000
'(0.000)
0.018*
'(0.007)
0.000
'(0.000)
-0.001
'(0.001)
0.000
'(0.004)
0.006
'(0.004)
0.001
'(0.003)
0.006
'(0.005)
-0.002
'(0.004)
0.006
'(0.004)
0.006
'(0.004)
0.011**
'(0.004)
0.006
'(0.004)
0.528
313
-0.081
'(0.079)
0.005
'(0.003)
0.000
'(0.000)
-0.016*
'(0.007)
0.003
'(0.008)
-0.023
'(0.034)
0.000
'(0.000)
0.017*
'(0.007)
0.000
'(0.000)
-0.001
'(0.001)
-0.001
'(0.004)
0.004
'(0.004)
0.000
'(0.003)
0.005
'(0.005)
-0.001
'(0.004)
0.005
'(0.004)
0.005
'(0.004)
0.010**
'(0.004)
0.005
'(0.004)
0.533
313
Average Nonprofit Age
HHI (Revenue Concentration)
HHI
Both
Age*HHI
Intercept
Population (log)
Foundation Assets (log)
Revenue Proportion from Donations
Proportion Voting Republican
Gini Coefficient
Unemployment
Per Capita Income (log)
Percent Adults w/ College Education
Size of Government
MidAtlantic Region
MidSouth Region
Midwest Region
MountainWest Region
Northeast Region
PacificNorth Region
Plains Region
Southeast Region
Southwest Region
R-squared
N
-0.111
'(0.082)
0.005
'(0.003)
-0.001*
'(0.000)
-0.022**
'(0.007)
0.003
'(0.009)
-0.025
'(0.036)
0.000
'(0.001)
0.017*
'(0.007)
0.000
'(0.000)
-0.001
'(0.001)
0.001
'(0.004)
0.008
'(0.004)
0.001
'(0.004)
0.008
'(0.005)
-0.003
'(0.005)
0.010*
'(0.004)
0.003
'(0.005)
0.014***
'(0.004)
0.010*
'(0.004)
0.482
313
-0.071
'(0.078)
0.005
'(0.003)
0.000
'(0.000)
-0.017**
'(0.006)
0.003
'(0.008)
-0.025
'(0.034)
0.000
'(0.000)
0.018*
'(0.007)
0.000
'(0.000)
-0.001
'(0.001)
-0.001
'(0.004)
0.005
'(0.004)
0.001
'(0.003)
0.006
'(0.005)
-0.002
'(0.004)
0.006
'(0.004)
0.006
'(0.004)
0.011**
'(0.004)
0.006
'(0.004)
0.531
313
w/Interaction
OLS MODEL: EXIT RATES
Age
HHI
Both
w/Interaction
Pop. Growth 2000-2010
-0.014
'(0.010)
-0.001*
'(0.000)
-0.006
'(0.009)
0.000
'(0.001)
-0.014
'(0.010)
-0.001*
'(0.000)
0.000
'(0.001)
-0.014
'(0.010)
0.001
'(0.001)
-0.019**
'(0.006)
0.001**
'(0.000)
-0.020
'(0.072)
0.001
'(0.002)
-0.001
'(0.000)
-0.012*
'(0.006)
-0.010
'(0.007)
-0.038
'(0.032)
-0.001
'(0.000)
0.007
'(0.006)
0.000
'(0.000)
0.001
'(0.001)
0.005
'(0.003)
0.012**
'(0.004)
0.006
'(0.003)
0.018***
'(0.004)
0.002
'(0.004)
0.009*
'(0.004)
0.012**
'(0.004)
0.015***
'(0.003)
0.015***
'(0.004)
0.22
313
-0.006
'(0.071)
0.001
'(0.002)
0.000
'(0.000)
-0.010
'(0.006)
-0.010
'(0.007)
-0.038
'(0.031)
-0.001
'(0.000)
0.008
'(0.006)
0.000
'(0.000)
0.001
'(0.001)
0.005
'(0.003)
0.011**
'(0.004)
0.006
'(0.003)
0.017***
'(0.004)
0.002
'(0.004)
0.007
'(0.004)
0.013**
'(0.004)
0.014***
'(0.003)
0.013***
'(0.004)
0.232
313
-0.030
'(0.071)
0.000
'(0.002)
0.000
'(0.000)
-0.006
'(0.006)
-0.010
'(0.007)
-0.032
'(0.031)
0.000
'(0.000)
0.006
'(0.006)
0.000
'(0.000)
0.001
'(0.001)
0.004
'(0.003)
0.009*
'(0.004)
0.004
'(0.003)
0.015***
'(0.004)
0.003
'(0.004)
0.005
'(0.004)
0.012**
'(0.004)
0.012***
'(0.003)
0.012**
'(0.004)
0.257
313
Average Nonprofit Age
HHI (Revenue Concentration)
Age*HHI
Intercept
Population (log)
Foundation Assets (log)
Revenue Proportion from Donations
Proportion Voting Republican
Gini Coefficient
Unemployment
Per Capita Income (log)
Percent Adults w/ College Education
Size of Government
MidAtlantic Region
MidSouth Region
Midwest Region
MountainWest Region
Northeast Region
PacificNorth Region
Plains Region
Southeast Region
Southwest Region
R-squared
N
-0.005
'(0.071)
0.000
'(0.002)
0.000
'(0.000)
-0.009
'(0.005)
-0.010
'(0.007)
-0.038
'(0.031)
-0.001
'(0.000)
0.008
'(0.006)
0.000
'(0.000)
0.001
'(0.001)
0.005
'(0.003)
0.011**
'(0.004)
0.006
'(0.003)
0.017***
'(0.004)
0.002
'(0.004)
0.007
'(0.004)
0.013**
'(0.004)
0.014***
'(0.003)
0.013***
'(0.004)
0.232
313
OLS MODEL: GROWTH RATES
Age
HHI
Both
w/Interaction
Pop. Growth 2000-2010
0.048***
'(0.012)
-0.002**
'(0.000)
0.061***
'(0.012)
0.001
'(0.001)
0.047***
'(0.012)
-0.002**
'(0.000)
0.001
'(0.001)
0.047***
'(0.012)
-0.003**
'(0.001)
0.013
'(0.008)
-0.001
'(0.000)
-0.090
'(0.088)
0.005
'(0.003)
-0.001
'(0.000)
-0.010
'(0.007)
0.013
'(0.009)
0.013
'(0.039)
0.000
'(0.001)
0.009
'(0.008)
0.000
'(0.000)
-0.002
'(0.001)
-0.005
'(0.004)
-0.004
'(0.005)
-0.005
'(0.004)
-0.009
'(0.005)
-0.005
'(0.005)
0.002
'(0.005)
-0.009
'(0.005)
-0.001
'(0.004)
-0.005
'(0.005)
0.267
313
-0.065
'(0.087)
0.005
'(0.003)
0.000
'(0.000)
-0.007
'(0.007)
0.013
'(0.009)
0.012
'(0.038)
0.000
'(0.001)
0.010
'(0.008)
0.000
'(0.000)
-0.002
'(0.001)
-0.005
'(0.004)
-0.006
'(0.005)
-0.005
'(0.004)
-0.011*
'(0.005)
-0.004
'(0.005)
-0.001
'(0.005)
-0.007
'(0.005)
-0.003
'(0.004)
-0.008
'(0.005)
0.292
313
-0.051
'(0.088)
0.005
'(0.003)
0.000
'(0.000)
-0.009
'(0.007)
0.014
'(0.009)
0.009
'(0.038)
0.000
'(0.001)
0.011
'(0.008)
0.000
'(0.000)
-0.002
'(0.001)
-0.005
'(0.004)
-0.005
'(0.005)
-0.004
'(0.004)
-0.010
'(0.005)
-0.004
'(0.005)
0.000
'(0.005)
-0.006
'(0.005)
-0.003
'(0.004)
-0.007
'(0.005)
0.297
313
Average Nonprofit Age
HHI (Revenue Concentration)
Age*HHI
Intercept
Population (log)
Foundation Assets (log)
Revenue Proportion from Donations
Proportion Voting Republican
Gini Coefficient
Unemployment
Per Capita Income (log)
Percent Adults w/ College Education
Size of Government
MidAtlantic Region
MidSouth Region
Midwest Region
MountainWest Region
Northeast Region
PacificNorth Region
Plains Region
Southeast Region
Southwest Region
R-squared
N
-0.060
'(0.087)
0.004
'(0.003)
0.000
'(0.000)
-0.004
'(0.006)
0.013
'(0.009)
0.012
'(0.038)
0.000
'(0.001)
0.010
'(0.008)
0.000
'(0.000)
-0.002
'(0.001)
-0.005
'(0.004)
-0.005
'(0.004)
-0.005
'(0.004)
-0.011*
'(0.005)
-0.004
'(0.005)
-0.001
'(0.005)
-0.007
'(0.005)
-0.003
'(0.004)
-0.007
'(0.004)
0.289
313
References:
Agarwal, R. and M. Gort (1996). "The evolution of markets and entry, exit and survival of firms."
The review of Economics and Statistics: 489-498.
Aldrich, H. E. and J. Pfeffer (1976). "Environments of organizations." Annual review of sociology
2: 79-105.
Armsworth, P. R., I. S. Fishburn, Z. G. Davies, J. Gilbert, N. Leaver and K. J. Gaston (2012). "The
size, concentration, and growth of biodiversity-conservation nonprofits." BioScience 62(3): 271281.
Bartelsman, E., S. Scarpetta and F. Schivardi (2005). "Comparative analysis of firm demographics
and survival: evidence from micro-level sources in OECD countries." Industrial and Corporate
Change 14(3): 365-391.
Bartelsman, E. J., J. Haltiwanger and S. Scarpetta (2004). Microeconomic evidence of creative
destruction in industrial and developing countries, IZA discussion paper series.
Bonaccorsi, A. and P. Giuri (2000). Industry life cycle and the evolution of an industry network,
Laboratory of Economics and Management (LEM), Sant'Anna School of Advanced Studies, Pisa,
Italy.
Bonaccorsi, A. and P. Giuri (2000). "When shakeout doesn’t occur: the evolution of the turboprop
engine industry." Research Policy 29(7): 847-870.
Buenstorf, G. (2007). "Evolution on the shoulders of giants: entrepreneurship and firm survival in
the German laser industry." Review of Industrial Organization 30(3): 179-202.
Burger, R. and T. Owens (2013). "Receive Grants or Perish? The Survival Prospects of Ugandan
Non-Governmental Organisations." The Journal of Development Studies(ahead-of-print): 1-15.
Carree, M. A. and A. R. Thurik (2000). "The life cycle of the US tire industry." Southern
Economic Journal: 254-278.
Carroll, D. A. and K. J. Stater (2009). "Revenue Diversification in nonprofit organizations: does it
lead to financial stability?" Journal of Public Administration Research and Theory 19(4): 947-966.
Carroll, G. R. (1984). "Organizational ecology." Annual review of Sociology: 71-93.
Carroll, G. R. and M. T. Hannan (1995). Organizations in industry: Strategy, structure, and
selection, Oxford University Press.
Corbin, J. J. (1999). "A study of factors influencing the growth of nonprofits in social services."
Nonprofit and Voluntary Sector Quarterly 28(3): 296-314.
Day, G. S. and A. J. Fein (2003). "Shakeouts in digital markets." California Management Review
45(2).
Dougherty, K. D., J. Maier and B. V. Lugt (2008). "When the final bell tolls: Patterns of Church
closings in two Protestant denominations." Review of Religious Research: 49-73.
Duckies, B. M., M. A. Hager and J. Galaskiewicz (2005). HOW NONPROFITS CLOSE.
Qualitative Organizational Research: Best Papers from the Davis Conference on Qualitative
Research (HC), IAP.
Fernandez, J. J. (2008). "Causes of dissolution among Spanish nonprofit associations." Nonprofit
and Voluntary Sector Quarterly 37(1): 113-137.
Galaskiewicz, J. and W. Bielefeld (1998). Nonprofit organizations in an age of uncertainty: A
study of organizational change, Transaction Books.
Grønbjerg, K. A. and L. Paarlberg (2001). "Community variations in the size and scope of the
nonprofit sector: Theory and preliminary findings." Nonprofit and Voluntary Sector Quarterly
30(4): 684-706.
Hager, M. A. (2001). "Explaining Demise among Nonprofit Organizations Abstract." Nonprofit
and Voluntary Sector Quarterly 30(4): 795-796.
Hager, M. A., J. Galaskiewicz and J. A. Larson (2004). "Structural embeddedness and the liability
of newness among nonprofit organizations." Public Management Review 6(2): 159-188.
Hannan, M. T. and J. Freeman (1977). "The population ecology of organizations." American
journal of sociology: 929-964.
Hannan, M. T. and J. Freeman (1987). "The ecology of organizational founding: American labor
unions, 1836-1985." American Journal of Sociology: 910-943.
Hannan, M. T. and J. Freeman (1993). Organizational ecology, Harvard University Press.
Harrison, T. D. and C. A. Laincz (2008). "Entry and exit in the nonprofit sector." The BE Journal
of Economic Analysis & Policy 8(1).
Harrison, T. D. and J. Thornton (2012). "Entry Thresholds and Competitive Behavior Among
Nonprofit Firms." Available at SSRN 1964437.
Horvath, M., F. Schivardi and M. Woywode (2001). "On industry life-cycles: delay, entry, and
shakeout in beer brewing." International Journal of Industrial Organization 19(7): 1023-1052.
Jovanovic, B. and G. MacDonald (1994). The life-cycle of a competitive industry, National Bureau
of Economic Research.
Klepper, S. (1997). "Industry life cycles." Industrial and corporate change 6(1): 145-182.
Lecy, J. D. and D. M. Van Slyke (2013). "Nonprofit Sector Growth and Density: Testing Theories
of Government Support." Journal of Public Administration Research and Theory 23(1): 189-214.
Luksetich, W. (2008). "Government funding and nonprofit organizations." Nonprofit and
Voluntary Sector Quarterly 37(3): 434-442.
Maier, J. E. (2010). Congregational growth, closure, identity, and diversity.
Matsunaga, Y. and N. Yamauchi (2004). "Is the government failure theory still relevant? A panel
analysis using US state level data." Annals of Public and Cooperative Economics 75(2): 227-263.
McClellan, S. T. (1984). The coming computer industry shakeout: winners, losers, and survivors,
John Wiley & Sons, Inc.
Murmann, J. P. and E. Homburg (2001). "Comparing evolutionary dynamics across different
national settings: the case of the synthetic dye industry, 1857–1914." Journal of Evolutionary
Economics 11(2): 177-205.
Nelson, R. R. (1982). An evolutionary theory of economic change, Harvard University Press.
Pandey, L., D. L. Sjoquist and M. B. Walker (2009). "An Analysis of Private School Closings."
Education 4(1): 34-59.
Roeger, K. L., A. Blackwood and S. L. Pettijohn (2012). The Nonprofit Almanac 2012, Urban
Institute Press.
Salamon, L. M., S. W. Sokolowski and S. L. Geller (2012). "Holding the Fort: Nonprofit
Employment During a Decade of Turmoil." Johns Hopkins University: Center for Civil Society
Studies.
Saxton, G. D. and M. A. Benson (2005). "Social capital and the growth of the nonprofit sector*."
Social Science Quarterly 86(1): 16-35.
Schumpeter, J. A. (1942). Capitalism, socialism, and democracy, HarperCollins.
Seokki, A. (2010). Heterogeneous Population and Evolutionary Dynamics of Subpopulations: The
Entry of Nonprofit Organizations in Florida Counties 1994-2007.
Skocpol, T. (2013). Diminished democracy: From membership to management in American civic
life, University of Oklahoma press.
Straughan, B. (2008). "The broader movement: Nonprofit environmental and conservation
organizations, 1989-2005."
Stretesky, P. B., S. Huss, M. J. Lynch, S. Zahran and B. Childs (2011). "The founding of
environmental justice organizations across US counties during the 1990s and 2000s: civil rights
and environmental cross-movement effects." Social Problems 58(3): 330-360.
Tremblay, V. J., N. Iwasaki and C. H. Tremblay (2005). "The dynamics of industry concentration
for US micro and macro brewers." Review of Industrial Organization 26(3): 307-324.
Van Kranenburg, H. L., F. C. Palm and G. A. Pfann (1998). "The life cycle of daily newspapers in
the Netherlands: 1848–1997." De Economist 146(3): 475-494.
Van Kranenburg, H. L., F. C. Palm and G. A. Pfann (2002). "Exit and survival in a concentrating
industry: the case of daily newspapers in the Netherlands." Review of Industrial Organization
21(3): 283-303.
Vance, D. L. (2010). Government funding and failure in nonprofit organizations, Indiana
University.
Walker, E. T. and J. D. McCarthy (2010). "Legitimacy, strategy, and resources in the survival of
community-based organizations." Social problems 57(3): 315-340.
Walter, J. (2005). "Depression-Era Bank Failures: The Great Contagion or the Great Shakeout?"
FRB Richmond Economic Quarterly 91(1): 39-54.
Wollebaek, D. (2009). "Survival in local voluntary associations." Nonprofit Management and
Leadership 19(3): 267-284.