Content uploaded by Qiushi Wang
Author content
All content in this area was uploaded by Qiushi Wang on Nov 29, 2022
Content may be subject to copyright.
ORIGINAL PAPER
Does a More Diversified Revenue Structure Lead to Greater
Financial Capacity and Less Vulnerability in Nonprofit
Organizations? A Bibliometric and Meta-Analysis
Jiahuan Lu
1
•Weiwei Lin
1
•Qiushi Wang
2
International Society for Third-Sector Research and The Johns Hopkins University 2019
Abstract This article explores how and to what extent
revenue diversification and concentration strategies affect
financial performance, particularly financial capacity and
vulnerability, in nonprofit organizations. Using a sample
collected from a systematic literature search of all major
databases, we first conducted a bibliometric analysis of 86
existing studies to visualize the clusters of major topics in
this area and to explore the connections between existing
studies. We then employed a meta-analysis to quantita-
tively synthesize 258 effect sizes from 23 existing empir-
ical studies. We found that diversification had little effect
on financial vulnerability, but it had a slightly negative
effect on financial capacity. The article finally uses a meta-
regression to discuss some of the theoretical and practical
reasons why there is inconsistency in the results across
existing studies and calls for more discussion of the
assumptions and effectiveness of revenue diversification
among nonprofit scholars and practitioners.
Keywords Revenue diversification Financial capacity
Financial vulnerability Bibliometric analysis Meta-
analysis
Introduction
What is the best revenue strategy for a nonprofit organization
to enhance its financial stability and achieve better financial
performance? Scholars are split into two camps on this issue.
Based on modern portfolio theory (Markowitz 1952),
resource dependence theory (Pfeffer and Salanick 2003), and
institutional theory (Scott 1987), one camp suggests that
revenue diversification can achieve a more efficient portfolio
(Grasse et al. 2016;Kingma1993), avoid excessive depen-
dence on any single funding source (Froelich 1999; Hodge
and Piccolo 2005), enhance financial stability and capacity
(Carroll and Stater 2009; Lam and McDougle 2016), and
reduce the risk of financial vulnerability in the face of fiscal
shocks (Chang and Tuckman 1996;GreenleeandTrussel
2000; Lin and Wang 2016; Tevel et al. 2015). Therefore, it is
desirable for nonprofits to establish and maintain multiple
funding streams including government contracts, individual
contributions, earned income, and so on. The other camp,
however, advocates for revenue concentration by emphasiz-
ing the higher transaction costs (Frumkin and Keating 2011;
Grønbjerg 1993), the possible crowding-out effect (Brooks
2000), and the added administrative complexity (de Los
Mozos et al. 2016). Over the past three decades, scholars from
both camps have found empirical evidence to support their
arguments. In short, there is no consensus position on this
issue.
It is of great importance for both nonprofits and gov-
ernments to find a solid answer to this question and settle
this long-standing debate between these two mutually
&Qiushi Wang
wangqsh3@sysu.edu.cn
Jiahuan Lu
jiahuan.lu@rutgers.edu
Weiwei Lin
weiwei.lin@rutgers.edu
1
School of Public Affairs and Administration, Rutgers
University-Newark, 111 Washington Street, Newark,
NJ 07102, USA
2
School of Government and Center for Chinese Public
Administration Research, Sun Yat-sen University, 135
Xingang Xi Road, Guangzhou 510275,
People’s Republic of China
123
Voluntas
https://doi.org/10.1007/s11266-019-00093-9
exclusive revenue strategies. On the nonprofit side, revenue
strategy is clearly one of the top concerns of nonprofit
management. Over the past few decades, nonprofit scholars
have discussed such important topics as fundraising effi-
ciency, organizational effectiveness, the crowding-in/out
effect of government grants, and the government financing
of nonprofit activities. All of these areas are related to
revenue strategy to a greater or lesser extent. Indeed, rev-
enue strategy shapes nonprofits’ ability to generate enough
revenue for organizational survival and growth. Obviously,
nonprofits need a clear revenue strategy to direct their
fundraising efforts and other aspects of their operations.
This issue is equally important on the government side.
Under governance models such as third-party government
and collaborative governance, the nonprofit sector plays an
increasingly critical role in providing various public goods
and services such as health and social services (Bryson
et al. 2006; Salamon 1995). Over time, governments have
become largely dependent on nonprofits to deliver services
to serve citizens and implement policies that advance
public priorities. In this context, a healthy and viable
nonprofit sector, which includes effective revenue strate-
gies, is necessary to help governments achieve better per-
formance and higher public trust (Light 2004). All put
together, revenue strategy is a central and pressing issue for
both public administration and nonprofit researchers.
This article does not attempt to produce additional
empirical results. Instead, it aims to synthesize and assess
all extant studies in order to reconcile the observed dis-
crepancies in the literature and to find more reliable esti-
mates of the relationship between revenue diversification
and concentration and nonprofit financial performance. Our
research contributes to the current literature in several
important ways. First, we conducted a systematic literature
search in all major databases using complementary strate-
gies to identify as many relevant studies as possible. The
resulting sample is unparalleled in the field. Second, we
employed a relatively new bibliometric technology,
CiteSpace, to systematically categorize existing studies and
visualize the connections among these studies. Third, we
performed a meta-analysis using effect sizes from a refined
sample of the most relevant empirical studies and con-
ducted a conclusive assessment of the impact of revenue
diversification and concentration on nonprofit financial
performance. Last, we conducted a meta-regression anal-
ysis to explore the moderators that would account for the
variations of effect size estimates across the studies in our
sample.
The remainder of this article is organized as follows: In
Sect. 2, we summarize the existing theories and review
previous empirical literature on revenue diversification and
concentration strategies. In Sect. 3, we explain our bib-
liometric analysis and discuss the meta-analysis method we
used for this study. We then present the results of our
average effect size analysis and moderator analysis in
Sect. 4. In the final section, we present our conclusions and
discuss the limitations of our findings.
Theoretical Framework
This section first reviews the theoretical foundations of the
revenue diversification and concentration strategies, and
then summarizes the empirical findings from the studies
testing these theories.
Major Theories Concerning Revenue Diversification
and Revenue Concentration
Originally developed by Markowitz (1952), modern port-
folio theory (MPT) describes how a risk-averse investor
can select an optimal investment portfolio to maximize
expected returns and minimize volatility. Diversification,
through the law of large numbers, can keep actual returns
close to the amount of anticipated returns, and therefore
reduce overall portfolio volatility for a given expected
return (Markowitz 1952). Although the nonprofit sector is
unique in its fundraising methods (Jegers and Verschueren
2006; Steinberg 1990), the concept of revenue diversifi-
cation is ‘‘nonetheless applicable as a prudent revenue
generation strategy to potentially minimize the volatility of
revenue portfolios managed by nonprofits’’ (Carroll and
Stater 2009, p. 949). The underlying rationale for the
revenue diversification strategy is that, as Tuckman and
Chang (1991, p. 452) argued, ‘‘a nonprofit is more vul-
nerable to revenue downturns if its revenue sources are
limited than if they are diverse…because a shock is more
likely to affect one revenue source than it is to affect all
sources at once.’’ As such, the principles set forth by MPT
have been widely applied by nonprofits as part of their
resource acquisition strategies (Grasse et al. 2016; Kingma
1993). Many scholars have suggested that nonprofits
should have a more diversified revenue base that includes
government funding, private donations, fees, and other
sources in order to minimize potential instability of funding
streams and to achieve superior performing revenue port-
folios (e.g., Chang and Tuckman 1996; Froelich 1999;
Greenlee and Trussel 2000; Trussel 2002; Tevel et al.
2015).
Resource dependence theory provides another perspec-
tive for understanding the revenue portfolio development
of nonprofits. According to Pfeffer and Salanick (2003), the
survival of an organization depends on its ability to acquire
and maintain resources, but resources are almost always
inadequate, unstable, or uncertain. Consequently, organi-
zations are constrained by the funding environment
Voluntas
123
because of their resource needs and interdependence
between themselves and others. The degree of resource
dependence is determined by the importance and concen-
tration of the resources provided by funders. Organizations
relying on only a few sources for vital inputs will become
more dependent on resource suppliers for survival com-
pared to those relying on more sources. Resource suppliers,
however, often exert a significant impact on nonprofit
operations in areas such as service delivery and resource
use. Therefore, reliance on one particular stream of rev-
enue, whether donations, grants, or earned income, will
greatly affect organizational autonomy and financial sta-
bility. As a result, nonprofits tend to have an incentive to
diversify their revenue sources in order to maintain orga-
nizational autonomy by reducing dependence on their pri-
mary funding sources (Froelich 1999; Hodge and Piccolo
2005).
The third theoretical perspective is institutional theory.
Unlike resource dependence theory, it is not mainly based
on market-driven rational or instrumental considerations.
Instead, institutional theory emphasizes the non-market-
driven social and normative demands that environments
impose on organizations (Scott 1987). From this perspec-
tive, organizational structures are seen as shaped by such
factors as the imitation of successful organizations and the
normative transmission of social prescriptions (DiMaggio
and Powell 1983) as well as the need for organizational
legitimacy (Meyer and Scott 1992). To imitate other suc-
cessful organizations in response to uncertainty or to
maintain their legitimacy, organizations will fulfill or
match the expectations of the environment and engage in
more activities designed to enhance their identification and
alignment with legitimated aspects of their environment.
According to this institutional logic, nonprofits whose
funding sources are more heterogeneous are in a better
position to build institutional linkages to the community
and bolster their reputations and desirability as fund
recipients; therefore, they are more likely to stabilize their
resource flows and achieve better organizational sustain-
ability (Bielefeld 1992; Kerlin and Pollak 2011).
On the other end of the spectrum is the transaction costs
line of reasoning. Some scholars argue that revenue
diversification will substantially increase transaction costs
in nonprofit operations, which undermines the benefits of
revenue diversification as an effective financial manage-
ment tool. First, incorporating and maintaining multiple
funding relationships will necessarily incur significantly
greater costs, such as increased administrative monitoring,
communication, and reporting costs (Frumkin and Keating
2011; Grønbjerg 1993). Second, the added complexity of
managing multiple revenue streams and the uncertainty of
some revenue sources might reduce financial predictability
(de Los Mozos et al. 2016; Kingma 1993). All these extra
costs associated with diversification could prevent non-
profits from efficiently managing any of their funding
relationships. Third, diversification in the nonprofit sector
often involves generating revenue from unique funding
sources. The reliance of some funding sources (e.g., com-
mercial income) may undermine some nonprofits’ legiti-
macy and ability to carry out their missions (Tuckman and
Chang 1992; Weisbrod 1998). Meanwhile, a mixture of
different revenue sources may lead to a crowding-out effect
between revenue sources and further undermine revenue
growth (Brooks 2000;Lu2016). For these reasons, non-
profits can probably economize on the transaction costs and
increase organizational efficiency by concentrating revenue
on a few sources (Chikoto and Neely 2014; Foster and Fine
2007; Frumkin and Keating 2011).
Empirical Evidence
Employing the four main theories explained above, previ-
ous empirical studies have produced mixed results, pro-
viding evidence for either a positive or a negative
association between revenue diversification and nonprofit
financial performance.
Many early studies found that a higher level of revenue
diversification, no matter how measured, was positively
associated with various indicators of better financial per-
formance (e.g., Chang and Tuckman 1994; Trussel 2002;
Greenlee and Trussel 2000; Keating et al. 2005; Carroll
and Stater 2009; Tuckman and Chang 1991). For instance,
Tuckman and Chang (1991) and Chang and Tuckman
(1994) showed that revenue diversification was positively
correlated with higher operating margins and larger net
assets. By examining a sample of US arts organizations,
Hager (2001) found that revenue diversification decreases
the likelihood of organizational closure in the face of
financial shocks. Similarly, Carroll and Stater (2009)
demonstrated that revenue diversification can reduce rev-
enue volatility and lead to better financial stability. In the
non-US context, Wicker and Breuer (2014) examined a
sample of German sports organizations and revealed a
positive relationship between revenue diversification and
organizational financial conditions. Tevel et al. (2015) also
found empirical evidence that revenue diversification
reduced financial vulnerability in Israeli nonprofits.
Many recent studies, however, demonstrated a negative
or null association between revenue diversification and
financial effectiveness (e.g., Frumkin and Keating 2011;
Chikoto and Neely 2014; Mayer et al. 2014; de los Mozos
et al. 2016; Lin and Wang 2016; von Schnurbein and Fritz
2017). For example, Frumkin and Keating (2011) discov-
ered that revenue diversification lead to lower organiza-
tional efficiency. Similarly, de los Mozos et al. (2016)
found a negative impact of revenue diversification on
Voluntas
123
fundraising efficiency. Moreover, Mayer et al. (2014)
suggested that diversified revenue does not necessarily
lower volatility for US nonprofits. Lin and Wang (2016)
showed that revenue diversification might even aggravate
nonprofits’ fiscal stress during an economic downturn. In
addition, Chikoto and Neely (2014) found that a more
concentrated revenue structure can boost total revenue
growth in US nonprofits. The same result was reported in
von Schnurbein and Fritz’s (2017) study of Swiss
organizations.
Methods and Data
Given the competing theories and mixed empirical results
in the literature, one might wonder what the true effect of
revenue diversification on nonprofit financial performance
is? In this article, we follow a two-step procedure to answer
this question.
The first step is to conduct a broad bibliometric analysis
of all scholarly articles on financial performance in general
and revenue diversification and concentration in specific.
Bibliometric methods are frequently used to provide a
quantitative analysis of academic literature. The central
idea is that the development of an academic field can be
traced by studying its footprints in scholarly publications
(Chen 2004). There are several types of bibliometric
studies, including collaboration network analysis, co-word
analysis, author co-citation analysis, text and geospatial
visualization, and so forth. In this study, we focus on the
document co-citation analysis (DCA) using a professional
software called CiteSpace.
1
DCA studies a network of co-cited reference (Small
2003). The fundamental assumption is that co-citation
clusters can reveal underlying intellectual structures.
CiteSpace first takes a set of bibliographic records
2
as its
input and models the intellectual structure of the underly-
ing domain in terms of a synthesized network based on
time series of networks derived from each year’s publica-
tions on relevant topics. The synthesized network is then
divided into co-citation clusters of references. Citers to
these references are considered as the research fronts
associated with these clusters. Each cluster represents the
intellectual base of the underlying knowledge domain. To
characterize the nature of an already identified cluster,
CiteSpace next extracts noun phrases from the titles, key-
word lists, or abstracts of articles that cited the particular
cluster. Once the process is completed, cluster labels based
on the selection algorithm (usually log-likelihood ratio)
will be displayed. Additionally, CiteSpace can also high-
light nodes with high betweenness centrality with purple
trims.
3
The thickness of a purple betweenness centrality
trim indicates how strong its betweenness centrality is.
Thus, by visualizing a knowledge domain’s co-citation
network, the bibliometric method can identify intellectu-
ally significant articles and the major areas of research
(clusters), reveal the connections between these major
areas (lines linking the nodes), validate the contributions of
the leading scholars in the field (size of the nodes and
thickness of the trims), and pinpoint the key papers for a
given area (label of the nodes).
4
The second step of the empirical inquiry is to narrow our
focus using the results of the bibliometric analysis to
conduct a meta-analysis. This allows us to take stock of the
most relevant empirical findings within the literature and
estimate a generalized effect of revenue diversification on
financial capacity and vulnerability across studies. Meta-
analysis is a quantitative research synthesis method which
enables researchers to statistically combine empirical
results from different studies for the purpose of scientific
generalization (Glass 1976). Since any single study inevi-
tably suffers from sampling error and other artifacts,
aggregating findings from multiple studies is more likely to
produce estimates that are closer to the true underlying
relationships and have higher levels of external validity
(Schmidt and Hunter 2014). In our case, meta-analysis can
help us estimate an average effect size of revenue diver-
sification on financial capacity and vulnerability to form a
generalized knowledge and explore the conditions under
which we could expect different directions and magnitudes
of this effect.
In sum, the bibliometric analysis lays the groundwork
for the meta-analysis, while the latter condenses and
quantitatively generalizes the findings of the former. The
combination of the two steps will provide us with both a
panoramic view of the literature developments in the field
and a synthesized result of different studies.
Literature Search and Bibliometric Analysis
To develop our sample for the bibliometric analysis and the
subsequent meta-analysis, we started the analysis with a
1
Other widely used science mapping tools include VOSViewer,
HistCite, SciMAT, and Sci2. In many ways, they are similar to
CiteSpace.
2
These bibliographic records refer to standard word profiles such as
title, abstract, and keywords that characterize the nature of an article.
Popular databases, such as Web of Science and Scopus, are often used
for collecting these bibliographic records. We will describe the details
of data collection for this study in the next section.
3
Each node represents an article. A node of high betweenness
centrality is usually one that connects two or more large groups of
nodes with the node itself in-between, hence the term betweenness.
4
For more details about the bibliometric method and the CiteSpace
software, one can refer to Chen (2004) and the CiteSpace Manual
available from https://leanpub.com/howtousecitespace.
Voluntas
123
comprehensive literature search using several comple-
mentary strategies to identify as many relevant studies as
possible (Reed and Baxter 2009). To begin with, we con-
ducted database searches in Web of Science (for journal
articles), EBSCO (for journal articles), SCOPUS (for
journal articles), ProQuest (for theses and dissertations),
and SSRN (for working papers). Afterward, we searched
relevant articles in three core nonprofit journals, Nonprofit
and Voluntary Sector Quarterly,Nonprofit Management
and Leadership, and Voluntas: International Journal of
Voluntary and Nonprofit Organizations. In both steps, we
searched literature using the abstract search profile (rev-
enue diversification OR revenue concentration) AND
financ*. Next, after collecting an initial set of studies, we
conducted an ancestor search by examining the reference
sections of the identified studies to seek other relevant
studies. Lastly, we used Google Scholar to do a descendant
search by reviewing later studies that cited the identified
studies. We iterated the third and fourth steps until no new
relevant studies were identified. The entire literature search
was concluded on September 30, 2017.
The above search process resulted in a total of 86
scholarly articles, covering the period 1991–2017. Using
this sample, we employed the CiteSpace 5.0 software
developed by Chen (2004) to visually analyze the co-ci-
tation network of the field. The major areas (‘‘clusters’’) of
research, the most important articles (‘‘nodes’’) in the field,
as well as the links between the articles, are shown in
Fig. 1. The modularity Qwas .61 which is relatively high.
This means that the network is reasonably divided into
clearly defined co-citation clusters. Additionally, the mean
silhouette score of .81 suggests that the homogeneity of
these clusters on average is also very high. Therefore, the
overall quality of the co-citation network is quite good.
As Fig. 1illustrates, nonprofit scholars have plowed the
field of nonprofit revenue diversification and concentration
since the early 1990s, but many influential works did not
appear until after the Great Recession of 2008.
5
Using the
log-likelihood ratio measure to generate cluster labels, the
largest (#0) and second largest (#1) clusters of literature are
‘‘Herfindahl index’’ and ‘‘nonprofit revenue concentra-
tion,’’ where researchers discuss the measurement of non-
profit revenue diversification or concentration and how
either is related to nonprofit financial stability and growth.
It should be noted that there is considerable overlap
between these two clusters, indicating that the two clusters
are closely related to each other. The third largest (#2)
cluster is related to how nonprofits can improve their rev-
enue strategies in order to better weather the negative
impacts of a major fiscal downturn such as the Great
Recession of 2008. The fourth (#3) and fifth (#4) clusters
include such topics as program ratio management, resource
dependence, the impact of overhead costs on financial
capacity, and other accountability issues in nonprofit
organizations. The last (#5) cluster consists of broader
financial issues that are related to nonprofit organizational
effectiveness, but it is only loosely linked to the other five
clusters.
Literature Refinement and Meta-Analysis
The bibliometric analysis in the previous section provides a
panoramic picture of the development of the literature in
the field, but it needs to be narrowed down and refined for
more in-depth analysis. To do so, we manually reviewed
the abstracts of the relevant studies for inclusion in the
meta-analysis. If an article’s eligibility for inclusion was
unclear from its abstract, we performed a full-text review
to determine whether the study met our criteria. To be
included in the sample, first, the study had to quantitatively
examine the effect of revenue diversification or concen-
tration on financial capacity and/or vulnerability. Second,
the unit of analysis of the study had to be individual
organizations. Third, the focal variable, revenue diversifi-
cation, had to be measured using the Herfindahl–Hirsch-
man index (HHI, with 0 denoting perfect concentration and
1 denoting perfect diversification) instead of only the
number of revenue sources (e.g., Despard et al. 2017;
Foster and Meinhard 2005), since the latter fails to account
for the dispersion of different sources (Tuckman and Chang
1991).
6
Fourth, the dependent variable, financial vulnera-
bility, had to be conceptualized as the extent to which a
nonprofit scales back its operations when experiencing a
financial shock such as an economic downturn and the loss
of a major donor (Cordery et al. 2013; Tuckman and Chang
1991; Greenlee and Trussel 2000), and, financial capacity,
had to be conceptualized as the extent to which a nonprofit
maintains or expands financial resources to support its
operations (Bowman 2011; Chikoto and Neely 2014).
Fifth, the study had to use financial measures of capacity
and vulnerability, rather than perceptional or other mea-
sures (e.g., Hager 2001; Wicker and Breuer 2013).
Table 1summarizes the measures of both dependent
variables included in our meta-analysis. After extensive
screening, the final set of studies included in the analysis
consisted of 17 journal articles, 3 working papers, and 3
dissertations. These 23 studies represent a wide variety of
year durations, nonprofit types, countries, policy fields, and
research designs, helping to enhance the external validity
5
In CiteSpace, this is indicated by different colors, but it may not be
distinctive in black and white.
6
We reverse coded the effects of revenue concentration in the
original studies to make comparable the information coded from all
the included studies.
Voluntas
123
of the meta-analysis. Detailed descriptions of the included
studies are provided in ‘‘Appendix.’’ While we cannot
guarantee the inclusion of every relevant study, we are
confident that we have identified a systematic sample to
explore our research question.
The next step is to extract information from these
studies. Specifically, we coded two groups of information:
effect size and moderator information (Lipsey 2009). In the
present analysis, effect size refers to a standardized asso-
ciation between revenue diversification and financial
capacity or vulnerability. Following common practice, we
employed a correlation-based effect size (i.e., Pearson’s r)
(Geyskens et al. 2009). To calculate the correlation-based
effect sizes from the original studies, we first coded the
studies’ regression parameters and statistics (regression
coefficient, sample size, number of independent variables
used, model specification method, etc.), and then followed
the calculation procedures suggested by Borenstein (2009),
Fleiss and Berlin (2009), and Ringquist (2013). In partic-
ular, when original studies reported multiple effect sizes
(because of different variable measurements, model spec-
ifications, or sample restrictions), all relevant effect sizes
were coded to maintain within-study variation (Ringquist
2013).
Finally, we captured a total of 258 effect sizes, with 76
effect sizes on the diversification–vulnerability relationship
and 182 effect sizes on the diversification–capacity rela-
tionship. Within the effect sizes on diversification–
Fig. 1 Clusters of literature on
nonprofit revenue diversification
and concentration and financial
performance 1991–2017. Note:
N = 86, Modularity = .61,
Silhouette = .81
Table 1 Measures included in the meta-analysis and example studies
Construct Measures Example studies
Financial vulnerability Reduction in program expenditure Greenlee and Trussel (2000) and Tuckman and Chang (1991)
Reduction in net assets Keating et al. (2005) and Trussel and Greenlee (2004)
Reduction in net earnings Gordon et al. (2013) and Keating et al. (2005)
Financial capacity Revenue growth Chikoto and Neely (2014) and von Schnurbein and Fritz (2017)
Surplus margin Chang and Tuckman (1994) and Wicker and Breuer (2014)
Fixed assets growth de Jong (2014) and Frumkin and Keating (2011)
Expense growth Frumkin and Keating (2011) and Lin and Wang (2016)
Others Calabrese (2012a) and Lam and McDougle (2016)
Voluntas
123
vulnerability relationship, 38 were positive, 5 were null,
and 33 were negative, ranging from -.3460 to .2865.
Within the effect sizes on the diversification–capacity
relationship, 67 were positive, 2 were null, and 113 were
negative, ranging from -.4897 to .3890. These competing
findings confirmed the disparity in the literature, under-
scoring the need for synthesis. Figure 2shows the distri-
bution of study-level effect sizes across the studies
included in the analysis.
After calculating the effect sizes from individual studies,
we coded the moderator information. Moderators are the
factors that contribute to effect size variability within and
across original studies. They also help explain why dif-
ferent studies reach different results or under what condi-
tions we should expect a particular relationship between
revenue diversification and financial capacity or vulnera-
bility. In the present analysis, we examined a series of
research design characteristics to examine whether the
relationship was consistent across studies with different
research designs. Specifically, we explored the potential
effect of moderators using the following empirical model:
yi¼b0þb1x1;iþb2x2;iþb3x3;iþb4x4;iþb5x5;iþb6x6;i
þei
where yis the correlation-based effect size in the original
study i,x1is the year duration of study i,x2is the number
of revenue sources study iuses in the calculation of HHI,
x3is whether study icontrols for policy field in the analysis
(yes = 1 and no = 0), x4is whether study icontrols for
organizational size in the analysis (yes = 1 and no = 0), x5
is whether study icontrols for organizational age in the
analysis (yes = 1 and no = 0), and x6is whether study iis a
peer-reviewed journal article (yes = 1 and no = 0). Table 2
reports the descriptive statistics of the moderators.
Meta-Analysis Results
Average Effect Size Analysis
We first aggregated individual effect sizes to estimate an
average effect size for diversification–vulnerability or
capacity relationships. Before combining individual effect
sizes, two methodological treatments were made. First,
individual effect sizes were corrected for sampling errors.
Individual effect sizes were weighted by an estimate of the
inverse of their variance (n-3, where n is the sample size
of each study) to give greater weight to more precise
estimates. In this way, effect sizes from studies with larger
samples were weighted more heavily, since such studies
tend to produce estimates that are closer to the population
parameters (Ringquist 2013; Shadish and Haddock 2009).
Second, all individual effect sizes in Pearson’s rcorrela-
tions were transformed into Fisher’s zcorrelations to cor-
rect the slightly skewed distribution of Pearson’s raround a
given population (Ringquist 2013). Once the average effect
sizes were calculated, the results in Fisher’s zwere con-
verted back to Pearson’s rfor easier interpretation.
Due to the high levels of variability observed in effect
sizes (see Fig. 2), random-effects models were employed in
computing average effects (Borenstein et al. 2010; Ring-
quist 2013).
7
For the diversification–vulnerability rela-
tionship, after combining 76 effect sizes, we calculated a
weighted average effect size of -.009 (z= 1.64, p[.1),
with a 95% confidence interval of [-.020, .002]. This
finding indicates revenue diversification has a small nega-
tive association with financial vulnerability,
8
which means
that organizations relying on more diversified revenue
portfolios seem less vulnerable to financial shocks. How-
ever, the association itself is not statistically significant at
the .1 level. Therefore, revenue diversification might have
limited effect on financial vulnerability. For the diversifi-
cation–capacity relationship, after combining 182 effect
sizes, we found a weighted average effect size of -.012
(z= 2.51, p\.05), with a 95% confidence interval of
[-.021, -.003]. This result reveals that revenue diversi-
fication is detrimental to capacity growth, since there is a
statistically significant and negative association between
the two, even though the magnitude of the relationship
seems very small.
Moderator Analysis
We next conducted a moderator analysis, using the meta-
regression model described above, to explore under what
conditions we should expect particular effect size esti-
mates. Two methodological challenges in specifying the
model are effect size heteroscedasticity and non-indepen-
dent observations. On the one hand, effect sizes were
estimated from original studies with different sample sizes.
The variance of the effect size estimates tends to decrease
as the sample size increases, thus violating the assumption
of homoscedasticity. On the other hand, we coded multiple
effect sizes from individual studies to retain within-study
variation information. For example, we drew 14 effect
sizes from Frumkin and Keating (2011) on the diversifi-
cation–vulnerability relationship using different
7
For effect sizes on the diversification–vulnerability relationship, we
found a Qstatistic of 10,236.41 (p\.000) and an I
2
statistic of
99.3%; for effect sizes on diversification–capacity relationship, we
found a Qstatistic of 17,564.8 (p\.000) and an I
2
statistic of 99.0%.
Both groups of effect sizes demonstrate high levels of heterogeneity
that is not attributable to sampling error.
8
According to Cohen’s (1988) effect size benchmarks, a correlation
coefficient below .1 is considered a small effect.
Voluntas
123
measurements of financial vulnerability and nonprofits
from different policy fields. As a result, the effect sizes
coded from the same study were not independent of each
other but rather were clustered within that study, violating
the statistical assumption of independence. Following
Ringquist (2013), this study employed weighted least
squares (WLS) regression with clustered standard errors to
estimate the empirical model. This method weights each
effect size estimate by the inverse of the sample size to
address the heteroscedasticity concern and uses a clustered
robust parameter variance–covariance matrix to address the
(A) Studies on Revenue Diversification – Financial VulnerabilityRelationship
Andres-Alonso et al. (2015)
Cordery et al. (2013)
Frumkin & Keating (2011)
Gordon et al. (2013)
Greenlee & Trussel (2000)
Keating et al. (2005)
Rodríguez (2017)
Silva & Burger (2015)
Spyker & Deol (2014)
Tevel et al. (2015)
Trussel (2002)
Trussel & Greenlee (2004)
Study
0.10 (-0.03, 0.23)
-0.14 (-0.32, 0.05)
-0.03 (-0.06, 0.01)
0.01 (0.00, 0.02)
-0.05 (-0.08, -0.02)
-0.01 (-0.02, -0.01)
0.07 (-0.18, 0.32)
-0.16 (-0.36, 0.03)
0.02 (0.02, 0.03)
-0.32 (-0.64, 0.00)
-0.09 (-0.10, -0.08)
0.02 (-0.01, 0.04)
ES (95% CI)
0.10 (-0.03, 0.23)
-0.14 (-0.32, 0.05)
-0.03 (-0.06, 0.01)
0.01 (0.00, 0.02)
-0.05 (-0.08, -0.02)
-0.01 (-0.02, -0.01)
0.07 (-0.18, 0.32)
-0.16 (-0.36, 0.03)
0.02 (0.02, 0.03)
-0.32 (-0.64, 0.00)
-0.09 (-0.10, -0.08)
0.02 (-0.01, 0.04)
ES (95% CI)
0-.4 -.2 0.2 .4
(B) Studies on Revenue Diversification – Financial Capacity Relationship
Calabrese (2012a)
Calabrese (2012b)
Chang & Tuckman (1994)
Chikoto & Neely (2014)
Chikoto, Ling, & Neely (2016)
de Jong (2014)
Frumkin & Keating (2011)
Lam & McDougle (2016)
Lin & Wang (2016)
Myser (2016)
von Schnurbein & Fritz (2017)
Wicker & Breuer (2014)
Study
-0.01 (-0.03, 0.00)
-0.00 (-0.00, 0.00)
-0.06 (-0.11, -0.02)
-0.02 (-0.03, -0.01)
-0.02 (-0.04, -0.00)
-0.03 (-0.07, 0.01)
0.06 (0.03, 0.10)
0.08 (-0.05, 0.21)
-0.02 (-0.13, 0.08)
-0.03 (-0.05, -0.00)
-0.15 (-0.30, -0.01)
0.05 (0.02, 0.08)
ES (95% CI)
-0.01 (-0.03, 0.00)
-0.00 (-0.00, 0.00)
-0.06 (-0.11, -0.02)
-0.02 (-0.03, -0.01)
-0.02 (-0.04, -0.00)
-0.03 (-0.07, 0.01)
0.06 (0.03, 0.10)
0.08 (-0.05, 0.21)
-0.02 (-0.13, 0.08)
-0.03 (-0.05, -0.00)
-0.15 (-0.30, -0.01)
0.05 (0.02, 0.08)
ES (95% CI)
0-.4 -.2 0 .2 .4
Fig. 2 Distribution of study-
level effect sizes across existing
studies. aStudies on revenue
diversification–financial
vulnerability relationship.
bStudies on revenue
diversification–financial
capacity relationship. Note:ES
effect size, CI confidence
interval
Voluntas
123
‘‘cluster’’ nature of our data.
9
Table 3presents the meta-
regression results, and we discuss them as follows:
Timing is always a big challenge in the analysis of
financial behaviors, because nonprofit data are typically
reported on a yearly basis, but the effect of certain financial
behaviors may not fall clearly within the same one-year
period or it may last for more than just a single year. For
example, Greenlee and Trussel (2000) explored financial
vulnerability over a three-year period, using revenue
diversification in 1992 to examine financial vulnerability in
1995. Chikoto and Neely (2014) employed five-year rev-
enue growth to explore the effect of revenue diversification
in 1998 on change in financial capacity from 1998 to 2003.
Indeed, in nonprofit studies the distinction between a short-
term period and a long-term period is not clear (Bowman
2011). We included one moderator, year duration, in the
meta-regression model to explore this timing concern, with
an attempt to see whether the effect of diversification on
vulnerability or capacity differs by the length of time under
study. We find some support for the moderating effect of
year duration. In the diversification–capacity model, year
duration has a statistically significant effect (p\.05) on
the impact of revenue diversification on financial capacity:
The longer period a study examines, the smaller the
diversification effect the study finds. The effect of revenue
diversification on financial capacity thus seems to become
smaller in the long term. In contrast, in the diversification–
vulnerability model, the effect of year duration is not sta-
tistically significant (p[.1), indicating that the effect of
revenue diversification on financial vulnerability seems
consistent irrespective of the length of years under study.
Revenue diversification thus has a similar effect on finan-
cial vulnerability in both the short and the long term.
Although revenue diversification was predominately
measured by HHI in the sampled studies, the way HHI was
calculated (i.e., the number of revenue sources used) was
not always the same. For example, Frumkin and Keating
(2011) used three revenue streams (donations, earned
income, and investment income), von Schnurbein and Fritz
(2017) relied on four revenue sources (government grants,
donations, program revenue, and investment income), and
de Andre
´s-Alonso et al. (2015) used five streams (volun-
tary income, generating income, investment income,
charitable income, and other income). Literally, the inclu-
sion of more revenue sources in the HHI calculation is
more likely to better capture a nonprofit’s revenue diver-
sification efforts, which would consequently lead to more
nuanced analyses of the effect of revenue diversification.
Chikoto et al. (2016) suggest that the impact of revenue
diversification on financial performance is sensitive to the
measurement of HHI: The more revenue sources used in
the HHI calculation, the more robust effect revenue
diversification has. We addressed the sensitivity of the HHI
calculation in the meta-regression by including a modera-
tor, number of revenue sources, and concurred with Chi-
koto et al. (2016) argument. In both models, the number of
revenue sources used in the HHI calculation has a signifi-
cant effect (p\.01) on moderating the relationship
between diversification and capacity and vulnerability.
Particularly, the more revenue sources used in the HHI
calculation, the smaller the effect of diversification on
capacity and vulnerability. In sum, how HHI is constructed
matters in the exploration of the financial impacts of rev-
enue diversification.
We also examined the moderating effect of controlling
for policy field. It has been widely acknowledged that
nonprofits in different policy fields face different policy
and resource environments, which would influence their
behaviors (Stone and Sandfort 2009). For example, in the
social and human services fields, government funding
represents a significant share of nonprofit revenue, while in
the arts and culture fields, it only plays a minor role.
Table 2 Descriptive statistics Variable NMean Median SD Min Max
Effect size (diversification–capacity) 182 -.0092 -.0092 .1089 -.4898 .389
Effect size (diversification–vulnerability) 76 -.0093 .0023 .1185 -.346 .2865
Year duration (number of lagged years) 258 1.3643 1 1.7035 0 7
Number of revenue sources 258 6.2481 5 3.4405 3 19
Control for policy field 258 .5349 1 .4998 0 1
Control for size 258 .5349 1 .4998 0 1
Control for age 258 .1744 0 .3802 0 1
Peer-reviewed article 258 .7558 1 .4304 0 1
9
We also used generalized estimating equation (GEE) to test the
robustness of our meta-regression results in Table 3. The GEE
method addresses cluster observations by placing less emphasis on
effect sizes from studies generating more effect sizes and thus guards
against a few studies dominating the meta-regression results (Liang
and Zeger 1986). A detailed methodological comparison of WLS and
GEE in meta-regression can be seen in Ringquist (2013). Given that
the GEE results are consistent with the WLS results (in terms of signs
and statistical significance), we do not report the GEE results here to
save space.
Voluntas
123
Nonprofits in different fields thus have to develop different
strategies to confront unique landscapes in order to diver-
sify their revenue. Another reason why it is imperative to
control for policy field is that the degree to which a non-
profit diversifies its revenue mix is closely associated with
its mission and the nature of services provided (Chang and
Tuckman 1994; Fischer et al. 2011). Many studies use
policy field as a proxy for organization mission and service
type (e.g., Sua
´rez and Hwang 2008). In the meta-regression
analysis, we included a moderator, control for policy field,
to examine whether studies controlling for policy field
produced different effect sizes. In the diversification–vul-
nerability model, studies using a policy field control vari-
able reported significantly smaller (p\.1) average effects
compared to those without such a control. In the diversi-
fication–capacity model, studies using a policy field control
variable on average seemed to produce smaller effects
compared to those without such a control, but the differ-
ence between these two groups of effect sizes was not
significant at the .1 level. In sum, studies failing to control
for policy field might be at the risk of overestimating the
effect of revenue diversification on financial vulnerability.
We next explored whether studies controlling for size or
age made a difference. Indeed, organizational studies have
demonstrated that established organizations are less subject
to the liability of smallness and the liability of newness,
and thus are more likely to achieve survival and prosperity
(Aldrich and Auster 1986; Hannan and Freeman 1983).
Hager et al. (2004) found larger and older nonprofits in the
Minneapolis–St Paul region were less likely to close.
Similarly, Chikoto-Schultz and Neely (2016) reported that
older and larger nonprofits would enjoy better financial
stability and growth over time. In this line of reasoning,
any examination of organizational performance needs to
control for the impacts of size and age. We included two
moderators in the meta-regression analysis, control for size
and control for age, to see whether these two controls made
a difference. In both models, whether a study controls for
size matters: The effect sizes from studies controlling for
size are on average significantly larger (p\.05) than those
from studies that did not control for size. In other words,
studies that did not control for size would likely underes-
timate the effect of revenue diversification on financial
capacity and vulnerability. In contrast, controlling for age
seemed to make little difference. In both models, although
studies with a control for age produced smaller average
effect sizes than the ones without a control, the difference
was not statistically significant at the .1 level.
Finally, we included a dummy variable, peer review
article, in the meta-regression to compare the effect sizes
from published and unpublished studies to check whether
these two groups of studies produced different results. The
goal of this treatment is to address the concern for publi-
cation bias in meta-analysis: Given that studies with sta-
tistically significant findings are more likely to be
published, meta-analysis may distort the findings by giving
too much weight to the published studies (Rothstein et al.
2006). In both models, despite the slight difference in
Table 3 Meta-regression
predicting moderators of effect
size heterogeneity
Moderator Diversification–vulnerability Diversification–capacity
Year duration .0085 -.0121**
(.0078) (.0050)
Number of revenue sources -.0168*** -.0093***
(.0052) (.0023)
Control for policy field -.0822* -.0356
(.0451) (.0229)
Control for size .0499** .0456**
(.0175) (.0195)
Control for age -.0043 -.0202
(.0437) (.0244)
Peer-reviewed article .0214 .0049
(.0215) (.0317)
Constant -.1321 .0618*
(.1069) (.0313)
No. of effect sizes 76 182
No. of studies 12 12
F352 19.15
R
2
.0805 .1521
Weighted least squares (WLS) regression was used. Clustered robust standard errors (by study) in
parentheses. *p\.10. **p\.05. ***p\.01 (two-tailed test)
Voluntas
123
average effects between published and unpublished studies,
the difference was not statistically significant (p[.1). This
indicates our analysis was not seriously undermined by
publication bias. We also performed other statistical tests
to check the robustness of this finding. In both the Begg
and the Egger tests, we were not able to reject the null
hypothesis of no publication bias.
10
As a result, there was
no need for any corrections for publication bias in our
analysis.
To sum up, the average effect size analysis and mod-
erator analysis jointly suggested the following findings
across existing studies: (1) Revenue diversification overall
has no significant association with financial vulnerability,
but it does have a small negative association with financial
capacity. (2) The length of year under study moderates the
effect of revenue diversification on financial capacity: The
longer the time period under study, the smaller the effect
found. (3) The calculation of HHI matters to the effect of
revenue diversification: The more revenue sources used in
HHI calculation, the more accurate HHI in representing
revenue diversification and the smaller the effect revenue
diversification has on both financial vulnerability and
capacity. (4) Whether a study controls for policy field and
size makes a difference: Studies failing to control for
policy field may overestimate effect size estimates, and
studies failing to control for organization size could
underestimate effect size estimates. Put together, it seems
that revenue diversification has little effect on financial
vulnerability but has a slight negative effect on financial
capacity. Further, this effect on financial capacity becomes
smaller in the longer term and when a more comprehensive
measure of revenue diversification is used.
Discussion and Conclusion
Most nonprofits operate under financial austerity with
limited budgets. Thus, developing effective revenue
strategies to promote financial health and organizational
sustainability constitutes a pressing managerial challenge
for nonprofit leaders. Specifically, given that nonprofits
typically rely on a number of sources for funding, how to
achieve an optimal mix of nonprofit revenue becomes a
prominent issue. One strategy that has been widely dis-
cussed by nonprofit researchers and practitioners is revenue
diversification. Since each revenue stream has its own
benefits and risks, an intuitively reasonable practice is to
diversify across different sources to reduce financial
dependence on any single source.
11
For example, after a
thorough examination of the pros and cons of government
funding, Rushton and Brooks (2007, p. 89) wrote, ‘‘non-
profits are wise to diversifying funding sources to the
extent that they can, not relying excessively on government
funding.’’
However, although revenue diversification has been
advocated by some scholars and practitioners for a long
time, recent studies suggest that the implementation of the
diversification strategy in real-world settings is more
complex than expected. Indeed, existing empirical evi-
dence on the impact of revenue diversification on financial
health is highly inconsistent, which calls for a synthesis to
make sense of existing studies and integrate their findings
to estimate a generalized effect of revenue diversification
across studies. In this study, we first employed a biblio-
metric analysis to systematically categorize and visualize
existing studies. We then conducted a meta-analysis to
analyze the effect of revenue diversification on financial
capacity and vulnerability. Through systematically
reviewing a total of 23 existing studies and 258 effect sizes,
we found that revenue diversification has a slight detri-
mental impact on financial capacity and almost null effect
on financial vulnerability. Further, through moderator
analysis, we note that the length of years under study, the
number of revenue sources used to calculate HHI, control
for policy field, and control for organization size all con-
tribute to the variations of findings in existing studies.
Our study has a number of theoretical and practical
contributions. To begin with, we provide an empirical
integration of the extant empirical evidence and offer a
cumulative knowledge on the financial impacts of revenue
diversification across studies. Indeed, although revenue
diversification has been frequently referred to in the non-
profit literature, the empirical findings to date concerning
the financial effectiveness of revenue diversification have
been highly mixed, with few research efforts made to
integrate these disparate findings. As such, our results
consolidate previous empirical findings and provide
empirical generalizations of the effects of revenue diver-
sification on the two most studied financial outcomes (i.e.,
capacity and vulnerability) in the literature. In this way, our
meta-analysis extends the literature by bridging the diver-
gences in the literature and offering a more systematic
synthesis of previous findings. The results thus should lay
the foundation for further explorations of nonprofit finan-
cial strategies and performance.
10
In the Begg test, we found p= .86 for effect sizes on diversifi-
cation–capacity relationship, and p= .73 for effect sizes on diversi-
fication–vulnerability relationship. In the Egger test, we found p= .76
for effect sizes on diversification–capacity relationship and p= .28
for effect sizes on diversification–vulnerability relationship.
11
Here, it is assumed that nonprofits are legally allowed and have the
ability to diversify their revenue sources as they like. In reality,
however, this may not always be the case, especially for nonprofits in
some European countries. We thank an anonymous reviewer for
pointing this out.
Voluntas
123
Moreover, our findings indicate that the benefits of
revenue diversification suggested in the literature might be
overstated. Although revenue diversification has been
advocated by some scholars as an indispensable prerequi-
site for financial stability, our combination of existing
empirical evidence reveals that it has almost null effect on
financial stability or vulnerability. In addition, we find that
revenue diversification slightly damages financial capacity,
which seems to be consistent with several previous studies
(Chikoto and Neely 2014; Frumkin and Keating 2011; von
Schnurbein and Fritz 2017). In short, if nonprofits attempt
to both reduce financial vulnerability and enhance financial
capacity, some sort of concentration might turn out to be a
more appropriate revenue strategy. Therefore, the reason-
ing underlying the benefits of revenue diversification,
although theoretically appealing, may be empirically
unachievable. Part of the reason could be that nonprofits
with more diversified revenue portfolios may encounter
difficulties in simultaneously managing an array of funding
sources with different characteristics. A nonprofit in such a
difficult situation would not only incur high transaction
costs but it would also be prevented from focusing on its
main funding sources for substantive growth. This might be
especially troublesome for small- or middle-sized organi-
zations whose managers might lack the capability to
effectively handle high levels of revenue diversification.
More broadly, revenue diversification has been used
extensively as an explanatory variable to study a wide
range of nonprofit financial and non-financial behaviors,
but comparatively fewer studies have been devoted to
empirically examining the real effectiveness of revenue
diversification.
12
Our synthesis of existing studies indicates
that revenue diversification has nearly null effect on
financial vulnerability and only a small negative effect on
financial capacity. The moderator analysis further suggests
that the negative effect would become smaller when more
comprehensive measures of revenue diversification and
longer time period are taken into consideration. Overall,
the effect of revenue diversification on financial vulnera-
bility and capacity seems to be trivial. If so, nonprofit
scholars and practitioners might need to rethink the basic
assumptions they hold about the desirability and impor-
tance of revenue diversification or concentration. Clearly,
we are not in a position to discredit the body of literature
on revenue diversification, but more nuanced examination
of its theoretical foundation and empirical reality is needed.
Our findings also come with several limitations. First,
financial performance is a multi-dimensional construct
(Bowman 2011). However, we only examined the effects
of diversification on financial capacity and vulnerability
because these two financial outcomes were the most widely
studied in the revenue diversification literature where
enough effect sizes could be extracted for the meta-anal-
ysis. Other outcome indicators such as financial volatility
might be significantly affected by revenue diversification,
but we could not include them in the present analysis
because the number of relevant empirical studies was too
limited to warrant any meaningful meta-analysis. Second,
meta-analysis is not developed to explore causality; it only
examines associations between variables. Therefore, our
findings should be best understood as identifying correla-
tive rather than causal relationships. Future studies are
needed to deal with the causality issues. Third, our sample
mostly covered US charities and spanned a wide time range
during which the nonprofit sector evolved and changed
rapidly. A more balanced and homogenous sample may be
necessary to generate more convincing results. Finally, we
only explored the linear association between revenue
diversification and capacity and vulnerability, since all the
included studies did so and meta-analysis can only build on
existing studies. However, it is possible that the effect of
revenue diversification is not linear but rather curvilinear,
following a U- or inverted U-shaped pattern. For example,
some levels of diversification have a favorable impact, but
after a certain tipping point the negative effect tends to
emerge. Further research is needed to explore this
possibility.
Acknowledgements An earlier version of this manuscript was pre-
sented at the 2017 ARNOVA conference in Grand Rapids, MI. The
authors appreciated the comments from the panel audience, including
William Suhs Cleveland, Mark Hager, Georg von Schnurbein, and
others.
Compliance with Ethical Standards
Conflict of interest The authors declare that they have no conflict of
interest.
Appendix 1
See Table 4.
12
A search in Google Scholar with the term ‘‘nonprofit revenue
diversification’’ produced nearly 30,000 studies (search conducted on
November 13, 2017). However, a more systematic search in the
present project only located 23 existing studies that met our inclusion
criteria for the meta-analysis.
Voluntas
123
Table 4 Review of existing studies included in meta-analysis
#Study Publication
type
Year
coverage
Data sources Sample size Country
of
nonprofits
Policy field Field
control
Vulnerability
(lagged years)
Capacity
(lagged
years)
Size
control
Age
control
HHI
calculation
#ES
coded
1 de Andre
´s-
Alonso
et al. (2015)
Journal
Article
2008–2012 Financial
statements
228 Britain International
development
Yes Financial
Vulnerability
Index (3 year)
N/A Yes No 5 4
2 Calabrese
(2012a)
Journal
Article
1998–2003 NCCS 5350–72,691 United
States
501(c)(3)
public
charities
Yes N/A Unrestricted
net assets
(0 year)
Yes No 11 14
3 Calabrese
(2012b)
Journal
Article
1998–2003 NCCS 520,349 United
States
501(c)(3)
public
charities
No N/A Operating
reserves
(1 year)
Yes No 4 4
4 Chang and
Tuckman
(1994)
Journal
Article
1986 NCCS 113,525 United
States
501(c)(3)
public
charities
Yes N/A surplus
margin
(0 year)
No No 9 54
5 Chikoto and
Neely
(2014)
Journal
Article
1998–2003 NCCS 50,000–108,000 United
States
501(c)(3)
public
charities
Yes N/A Revenue
growth,
fund
balance
growth
(5 year)
Yes 0 3,4,13 18
6 Chikoto,
Ling, and
Neely
(2016)
Journal
Article
1998–2003 NCCS 3603, 16,298,
25,175,
103,701
United
States
501(c)(3)
public
charities
Yes N/A Revenue
growth
(5 years)
Yes Yes 3, 4, 7, 13 16
7 Cordery et al.
(2013)
Journal
Article
4 years not
specified
Financial
statements
227 New
Zealand
Sports club Yes Reduction in net
earnings
(3 years)
N/A Yes No 4 2
8 de Jong
(2014)
Dissertation 2005–2012 Financial
statements
1390–3413 Netherlands Fundraising
organizations
No N/A Revenue
growth,
program
expense
growth, and
fixed assets
growth
(1 year)
Yes No 10 20
9 Frumkin and
Keating
(2011)
Journal
Article
12 years not
specified
NCCS 56,870 United
States
501(c)(3)
public
charities
Yes Reduction in net
assets,
reduction in
program
expenditure
(1 year)
Revenue
growth,
fixed assets
growth,
program
expense
growth,
surplus
margin
(0 year)
No No 3 49
Voluntas
123
Table 4 continued
#Study Publication
type
Year
coverage
Data sources Sample size Country of
nonprofits
Policy field Field
control
Vulnerability
(lagged years)
Capacity
(lagged
years)
Size
control
Age
control
HHI
calculation
#ES
coded
10 Gordon et al.
(2013)
Journal
Article
2000–2003 NCCS 311,977 United
States
501(c)(3)
public
charities
No Negative net
assets (0 year)
N/A Yes No 3 5
11 Greenlee and
Trussel
(2000)
Journal
Article
1985–1995 NCCS 5918 United
States
501(c)(3)
public
charities
No Reduction in
program
expenditure
(3 years)
N/A No No 5 2
12 Keating et al.
(2005)
Working
Paper
1998–2000 NCCS 290,579 United
States
501(c)(3)
public
charities
No Negative net
assets
(0 year),
reduction in
program
expenditure
(1 year),
reduction in
net assets
(1 year)
N/A Yes No 3 12
13 Lam and
McDougle
(2016)
Journal
Article
2005–2007 NCCS 222 United
States
(San
Diego)
Human
services
Yes N/A Equity ratio,
months of
spending,
return on
assets,
mark up,
month of
liquidity
(3 year)
Yes Yes 3 15
14 Lin and
Wang
(2016)
Journal
Article
2008–2011 Self-
administered
survey, IRS
990 forms,
state
administrative
data
363–364 US (New
Jersey)
Human
services
Yes N/A Revenue
growth,
expense
growth
(3 year)
Yes Yes 3 2
15 Myser (2016) Dissertation 2011 NCCS 6780 United
States
501 (c) (3)
public
charities
No N/A Return on
assets
(0 year)
Yes Yes 3 1
16 Rodrı
´guez
(2017)
Dissertation 2011–2013 Financial
statement
65 Spain International
development
Yes Reduction in net
assets
(2 years)
N/A Yes Yes 7 17
17 Silva and
Burger
(2015)
Working
Paper
2000–2001,
2006–2007
Self-
administered
survey
67, 148 Uganda International
development
Yes Reduction in net
earnings
(1 years)
N/A Yes No 6 7
18 Spyker and
Deol (2014)
Working
Paper
2000–2009 Annual
Information
Return
602,629 Canada Registered
charities
No Reduction in net
assets
(3 years)
N/A Yes No 3 6
Voluntas
123
References
Aldrich, H., & Auster, E. R. (1986). Even dwarfs started small:
Liabilities of age and size and their strategic implications.
Research in Organizational Behavior, 8, 165–198.
Bielefeld, W. (1992). Non-profit funding environment relations:
Theory and application. VOLUNTAS: International Journal of
Voluntary and Nonprofit Organizations, 3(1), 48–70.
Borenstein, M. (2009). Effect sizes for continuous data. In H. Cooper,
L. V. Hedges, & J. C. Valentine (Eds.), The handbook of
research synthesis and meta-analysis (2nd ed., pp. 221–236).
New York: Russell Sage Foundation.
Borenstein, M., Hedges, L. V., Higgins, J., & Rothstein, H. R. (2010).
A basic introduction to fixed-effect and random-effects models
for meta-analysis. Research Synthesis Methods, 1(2), 97–111.
Bowman, W. (2011). Financial capacity and sustainability of
ordinary nonprofits. Nonprofit Management and Leadership,
22(1), 37–51.
Brooks, A. C. (2000). Is there a dark side to government support for
nonprofits? Public Administration Review, 60(3), 211–218.
Bryson, J. M., Crosby, B. C., & Stone, M. M. (2006). The design and
implementation of cross-sector collaborations: Propositions
from the literature. Public Administration Review, 66(S1),
44–55.
Calabrese, T. (2012a). Running on empty: The operating reserves of
U.S. nonprofit organizations. Nonprofit Management and Lead-
ership, 23(3), 281–302.
Calabrese, T. D. (2012b). The accumulation of nonprofit profits: A
dynamic analysis. Nonprofit and Voluntary Sector Quarterly,
41(2), 300–324.
Carroll, D. A., & Stater, K. J. (2009). Revenue diversification in
nonprofit organizations: Does it lead to financial stability?
Journal of Public Administration Research and Theory, 19(4),
947–966.
Chang, C. F., & Tuckman, H. P. (1994). Revenue diversification
among non-profits. Voluntas: International Journal of Voluntary
and Nonprofit Organizations, 5(3), 273–290.
Chang, C. F., & Tuckman, H. P. (1996). The goods produced by
nonprofit organizations. Public Finance Quarterly,24(1),
25–43.
Chen, C. (2004). Searching for intellectual turning points: Progres-
sive knowledge domain visualization. Proceedings of the
National Academy of Sciences, 101(s1), 5303–5310.
Chikoto, G. L., Ling, Q., & Neely, D. G. (2016). The adoption and
use of the Hirschman-Herfindahl Index in nonprofit research:
Does revenue diversification measurement matter? Voluntas:
International Journal of Voluntary and Nonprofit Organiza-
tions, 27(3), 1425–1447.
Chikoto, G. L., & Neely, D. G. (2014). Building nonprofit financial
capacity: The impact of revenue concentration and overhead
costs. Nonprofit and Voluntary Sector Quarterly, 43(3),
570–588.
Chikoto-Schultz, G. L., & Neely, D. G. (2016). Exploring the nexus
of nonprofit financial stability and financial growth. Voluntas:
International Journal of Voluntary and Nonprofit Organiza-
tions, 27(6), 2561–2575.
Cohen, J. (1988). Statistical power analysis for the behavior science.
Hillside: Lawrance Eribaum Association.
Cordery, C. J., Sim, D., & Baskerville, R. F. (2013). Three models,
one goal: Assessing financial vulnerability in New Zealand
amateur sports clubs. Sport Management Review, 16(2),
186–199.
de Andre
´s-Alonso, P., Garcia-Rodriguez, I., & Romero-Merino, M.
E. (2015). The dangers of assessing the financial vulnerability of
Table 4 continued
#Study Publication
type
Year
coverage
Data sources Sample size Country of
nonprofits
Policy field Field
control
Vulnerability
(lagged years)
Capacity
(lagged
years)
Size
control
Age
control
HHI
calculation
#ES
coded
19 Tevel et al.
(2015)
Journal
Article
2009–2011 Financial reports 41 Israel performing arts Yes Tuckman and
Change index
(2 years)
N/A Yes 0 3 1
20 Trussel and
Greenlee
(2004)
Journal
Article
1992–1995 NCCS 6795 United
States
501 (c) (3)
public
charities
Yes Reduction in net
assets
(3 years)
N/A Yes 0 5 4
21 Trussel
(2002)
Journal
Article
1996–1999 NCCS 94,002 United
States
501 (c) (3)
public
charities
Yes Reduction in net
assets
(3 years)
N/A Yes 0 5 2
22 von
Schnurbein
and Fritz
(2017)
Journal
Article
(2005–2012) Audited annual
reports
191 Swiss Certified
fundraising
charities
No N/A Revenue
growth (7-
year)
Yes Yes 4 1
23 Wicker and
Breuer
(2014)
Journal
Article
2011 Self-
administered
survey
1080 Germany Sports
governing
bodies
Yes N/A Surplus
(0 year)
Yes No 19 2
Voluntas
123
nonprofits using traditional measures. Nonprofit Management
and Leadership, 25(4), 371–382.
de Jong, F. (2014). Revenue diversification in Dutch charity
organizations: Does it lead to growth? (Doctoral dissertation).
Tilburg University.
de los Mozos, I. S. L., Duarte, A. R., & Ruiz, O. R. (2016). Resource
dependence in non-profit organizations: Is it harder to fundraise
if you diversify your revenue structure? Voluntas: International
Journal of Voluntary and Nonprofit Organizations, 27(6),
2641–2665.
Despard, M. R., Nafziger-Mayegun, R. N., Adjabeng, B. K., &
Ansong, D. (2017). Does revenue diversification predict financial
vulnerability among non-governmental organizations in sub-
Saharan Africa? Voluntas: International Journal of Voluntary
and Nonprofit Organizations, 28(5), 2124–2144.
DiMaggio, P., & Powell, W. W. (1983). The iron cage revisited:
Collective rationality and institutional isomorphism in organi-
zational fields. American Sociological Review, 48(2), 147–160.
Fischer, R. L., Wilsker, A., & Young, D. R. (2011). Exploring the
revenue mix of nonprofit organizations: Does it relate to
publicness? Nonprofit and Voluntary Sector Quarterly, 40(4),
662–681.
Fleiss, J. L., & Berlin, J. A. (2009). Effect sizes for dichotomous data.
In H. Cooper, L. V. Hedges, & J. C. Valentine (Eds.), The
handbook of research synthesis and meta-analysis (2nd ed.,
pp. 237–254). New York: Russell Sage Foundation.
Foster, W., & Fine, G. (2007). How nonprofits get really big. Stanford
Social Innovation Review, 5(2), 46–55.
Foster, M. K., & Meinhard, A. G. (2005). Diversifying revenue
sources in Canada: Are women’s voluntary organizations
different? Nonprofit Management and Leadership, 16(1), 43–60.
Froelich, K. A. (1999). Diversification of revenue strategies: Evolving
resource dependence in nonprofit organizations. Nonprofit and
Voluntary Sector Quarterly, 28(3), 246–268.
Frumkin, P., & Keating, E. K. (2011). Diversification reconsidered:
The risks and rewards of revenue concentration. Journal of
Social Entrepreneurship, 2(2), 151–164.
Geyskens, I., Krishnan, R., Steenkamp, J.-B. E. M., & Cunha, P. V.
(2009). A review and evaluation of meta-analysis practices in
management research. Journal of Management, 35(2), 393–419.
Glass, G. V. (1976). Primary, secondary, and meta-analysis of
research. Educational Researcher, 5(10), 3–8.
Gordon, T. P., Fischer, M., Greenlee, J., & Keating, E. K. (2013).
Warning signs: Nonprofit insolvency indicators. International
Research Journal of Applied Finance,4(3), 343–378.
Grasse, N. J., Whaley, K. M., & Ihrke, D. M. (2016). Modern
portfolio theory and nonprofit arts organizations: Identifying the
efficient frontier. Nonprofit and Voluntary Sector Quarterly,
45(4), 825–843.
Greenlee, J. S., & Trussel, J. M. (2000). Predicting the financial
vulnerability of charitable organizations. Nonprofit Management
and Leadership, 11(2), 199–210.
Grønbjerg, K. A. (1993). Understanding nonprofit funding: Managing
revenues in social services and community development orga-
nizations. San Francisco: Jossey-Bass.
Hager, M. A. (2001). Financial vulnerability among arts organiza-
tions: A test of the Tuckman-Chang measures. Nonprofit and
Voluntary Sector Quarterly, 30(2), 376–392.
Hager, M. A., Galaskiewicz, J., & Larson, J. A. (2004). Structural
embeddedness and the liability of newness among nonprofit
organizations. Public Management Review, 6(2), 159–188.
Hannan, M. T., & Freeman, J. (1983). The liability of newness: Age
dependence in organizational death rates. American Sociological
Review, 48(5), 692–710.
Hodge, M., & Piccolo, R. (2005). Funding source, board investment
techniques, and financial vulnerability in nonprofits. Nonprofit
Management & Leadership,16(2), 171–190.
Jegers, M., & Verschueren, I. (2006). On the capital structure of non-
profit organisations: An empirical study for Californian organ-
isations. Financial Accountability & Management, 22(4),
309–329.
Keating, E., Fischer, M., Gordon, T., & Greenlee, J. (2005). Assessing
financial vulnerability in the nonprofit sector (Faculty Research
Working Paper Series, Paper No. 27). Cambridge, MA: Hauser
Center for Nonprofit Organizations.
Kerlin, J. A., & Pollak, T. H. (2011). Nonprofit commercial revenue:
A replacement for declining government grants and private
contributions? American Review of Public Administration, 41(6),
686–704.
Kingma, B. R. (1993). Portfolio theory and nonprofit financial
stability. Nonprofit and Voluntary Sector Quarterly, 22(2),
105–119.
Lam, M., & McDougle, L. (2016). Community variation in the
financial health of nonprofit human service organizations: An
examination of organizational and contextual effects. Nonprofit
and Voluntary Sector Quarterly, 45(3), 500–525.
Liang, K. Y., & Zeger, S. L. (1986). Longitudinal data analysis using
generalized linear models. Biometrika, 73(1), 13–22.
Light, P. C. (2004). Sustaining nonprofit performance: The case for
capacity building and the evidence to support it. Washington,
DC: Brookings Institution Press.
Lin, W., & Wang, Q. (2016). What helped nonprofits weather the
great recession? Evidence from human services and community
improvement organizations. Nonprofit Management & Leader-
ship,26(3), 257–276.
Lipsey, M. W. (2009). Identifying interesting variables and analysis
opportunities. In H. Cooper, L. V. Hedges, & J. C. Valentine
(Eds.), The handbook of research synthesis and meta-analysis
(2nd ed., pp. 147–158). New York: Russell Sage Foundation.
Lu, J. (2016). The philanthropic consequence of government grants to
nonprofit organizations. Nonprofit Management and Leadership,
26(4), 381–400.
Markowitz, H. (1952). Portfolio seclection. The Journal of Finance,
7(1), 77–91.
Mayer, W. J., Wang, H., Egginton, J. F., & Flint, H. S. (2014). The
impact of revenue diversification on expected revenue and
volatility for nonprofit organizations. Nonprofit and Voluntary
Sector Quarterly, 43(2), 374–392.
Meyer, J. W., & Scott, W. R. (1992). Organizational environments:
Ritual and rationality. Newcastle upon Tyne: Sage.
Myser, S. (2016). Financial health of nonprofit organizations
(Doctoral dissertation). University of Kansas.
Pfeffer, J., & Salanick, G. R. (2003). The external control of
organizations: A resource dependence perspective. Stanford,
CA: Stanford University Press.
Reed, J. G., & Baxter, P. M. (2009). Using reference databases. In H.
Cooper, L. V. Hedges, & J. C. Valentine (Eds.), The handbook of
research synthesis and meta-analysis (2nd ed., pp. 73–101). New
York: Russell Sage Foundation.
Ringquist, E. (2013). Meta-analysis for public management and
policy. Hoboken: Wiley.
Rodrı
´guez, I
´. G. (2017). Governance and financial vulnerability of
nonprofit organizations: An analysis of the NGDOs from Spain
and the United Kingdom (Doctoral dissertation). University of
Burgos.
Rothstein, H. R., Sutton, A. J., & Borenstein, M. (2006). Publication
bias in meta-analysis: Prevention, assessment and adjustments.
Hoboken: Wiley.
Rushton, M., & Brooks, A. C. (2007). Government funding of
nonprofit organizations. In D. Young (Ed.), Financing
Voluntas
123
nonprofits: Putting theory into practice (pp. 69–91). Lanham:
Altamira Press.
Salamon, L. M. (1995). Partners in public service: Government-
nonprofit relations in the modern welfare state. Baltimore, MD:
Johns Hopkins University Press.
Schmidt, F. L., & Hunter, J. E. (2014). Methods of meta-analysis:
Correcting error and bias in research findings. Thousand Oaks,
CA: Sage publications.
Scott, W. R. (1987). Organizations: Rational, natural, and open
systems. Englewood Cliffs, NJ: Prentice-Hall.
Shadish, W., & Haddock, C. K. (2009). Combining estimates of effect
size. In H. Cooper, L. V. Hedges, & J. C. Valentine (Eds.), The
handbook of research synthesis and meta-analysis (2nd ed.,
pp. 257–278). New York: Russell Sage Foundation.
Silva, B., & Burger, R. (2015). Financial vulnerability: An empirical
study of Ugandan NGOs (No. 1515). Liege: CIRIEC-Universite
´
de Lie
`ge.
Small, H. (2003). Paradigms, citations, and maps of sciences: A
personal history. Journal of the American Society for Informa-
tion Science and Technology, 54(5), 394–399.
Spyker, C., & Deol, H. (2014). Doing too much? Financial
vulnerability of Canadian registered charities. In Presentation
at 2014 Canadian Academic Accounting Association (CAAA)
annual conference.
Steinberg, R. (1990). Profits and incentive compensation in nonprofit
firms. Nonprofit Management and Leadership, 1(2), 137–151.
Stone, M. M., & Sandfort, J. R. (2009). Building a policy fields
framework to inform research on nonprofit organizations.
Nonprofit and Voluntary Sector Quarterly, 38(6), 1054–1075.
Sua
´rez, D. F., & Hwang, H. (2008). Civic engagement and nonprofit
lobbying in California, 1998-2003. Nonprofit and Voluntary
Sector Quarterly, 37(1), 93–112.
Tevel, E., Katz, H., & Brock, D. M. (2015). Nonprofit financial
vulnerability: Testing competing models, recommended
improvements, and implications. Voluntas: International
Journal of Voluntary and Nonprofit Organizations, 26(6),
2500–2516.
Trussel, J. M. (2002). Revisiting the prediction of financial vulner-
ability. Nonprofit Management & Leadership,13(1), 17–31.
Trussel, J. M., & Greenlee, J. (2004). A financial rating system for
nonprofit organizations. Research in Government and Nonprofit
Accounting,11(1), 105–128.
Tuckman, H. P., & Chang, C. F. (1991). A methodology for
measuring the financial vulnerability of charitable nonprofit
organizations. Nonprofit and Voluntary Sector Quarterly, 20(4),
445–460.
Tuckman, H. P., & Chang, C. F. (1992). Nonprofit equity: A
behavioral model and its policy implications. Journal of Policy
Analysis and Management,11(1), 76–87.
von Schnurbein, G., & Fritz, T. M. (2017). Benefits and drivers of
nonprofit revenue concentration. Nonprofit and Voluntary Sector
Quarterly, 46(5), 922–943.
Weisbrod, B. A. (1998). To profit or not to profit: The commercial
transformation of the nonprofit sector. New York: Cambridge
University Press.
Wicker, P., & Breuer, C. (2013). Understanding the importance of
organizational resources to explain organizational problems:
Evidence from nonprofit sport clubs in Germany. Voluntas:
International Journal of Voluntary and Nonprofit Organizations,
24(2), 461–484.
Wicker, P., & Breuer, C. (2014). Examining the financial condition of
sport governing bodies: The effects of revenue diversification
and organizational success factors. VOLUNTAS: International
Journal of Voluntary and Nonprofit Organizations,25(4),
929–948.
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Voluntas
123
