Does Government Punish Nonprofits for High Administrative Costs in Contracting Decisions?

Abstract

As government financing of nonprofit organizations to deliver services and implement policies has become a common practice in the public administration landscape, the question of what factors affect government’s source selection has emerged as a significant one. Within this strand of research, how nonprofits’ administrative costs affect their receipt of government contracts is still not fully understood. This article explores that relationship using a large panel data set of U.S.-based international development nonprofits from 1967 to 2014. Different model specifications consistently demonstrate an inverted U-shaped relationship between a nonprofit’s level of administrative costs and its amount of government contracts. In particular, as a nonprofit’s level of administrative costs increases, its amount of government contracts will initially increase, but after its level of administrative costs reaches approximately 16% to 18% of total expenses, further increases in the nonprofit’s level of administrative costs will reduce its amount of government contracts. These findings have implications for both public and nonprofit management.

Keywords

government contracting government–nonprofit relations administrative costs overhead costs

Introduction

Government financing of nonprofit organizations for service delivery and policy implementation has been a common practice in the public administration landscape in recent decades (Salamon, 1995; Smith & Lipsky, 1993). Through “hollowing out” of the state, a broad array of public services is increasingly provided by nonprofit contractors and grantees (Milward & Provan, 2000). For example, Steuerle et al. (2017) estimated that the U.S. nonprofit sector received approximately US$497 billion from the federal government in fiscal year (FY) 2015 for service delivery, which was 187% higher than in FY 1980. This close funding relationship has attracted significant attention from public and nonprofit management scholars (e.g., Amirkhanyan & Lambright, 2017; Brown et al., 2018; Lamothe, 2015; Salamon & Toepler, 2015; Suárez & Esparza, 2017). In this body of government–nonprofit relations literature, a growing number of studies explore what factors affect government’s source selection or, framed differently, what type of nonprofits are more preferred by funding agencies (e.g., Garrow, 2010; Lu, 2015; Marwell & Gullickson, 2013; Stone et al., 2001; Suárez, 2010). This body of literature has examined the various organizational and environmental factors that influence nonprofits’ receipt of government funding.

Within this strand of research, findings regarding how nonprofits’ administrative costs affect their receipt of government funding are still inconclusive. The existing literature has proposed two competing lines of argument. On one hand, the price hypothesis suggests that administrative costs represent a diversion of organizational resources from program outputs; high levels of administrative costs are thus considered to be signals of inefficiency and waste (Tinkelman & Mankaney, 2007; Weisbrod & Dominguez, 1986). Therefore, nonprofits charging higher overhead costs in their funding proposals would be less preferred by funding agencies because these nonprofits might not use government funding in an efficient manner. On the other hand, the quality hypothesis indicates that administrative costs represent organizational capacity. Nonprofits investing more in administrative infrastructure and operations are more likely to improve their organizational effectiveness and provide quality services (Gregory & Howard, 2009; Hager et al., 2004), and thus should be perceived by funding agencies as more competent service providers. The disparity in the literature indicates the complex nature of government–nonprofit funding relationship and calls for empirical studies to provide a definitive answer.

Unfortunately, few empirical studies on this topic exist. Ashley and Van Slyke’s (2012) study represents the first effort to directly test this relationship. Using cross-sectional state-level grants data from Georgia, the authors examined the effect of a nonprofit’s level of administrative costs on the amount of government grants received. The study finds only a weak linear association between the two, a relationship that is not strong enough to support either the price or the quality hypothesis. Relatedly, Ashley and Faulk (2010) examined the effect of administrative costs on nonprofits’ amount of foundation grants and concluded that no significant linear relationship seems to exist. The authors of both studies determined that their findings were inconclusive and called for more future research. However, to our knowledge, there are no later empirical studies that address the relationship between nonprofits’ administrative costs and government funding, even though the research on nonprofit overhead costs and efficiency ratios has flourished in recent years (e.g., Charles, 2018; Ecer et al., 2017; Gneezy et al., 2014; Parsons et al., 2017; Wong & Ortmann, 2016).

Building on Ashley and Van Slyke (2012), this study continues to explore the effect of nonprofits’ administrative costs on their receipt of government contracts. The unit of analysis in the study is individual nonprofits. We argue that government contracting decisions involve a balance between price and quality, which implies that neither the price hypothesis nor the quality hypothesis could completely capture the dynamic relationship between the two variables. Specifically, as a nonprofit’s level of administrative costs increases, its amount of government contracts will initially increase (due to the quality concern), but after a tipping point, further increases in the nonprofit’s level of administrative costs reduce its amount of government contracts (due to the price concern). In sum, we hypothesize that the relationship between a nonprofit’s level of administrative costs and its amount of government contracts is not linearly positive or negative but rather curvilinear, following an inverted U-shaped pattern.

We test this hypothesis using a large panel data set of international development nonprofits registered with the United States Agency for International Development (USAID) for funding purposes from 1967 to 2014. Different model specifications consistently support the inverted U-shaped relationship between the two variables and indicate that the tipping point appears when administrative costs approximate 16% to 18% of an organization’s total expenses. The findings have implications for both public and nonprofit organizations in effectively managing the contractual relationship. In particular, for public managers, in addition to considering whether the limitations on contractors’ administrative costs may undermine nonprofits’ ability to function as capable partners, they might set minimum administrative costs rates as signals to inform nonprofits of the necessary administrative costs to act as competent service providers. For nonprofit managers, organizations that make reasonable investments in staffing and infrastructure to carry out their missions would be better positioned in the competition for government funding.

The remaining sections of the manuscript are arranged as follows. The section “Theoretical Framework” reviews the relevant literature and develops the theoretical framework. We then introduce the data, variables, and measurements in the section “Method.” In the section “Results,” we present the data analysis results and robustness tests. The final section discusses the implications of the findings as well as the study limitations to inform future studies.

Theoretical Framework

Administrative costs represent the proportion of an organization’s resources allocated to administrative operations such as rent, utilities, management systems, and administrative salaries. These administrative costs are usually included in government funding proposals as indirect costs—costs that are incurred for common objectives but cannot be readily tied to any particular programs (U.S. Government Accountability Office [GAO], 2010).¹ The U.S. Office of Management and Budget’s OMB Circular A-122 guidance on cost principles for nonprofits further classifies indirect costs into two broad categories: facilities (e.g., buildings and equipment) and administration (e.g., accounting and personnel). When competing for government funding, nonprofits typically need to include indirect costs information in their proposals based on their actual organizational expenditure information along with supporting documentation. Funding agencies are usually required to review applicants’ indirect costs as one criterion in their source selections and to negotiate indirect cost rates with nonprofits, paying attention to the percentage of government funding devoted to administrative costs (Cohen & Eimicke, 2008; Kelman, 2002). For example, Ashley and Van Slyke (2012) documented that some state agencies explicitly consider the ratio of direct serivce costs to administrative costs when evaluating grant proposals. In fact, although administrative costs are one common component of the government contracting process, how those costs affect government contracting decisions is still not well understood. The literature has proposed two directions for the relationship.

The Price Hypothesis

The price line of arguments builds primarily on Weisbrod and Dominguez’s (1986) classical work. To these authors, administrative costs represent organizational resources that are not used directly on charitable output—the true mission of nonprofits—and can be considered part of “the cost to a donor of purchasing one dollar’s worth of the organization’s output” (Weisbrod & Dominguez, 1986, p. 87). Administrative costs ratio, the ratio of an organization’s administrative costs to its total expenses, thus becomes an indicator of efficiency of a nonprofit in turning resources into final output: the larger the proportion of expenses an organization devotes to its administration, the smaller the proportion that is dedicated to the provision of charitable output and the higher the price of the output to funders. In this way, higher administrative costs ratios signal higher levels of inefficiency and waste, which are perceived negatively by donors and other constituents such as donors and watchdog organizations. Gneezy et al. (2014) termed this phenomenon “overhead aversion.” Indeed, empirical research documents that nonprofits with high levels of administrative costs ratios receive less private donations and constituent support (Bowman, 2006; Sloan, 2009; Tinkelman & Mankaney, 2007; Wong & Ortmann, 2016).

When governments deliberate funding proposals from nonprofits, it seems reasonable to argue that government funders will favor organizations that charge lower administrative costs, ceteris paribus, to reduce procurement costs and maximize service outputs. After all, one critical motivation for contracting out and engaging in cross-sectoral collaboration is to make service provision more cost-effective through market mechanisms (Bel et al., 2010; Gazley & Brudney, 2007; Savas, 2000). For example, Warner and Hefetz (2004, 2009) reported that in their multiple rounds of national surveys of U.S. local governments, public administrators consistently listed cost saving as the top reason for service contracting. In this sense, lower administrative costs signal higher organizational efficiency, which could be valued by funding agencies when seeking service providers. The Nonprofit Overhead Costs Project reported that nonprofits felt the highest pressure from government funders to limit their administrative costs, much higher than from other constituents including watchdogs, donors, and foundations (Wing & Hager, 2004). A similar finding was also documented in a recent survey of California nonprofits (Berlin et al., 2017). In sum, in the price line of reasoning, if government wants its dollar to maximize program output, organizations asking for lower administrative costs would be preferred. Thus, nonprofits with lower administrative costs will receive more government contracts.

The Quality Hypothesis

The pursuit of lower administrative costs in government contracting creates substantial pressures on nonprofit management. A serious consequence is that it pushes nonprofits to underinvest in infrastructure and operations such as offering noncompetitive staff salaries and benefits, underinvesting facilities and technology, and lacking professional management, thereby creating a “nonprofit starvation cycle” that compromises the scope and quality of services offered in the short-term and undermines the viability of the nonprofit sector as an effective partner in service delivery in the long-term (Gregory & Howard, 2009; Lecy & Searing, 2015). Indeed, nonprofits need a reasonable level of administrative costs to maintain their organizational effectiveness (GAO, 2010; Marwell & Calabrese, 2014). For example, Tuckman and Chang’s (1991) and Greenlee and Trussel’s (2000) seminal works indicate that nonprofits with lower administrative costs are more financially vulnerable and distressed. Chikoto and Neely (2014) demonstrated that nonprofits with low administrative costs experience significant declines in financial capacity over time. In the Nonprofit Overhead Costs Project, Hager et al. (2004, p. 3) observed that “nonprofits that spend too little on infrastructure have more limited effectiveness than those that spend more reasonably.”

Based on these observations, the quality line of argument suggests that government funders should prefer organizations charging reasonably high levels of administrative costs, ceteris paribus, for at least two reasons. First, in market exchanges, one gets what one pays for. Lowest price does not necessarily guarantee the best value. Rather, relying on nonprofits that possess the necessary administrative infrastructure and management expertise may help ensure satisfactory contracting performance and minimize the possibility of contracting failure. Second, the increasing interdependence between government and nonprofits in service delivery implies that government–nonprofit contracting extends beyond short-term spot transactions to long-term reciprocal collaborations (Saidel, 1991; Van Slyke, 2007). This institutional arrangement means that government performance becomes largely contingent on nonprofit contractors (Kettl, 2002) and that nonprofits are similarly dependent on government for reliable funding to achieve their missions (Salamon, 1995). In this context, a nonprofit sector with improved infrastructure and capacity would help governments achieve better performance and public trust (Kettl, 2017). There is empirical evidence to support the quality line of reasoning. For example, research documents that nonprofits equipped with more sophisticated organizational structures and a professional workforce receive significantly more government funding than their counterparts (e.g., Lu, 2015; Stone et al., 2001; Suárez, 2010). In sum, to ensure short-term service quality and long-term collaboration sustainability, government should favor organizations that charge reasonably high levels of administrative costs. If so, nonprofits with higher administrative costs will receive more government contracts.

A Possible Balance

We argue that neither of these hypotheses can completely capture the dynamic relationship between nonprofits’ administrative costs and their receipt of government contracts. Indeed, government contracting decisions always involve a balance between price and quality (Cohen & Eimicke, 2008). As Kelman (2002, p. 285) wrote, “government gets good prices and good performance from the firms with which it does business.” For example, the Federal Acquisition Regulation (FAR 15.304[c]) requires that “Price or cost to the Government shall be evaluated in every source selection” and that “The quality of the product or service shall be addressed in every source selection through consideration of one or more non-cost evaluation factors such as . . . technical excellence, management capacity, personnel qualifications, . . .” We therefore hypothesize that the relationship between a nonprofit’s level of administrative costs and their amount of government contracts is not linearly positive or negative but rather curvilinear.

All other things being the same, government awards should mostly go to organizations charging the lowest administrative costs (the price concern). However, when administrative costs are below a reasonable level, the concern for quality will tend to dominate contracting decisions. In this case, organizations proposing unreasonably low administrative costs will not be selected. Thus, we hypothesize a nonlinear, inverted U-shaped relationship between a nonprofit’s level of administrative costs and the amount of government contracts it receives. The initial upward slope of the inverted U-shaped function, indicating a positive relationship between administrative costs and government contracts, could represent the quality hypothesis. Nonprofits with extremely low administrative costs signal weak infrastructure and effectiveness, and thus may be less likely to receive government contracts. As a result, with other factors remaining the same, as a nonprofit’s administrative costs increase, it will receive more government contracts. In addition, the price hypothesis could help explain how the upward slope eventually changes direction. When an organization’s administrative costs reach a reasonable level (i.e., the organization possesses the necessary capacity to meet government funding requirements), the concern for price and efficiency may guide government contracting decisions. At this stage, other things being the same, government will prefer organizations with lower administrative costs to maximize service output and procurement efficiency. As a result, after a certain point, further increases in the administrative costs will reduce government contracts.

In sum, we hypothesize an underlying inverted U-shaped relationship between a nonprofit’s level of administrative costs and the amount of government contracts it receives: increases in the administrative costs initially produce increases in government contracts, but after a certain point, subsequent increases in the administrative costs lead to declines in government contracts. Figure 1 summarizes our argument.

Figure 1.

The relationship between administrative costs and government contracts.

Method

We test our hypothesis using a sample of international development nonprofits registered with USAID for funding purposes. Since the 1950s, USAID has partnered with international development nonprofits in the provision of international assistance and development services around the world to support U.S. foreign policy goals and address a broad range of development, humanitarian, and health challenges (Lindenberg, 1999; Lu & Zhao, 2019; McCleary & Barro, 2008; Stoddard, 2012). Instead of implementing development programs in-house, USAID typically employs funding mechanisms such as contracts and grants to acquire services from international development nonprofits to leverage their community connections and expertise (GAO, 2002). For example, in FY 2014, nearly half of USAID’s procurement went to these nonprofits (Tarnoff, 2015).

The data for this study were collected from USAID’s annual Report of Voluntary Agencies Engaged in Overseas Relief and Development (i.e., VolAg Report). The report details the expenditure and revenue information of each nonprofit registered with USAID in a given FY, based on the self-reported financial information (supported by audited financial statements) the organization provides through its annual registration process. In fact, nonprofits seeking to work with USAID must first apply to USAID’s Registry and go through a pre-screening process that determines the eligibility to compete for most USAID funding. During this process, nonprofits have to submit supporting documentation on their legal, governance, and financial conditions and are then subject to USAID review before entering into merit-based competitions for international assistance funding. According to the 2016 VolAg Report, in FY 2014, there were 485 U.S.-based nonprofits and 106 international-based nonprofits registered with USAID, and among these nonprofits, 38% of the registered nonprofits received a total of US$2.8 billion support from USAID through contracts, grants, and in-kind assistance.

We gained access to the VolAg Reports (1967–2016) that provide financial information on registered nonprofits for the period from FY 1967 to FY 2014. To ensure the comparability of the units in our sample, we examined only U.S.-based nonprofits, which according to USAID’s definition are organizations organized under U.S. laws that have their headquarters in the United States. Moreover, because the rules and procedures associated with different types of government funding (e.g., contracts, grants, and in-kind support) may not be fully comparable (Eger & McDonald, 2017), we focused on government contracts in this study.²

After pulling all the data from the reports, we cleaned the data in the following ways. First, we eliminated observations that report a negative value for revenue or expense items (e.g., government contracts, administrative expenses, and fundraising expenses). Second, we excluded observations with zero administrative and fundraising expenses because these data are highly controversial in the literature and considered to be unreliable (Krishnan et al., 2006; Tinkelman & Mankaney, 2007). Third, we deleted observations representing organizations with zero government contracts for all years for which the organization was included in the sample, as those organizations do not contribute to variation in government contracts (Andreoni & Payne, 2003). Fourth, we used the consumer price index to adjust all the financial data to 2014-year dollars for inflation. After dropping missing data, our final sample used in the regression consisted of a panel of 1,036 organizations with 7,360 observations for the entire time period under study.

Following the practice in existing literature (e.g., Ashley & Van Slyke, 2012; Lu, 2015; McCleary & Barro, 2008), the dependent variable, government contracts, is measured by an organization’s total amount of federal government contracts (in log form).³ The key explanatory variable, administrative costs, is calculated as the proportion of an organization’s total expenses used for administrative and management activities. The squared term, administrative costs², is used to examine the nonlinear relationship between administrative costs and government contracts. In addition, given that an organization’s revenue structure is strongly affected by its expenditure structure, following previous studies (e.g., Ashley & Faulk, 2010; Ashley & Van Slyke, 2012; Chikoto & Neely, 2014), we also include two expense variables as controls: program expenses (measured by the proportion of an organization’s total expenses used for domestic and international programs) and fundraising expenses (measured by the proportion of an organization’s total expenses used for fundraising). Finally, we control for the effect of organization size (measured by an organization’s total revenue in log form).

Tables 1 and 2 present the descriptive statistics and correlations for all the variables used in the analysis, respectively. The main reason for reporting percentile information instead of minimum and maximum values is that doing so allows us to better show the distribution of the variables. In particular, the dependent variable government contracts (in US$1,000,000) has a mean value of 1.04 and a median value of 1.08 and varies widely among organizations, with the 75th percentile being approximately 27 times greater than the 25th percentile. The key independent variable administrative costs has a mean value of 9.85% and a median value of 8.74%, and its distribution is quite flat, with the 75th percentile being approximately 3 times larger than the 25th percentile. The correlation analysis of administrative costs and government contracts finds a correlation coefficient (r) of .1403, which seems to imply a positive association between the two variables.

Table 1.

Descriptive Statistics (N = 7,360).

Variable	Mean	Median	25% percentile	75% percentile	SD
Government contracts (US$1,000,000)	1.04	1.08	0.21	5.69	3.4
Administrative costs (%)	9.85	8.74	4.23	14.25	6.98
Program expenses (%)	54.09	55.61	26.26	85.93	30.73
Fundraising expenses (%)	3.89	1.72	0.08	6.14	5.04
Organization size (log)	16.07	15.98	14.71	17.42	1.89

Table 2.

Correlations Between Variables.

Variables	(1)	(2)	(3)	(4)	(5)
Government contracts	1.0000
Administrative costs	0.1403	1.0000
Program expenses	0.0805	−0.7937	1.0000
Fundraising expenses	−0.3056	−0.0137	−0.5530	1.0000
Organization size	0.1205	−0.2613	0.1983	0.0337	1.0000

Results

We conduct a series of regression analyses to further explore the statistical relationship between administrative costs and government contracts. Following many previous studies (e.g., Ashley & Van Slyke, 2012; McCleary & Barro, 2008; Nikolova, 2015), all the right-hand-side variables are taken with a 1-year lag to address potential simultaneity issues because financial revenue in year t is most likely dependent on financial expenses in year t – 1. In particular, the administrative costs included in nonprofits’ funding proposals in year t are mostly likely based on the organization’s administrative costs in year t – 1. This treatment also allows us to better explore the causality between the variables.

We specify the regression model using fixed-effects and random-effects models. Table 3 presents the results. We first explore the linear association between administrative costs and government contracts by excluding the squared term. Both the fixed-effects estimation in Column 1 and the random-effects estimation in Column 3 find a positive linear association between administrative costs and government contracts, indicating that government contracts will grow as nonprofits increase their administrative expenses. However, the significance level for both associations is only 10%, which seems to concur with Ashley and Van Slyke’s (2012) finding of a weak linear relationship between the two variables.

Table 3.

Regression Results.

Variables	Fixed-effects		Random-effects
Variables	(1)	(2)	(3)	(4)
Administrative costs	0.0460* (0.0221)	0.0450* (0.0221)	0.0428* (0.0191)	0.0509** (0.0192)
Administrative costs²		−0.0014*** (0.0001)		−0.0015*** (0.0001)
Program expenses	0.0344(0.0218)	0.0332(0.0218)	0.0270(0.0188)	0.0273(0.0188)
Fundraising expenses	0.0215(0.0065)	0.0242(0.0065)	0.0223(0.0063)	0.0236(0.0062)
Organization size	0.5232*** (0.0410)	0.5201*** (0.0410)	0.7074*** (0.0270)	0.7052*** (0.0270)
Constant	0.0939(2.2253)	0.2730(2.2221)	−2.0584(1.8942)	−2.0452(1.8923)
R ²	.2101	.2182	.2201	.2235
Number of organizations	1,036	1,036	1,036	1,036
Observations	7,360	7,360	7,360	7,360

Note. Dependent variable is total amount of government contracts in log form at year t. All the right-side variables take the value at year t – 1. Organziation fixed-effects and year indicator variables are included in each regression. Standard errors in parentheses.

p < .1. **p < .05. ***p < .01.

We then investigate the nonlinear relationship between the two variables by including the squared term in Columns 2 and 4. The Hausman test produced a p value of .0000 with a chi-square value of 58.6, indicating that the fixed-effects model is more suitable than the random-effects model. As such, the regression results in Column 2 represent the core findings of the present study. After controlling for the same variables that were controlled in the linear models, we find not only that the coefficient of the linear term remains positive and statistically significant but also that the coefficient of the squared term is statistically significant. The significance level for the squared terms is 1%. This finding reveals an inverted U-shaped curve between administrative costs and government contracts. That said, a nonprofit’s government contracts will not increase consistently as its administrative costs grow. Rather, when administrative costs reach a certain point, government contracts will drop even if administrative costs continue to increase. Based on the fixed-effects estimation in Column 2, we calculate that the turning point occurs when a nonprofit’s administrative costs reach 16.07% (i.e., $- 0.0450 / 2 (0.0014)$ ).

We conduct two additional tests to examine the robustness of our findings. First, although we use the 1-year lagged model to explore the relationship, it could be argued that nonprofit receipt of government contracts might be affected by nonprofits’ previous financial activities and performance that span more than 1 year. For example, when preparing funding proposals and calculating administrative costs, nonprofits might not only use cost information from the prior year but also consider cost information from several previous years. Moreover, to evaluate the performance of nonprofits, public managers might examine the long-term trend, which is more reliable rather than data from a single year. To further explore this possibility and test the robustness of our findings, we use 3-year moving average values (i.e., the average of values of current year t and two previous years, t – 1 and t – 2) to measure our explanatory variables. We believe that 3 years can represent a long-term period that captures an organization’s financial fluctuations over time; thus, a 3-year average can help iron out the surge and slump of an organization’s financial condition. The results of the 3-year average model in Table 4 indicate the coefficients of the linear term are statistically significant (p < .01) and positive, and the coefficients of the squared term are statistically significant (p < .01) and negative. In other words, nonprofits’ receipt of government contracts will first increase as they expand administrative costs to meet funding requirements, but will decrease when their administrative costs reach a certain threshold. The Hausman test reports a p value of .0000 with a chi-square value more than 76.2, thereby suggesting that fixed-effects estimation would be more reliable. Using the fixed-effects estimation, we identify that the turning point occurs when administrative costs reach 15.61% (i.e., $- 0.0999 / 2 (- 0.0032)$ ).

Table 4.

Robustness Check: 3-Year Average Model.

Variables	Fixed-effects		Random-effects
Variables	(1)	(2)	(3)	(4)
Administrative costs	0.0471*** (0.0112)	0.0999*** (0.0187)	0.0607*** (0.0096)	0.0998*** (0.0157)
Administrative costs²		−0.0032*** (0.0004)		−0.0032*** (0.0004)
Program expenses	0.0402*** (0.0061)	0.0483*** (0.0067)	0.0425*** (0.0055)	0.0452*** (0.0055)
Fundraising expenses	−0.0104(0.0065)	−0.0121(0.0065)	−0.0221*** (0.0060)	−0.0218*** (0.0060)
Organization size	0.5972*** (0.0261)	0.6821*** (0.0305)	0.7195*** (0.0201)	0.7274*** (0.0201)
Constant	−4.1064*** (0.8670)	−3.5615*** (0.8042)	−5.6652*** (0.7607)	−6.1751*** (0.7672)
R ²	.2036	.2121	.2102	.2230
Number of organizations	1,036	1,036	1,036	1,036
Observations	7,360	7,360	7,360	7,360

Note. Dependent variable is total amount of government contracts in log form at year t. All the right-side variables take the average value of the observations at year t, year t – 1, and year t – 2. Organziation fixed-effects and year indicator variables are included in each regression. Standard errors in parentheses.

p < .1. **p < .05. ***p < .01.

Second, we understand that the previous models might raise the concern for endogeneity with respect to the regressors, which could potentially bias the estimation results.⁴ To address this concern, we also employ the general method of moments (GMM) estimation to check the robustness of our findings. Econometrics literature suggests that GMM can be a good specification method to address potential endogeneity of regressors (Arellano & Bond, 1991; Roodman, 2009). In our full specification, the p value for Sagan/Hansen J test is 18%, indicating that the null hypothesis of instruments validity cannot be rejected. The p value for second-order serial correlation test, Arrellano-Bond test, is 20%, showing that the null hypothesis of no autocorrelation cannot be rejected as well. In sum, the GMM estimation seems to be a valid method for our data. The results from the GMM estimation are reported in Table 5. In Column 2, the coefficient of the linear term is statistically significant (p < .05) and positive, and the coefficient of the squared term is statistically significant (p < .01) and negative. In other words, the GMM estimation also confirm the inverted U-shaped association between administrative costs and government contracts, with the tipping point appearing when administrative cost is 17.59% (i.e., $- 0.0739 / 2 (- 0.0021)$ ).

Table 5.

Robustness Check: General Method of Moments Model.

Variables	(1)	(2)
Government contracts (lagged)	0.6854*** (0.0812)	0.6948*** (0.0838)
Administrative costs	0.0705*** (0.0224)	0.0739** (0.0307)
Administrative costs²		−0.0021*** (0.0007)
Program expenses	−0.0015(0.0183)	0.0164(0.0179)
Fundraising expenses	−0.0119(0.0079)	−0.0008(0.007)
Organization size	0.3234*** (0.1052)	0.04355*** (0.1010)
Number of organizations	692	692
Observations	4,880	4,880

Note. Dependent variable is total amount of government contracts in log form at year t. All the right-side variables are the same as the fixed- and random-effects models in Table 3 except the inclusion of lagged dependent variable. The general method of moments (GMM) estimation is by first-difference specification with instrument of lagged government contracts. The p values for Sagan/Hansen J test in Specifications 1 and 2 are 22% and 18%, respectively, indicating the estimation cannot reject the null hypothesis of instruments validity. The p values for second-order serial correlation test, Arrellano-Bond test, in Specifications 1 and 2 are 26% and 20%, respectively, showing that the null hypothesis of no autocorrelation cannot be rejected. Therefore, no serious problems with specifications and estimations were found based on these tests. We have also done sensitive analysis using 2-year lagged values and the patterns are generally the same. Standard errors in parentheses.

p < .1. **p < .05. ***p < .01.

After combining all the evidence, we observe that different model specifications jointly demonstrate that the relationship between a nonprofit’s level of administrative costs and its amount of government contracts is curvilinear, following an inverted U-shaped pattern. The tipping point of the inverted U-shaped curve appears when administrative costs reach approximately 16% to 18% of total expenses.

Discussion and Conclusion

Over the past several decades, nonprofits have become increasingly involved in the delivery of government funded services through various financial mechanisms. This new governance model has attracted substantial scholarly attention via different lenses over the years. However, within this growing body of literature, the effect of nonprofit administrative costs on government’s source selection remains largely unclear. The previous literature has proposed different arguments but without strong and consistent empirical support. Our study thus extends the extant literature with a new theoretical framework and empirical evidence. In particular, we argue that government contracting decisions involve a balance of efficiency and quality, which means that neither the price nor the quality consideration could solely dominate source selection. From a more balanced standpoint, we propose an inverted U-shaped relationship between a nonprofit’s level of administrative costs and its amount of government contracts. Our data lend strong support for this hypothesis: as a nonprofit’s administrative costs increase, its government contracts will initially increase (due to the quality concern), but after a tipping point where its administrative costs approximate 16% to 18% of total expenses, further increases in the nonprofit’s administrative costs reduce its government contracts (due to the price concern). In sum, our study adds new knowledge on government–nonprofit funding relationship.

This study has practical implications for both public and nonprofit management. On the public management side, the heavy government dependence on nonprofit contractors for service delivery and policy implementation highlights the importance of contracting management to ensure high-quality results (Kelman, 2002; Kettl, 1993). When acquiring services from nonprofits, to maximize the proportion of funding devoted to programs, government presumably sets high (and sometimes unrealistic) expectations to push nonprofits to keep administrative costs down. A widely used practice is to set caps to severely limit the percentages of funding that can be used for administrative purposes (Berlin et al., 2017; GAO, 2010). This emphasis on efficiency in government procurements leads nonprofits to overlook the improvement of their infrastructure and operations, which further compromises their organizational effectiveness and service quality. Our study implies that public managers may need to reevaluate whether these caps are set artificially below the legitimate administrative costs that are necessary for nonprofits to provide quality services and function as capable contractors. More importantly, in addition to cautioning public managers regarding caps on administrative costs, we join Hager et al. (2004) to suggest that floors (i.e., minimum administrative cost rates) should also be used as signals to inform nonprofits of the necessary administrative costs to act as competent service providers. We find that federal agencies on average are most likely to fund nonprofits with an administrative cost ratio of approximately 16% to 18%. Clearly, we do not suggest that one defined optimal rate exists because funding agencies face different policy environments, but we encourage public managers to reconsider their policies and practices.

On the nonprofit management side, with government representing one major funder of the nonprofit sector, securing government funding and managing the funding relationship with government have become critical challenges for nonprofit managers. Conventional wisdom suggests that government prefers low administrative costs. As a result, nonprofits typically engage in a “race to the bottom,” competing to drive down their administrative costs and reporting lower administrative costs in funding proposals (Gregory & Howard, 2009). Our finding informs nonprofit management in two ways. First, government does not always pursue the lowest administrative costs in their contracting decisions; government favors lower administrative costs only when nonprofits have the necessary capacity to fulfill contractual obligations. Thus, nonprofits spending too little on infrastructure and operations are less likely to receive government contracts. Our finding indicates that nonprofits with an administrative cost ratio of approximately 16% to 18% are most preferred by funding agencies. In other words, nonprofits that have made a reasonable investment in staffing and infrastructure to carry out their missions are better positioned in the competition for government support. Nonprofits may also use their reasonable administrative costs in funding proposals to signal their potential to function as competent contractors.

Second, while working with government, nonprofits need to communicate frequently with their funders on administrative cost matters. Previous literature highlights that government–nonprofit contracting often follows a negotiation approach where relational exchange plays an important role in facilitating contract enforcement (e.g., Amirkhanyan et al., 2010; Sclar, 2000). Nonprofit managers thus should make government officials aware that any contractual exchange is a balance of price and quality and that their working relationships are mutually beneficial rather than just transactional. Meanwhile, funding agencies need to pay for a fair share of the administrative costs required to effectively run government programs and sustain nonprofit capacity to act as high-performing partners. Through these communications, nonprofits play an educational role, urging public managers to reconsider whether their existing policies and practices will motivate nonprofits to deliver quality services on behalf of government (Lu, 2018). Berlin et al. (2017) reported that public managers are often unaware of the overhead issues in government contracting process unless the issues are raised by nonprofits.

It is worth noting that our study is subject to several limitations and caveats. First, we cannot completely rule out the possibility of misreporting in the VolAg data, even though the self-reported financial data that nonprofits provided in the USAID registration had to be consistent with audited financial statements, which are believed to be more reliable. Given the documented evidence that nonprofits may misreport efficiency ratios to enhance perceived legitimacy (e.g., Krishnan et al., 2006; Parsons et al., 2017), the interpretation of our results should be undertaken with caution. Second, the VolAg data provide rich information about the revenue and expenditure breakdowns of registered nonprofits’ activities, but they do not provide information on the organizational characteristics and local conditions of nonprofits operating overseas. In this way, the study might suffer from omitted variable bias. Third, our analysis builds upon the organizational level rather than the contract level. Therefore, we can only explore the association between nonprofits’ administrative costs and their amount of government contracts to identify the “optimal” administrative cost ratio on average. However, we do recognize that administrative costs may vary by contract. Fourth, our findings are based on a data set of international development nonprofits, and their generalization to nonprofits in other policy fields cannot be guaranteed because different policy fields involve different policy and resource environments. We welcome future studies employing more nuanced data on other nonprofit settings to replicate our findings.

In sum, this study deepens our understanding of the funding relationship between government and nonprofits. It adds new knowledge to the literature by demonstrating a nonlinear relationship between nonprofits’ administrative costs and their receipt of government contracts and by offering managerial implications for public and nonprofit management in addressing the “overhead myth” in government contracting process.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: Jianzhi Zhao received funding support from the National Social Science Fund of China Grant 19ZDA072.

ORCID iD

Jiahuan Lu

Notes

Author Biographies

Jianzhi Zhao is an associate professor at the School of International Relations and Public Affairs of Fudan University. His research interests lie in international development, INGOs, and development economics. He was Shorenstein Postdoctoral Fellow at Stanford University and worked at multiple international organizations, including the World Bank, ADB and IADB. He received his PhD from the University of Maryland, College Park.

Jiahuan Lu is an assistant professor in the School of Public Affairs and Administration at Rutgers University-Newark. His research interests lie in the area of public and nonprofit management, with an emphasis on government contracting and nonprofit-government relations.

References

Amirkhanyan

A. A.

Kim

H. J.

Lambright

K. T.

(2010). Do relationships matter? Assessing the association between relationship design and contractor performance. Public Performance & Management Review, 34(2), 189–220.

Amirkhanyan

A. A.

Lambright

K. T.

(2017). Citizen participation in the age of contracting: When service delivery trumps democracy. Routledge.

Andreoni

Payne

(2003). Do government grants to private charities crowd out giving or fund-raising? The American Economic Review, 93(3), 792–812.

Arellano

Bond

(1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. The Review of Economic Studies, 58(2), 277–297.

Ashley

S. R.

Faulk

(2010). Nonprofit competition in the grants marketplace. Nonprofit Management and Leadership, 21(1), 43–57.

Ashley

S. R.

Van Slyke

D. M.

(2012). The influence of administrative cost ratios on state government grant allocations to nonprofits. Public Administration Review, 72(Suppl. 1), 47–56.

Bel

Fageda

Warner

M. E.

(2010). Is private production of public services cheaper than public production? A meta-regression analysis of solid waste and water services. Journal of Policy Analysis and Management, 29(3), 553–577.

Berlin

Masaoka

Schumann

M. J.

(2017). Two-legged stool: New findings from California on nonprofits and overhead. Nonprofit Policy Forum, 8(2), 165–181.

Brown

T. L.

Potoski

Van Slyke

D. M.

(2018). Complex contracting: Management challenges and solutions. Public Administration Review, 78(5), 739–747.

10.

Bowman

(2006). Should donors care about overhead costs? Do they care? Nonprofit and Voluntary Sector Quarterly, 35(2), 288-310.

11.

Charles

(2018). Nonprofit arts organizations: Debt ratio does not influence donations—Interest expense ratio does. The American Review of Public Administration, 48(7), 659–667.

12.

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.

13.

Cohen

Eimicke

(2008). The responsible contract manager: Protecting the public interest in an outsourced world. Georgetown University Press.

14.

Ecer

Magro

Sarpça

(2017). The relationship between nonprofits’ revenue composition and their economic-financial efficiency. Nonprofit and Voluntary Sector Quarterly, 46(1), 141–155.

15.

Eger

R. J.

McDonald

B. D.

(2017). Cost accounting for government grants. In Mohr

(Ed.), Cost accounting in government: Theory and applications (pp. 81–99). Routledge.

16.

Garrow

E. E.

(2010). Receipt of government revenue among nonprofit human service organizations. Journal of Public Administration Research and Theory, 21(3), 445–471.

17.

Gazley

Brudney

J. L.

(2007). The purpose (and perils) of government-nonprofit partnership. Nonprofit and Voluntary Sector Quarterly, 36(3), 389-415.

18.

Gneezy

Keenan

E. A.

Gneezy

(2014). Avoiding overhead aversion in charity. Science, 346(6209), 632–635.

19.

Greenlee

J. S.

Trussel

J. M.

(2000). Predicting the financial vulnerability of charitable organizations. Nonprofit Management and Leadership, 11(2), 199–210.

20.

Gregory

A. G.

Howard

(2009). The nonprofit starvation cycle. Stanford Social Innovation Review, 7(4), 49–53.

21.

Hager

M. A.

Pollak

Wing

(2004). Getting what we pay for: Low overhead limits nonprofit effectiveness. Urban Institute.

22.

Kelman

S. J.

(2002). Contracting. In Salamon

L. M.

(Ed.), The tools of government: A guide to the new governance (pp. 282–318). Oxford University Press.

23.

Kettl

D. F.

(1993). Sharing power: Public governance and private markets. Washington, DC: Brookings Institution Press.

24.

Kettl

D. F.

(2002). Managing indirect government. In Salamon

L. M.

(Ed.), The tools of government: A guide to the new governance (pp. 490–510). Oxford University Press.

25.

Kettl

D. F.

(2017). Can governments earn our trust? Polity Press.

26.

Krishnan

Yetman

M. H.

Yetman

R. J.

(2006). Expense misreporting in nonprofit organizations. The Accounting Review, 81(2), 399–420.

27.

Lamothe

(2015). How competitive is “competitive” procurement in the social services? The American Review of Public Administration, 45(5), 584–606.

28.

Lecy

J. D.

Searing

E. A.

(2015). Anatomy of the nonprofit starvation cycle: An analysis of falling overhead ratios in the nonprofit sector. Nonprofit and Voluntary Sector Quarterly, 44(3), 539–563.

29.

Lindenberg

(1999). Declining state capacity, voluntarism, and the globalization of the not-for-profit sector. Nonprofit and Voluntary Sector Quarterly, 28(Suppl. 1), 147–167.

30.

(2015). Which nonprofit gets more government funding? Nonprofits’ organizational attributes and their receipts of government funding. Nonprofit Management and Leadership, 25(3), 297–312.

31.

(2018). Fear the government? A meta-analysis of the impact of government funding on nonprofit advocacy engagement. The American Review of Public Administration, 48(3), 203–218.

32.

Zhao

(2019). How does government funding affect nonprofits’ program spending? Evidence from international development organizations. Public Administration and Development, 39(2), 69–77.

33.

Marwell

N. P.

Calabrese

(2014). A deficit model of collaborative governance: Government–nonprofit fiscal relations in the provision of child welfare services. Journal of Public Administration Research and Theory, 25(4), 1031–1058.

34.

Marwell

N. P.

Gullickson

(2013). Inequality in the spatial allocation of social services: Government contracts to nonprofit organizations in New York City. Social Service Review, 87(2), 319–353.

35.

McCleary

R. M.

Barro

R. J.

(2008). Private voluntary organizations engaged in international assistance, 1939–2004. Nonprofit and Voluntary Sector Quarterly, 37(3), 512–536.

36.

Milward

H. B.

Provan

K. G.

(2000). Governing the hollow state. Journal of Public Administration Research and Theory, 10(2), 359–380.

37.

Nikolova

(2015). Government funding of private voluntary organizations: Is there a crowding-out effect? Nonprofit and Voluntary Sector Quarterly, 44(3), 487–509.

38.

Parsons

L. M.

Pryor

Roberts

A. A.

(2017). Pressure to manage ratios and willingness to do so: Evidence from nonprofit managers. Nonprofit and Voluntary Sector Quarterly, 46(4), 705–724.

39.

Roodman

(2009). How to do xtabond2: An introduction to difference and system GMM in Stata. Stata Journal, 9(1), 86–136.

40.

Saidel

J. R.

(1991). Resource interdependence: The relationship between state agencies and nonprofit organizations. Public Administration Review, 51(6), 543–553.

41.

Salamon

L. M.

(1995). Partners in public service: Government–nonprofit relations in the modern welfare state. Johns Hopkins University Press.

42.

Salamon

L. M.

Toepler

(2015). Government–nonprofit cooperation: Anomaly or necessity? Voluntas: International Journal of Voluntary and Nonprofit Organizations, 26(6), 2155–2177.

43.

Savas

E. S.

(2000). Privatization and public–private partnerships. Chatham House.

44.

Sclar

(2000). You don’t always get what you pay for: The economics of privatization. Cornell University Press.

45.

Sloan

M. F.

(2009). The effects of nonprofit accountability ratings on donor behavior. Nonprofit and Voluntary Sector Quarterly, 38(2), 220–236.

46.

Smith

S. R.

Lipsky

(1993). Nonprofits for hire: The welfare state in the age of contracting. Harvard University Press.

47.

Steuerle

C. E.

Abramson

A. J.

Steele

Hodgkinson

(2017). Meeting social needs through charitable and government resources. In Boris

E. T.

Steuerle

C. E.

(Eds.), Nonprofits and government: Collaboration and conflict (pp. 71–102). Brookings Institution Press.

48.

Stoddard

(2012). International assistance. In Salamon

L. M.

(Ed.), The state of nonprofit America (pp. 329–361). Brookings Institution Press.

49.

Stone

M. M.

Hager

M. A.

Griffin

J. J.

(2001). Organizational characteristics and funding environments: A study of a population of United Way-affiliated nonprofits. Public Administration Review, 61(3), 276–289.

50.

Suárez

D. F.

(2010). Collaboration and professionalization: The contours of public sector funding for nonprofit organizations. Journal of Public Administration Research and Theory, 21(2), 307–326.

51.

Suárez

D. F.

Esparza

(2017). Institutional change and management of public–nonprofit partnerships. The American Review of Public Administration, 47(6), 648–660.

52.

Tarnoff

(2015). U.S. Agency for International Development (USAID): Background, operations, and issues. Congressional Research Service.

53.

Tinkelman

Mankaney

(2007). When is administrative efficiency associated with charitable donations? Nonprofit and Voluntary Sector Quarterly, 36(1), 41–64.

54.

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.

55.

U.S. Government Accountability Office. (2002). Foreign assistance: USAID relies heavily on nongovernmental organizations, but better data needed to evaluate approaches (GAO-02-471). Washington, DC: United States General Accounting Office.

56.

U.S. Government Accountability Office. (2010). Nonprofit sector: Treatment and reimbursement of indirect costs vary among grants, and depend significantly on federal, state, and local government practices (GAO-10-477). Washington, DC: United States General Accounting Office.

57.

Van Slyke

D. M

. (2007). Agents or stewards: Using theory to understand the government–nonprofit social service contracting relationship. Journal of Public Administration Research and Theory, 17(2), 157–187.

58.

Warner

M. E.

Hefetz

(2004). Pragmatism over politics: Alternative service delivery in local government, 1992–2002. In International City/County Management Association (Ed.), The municipal yearbook 2004 (pp. 8–16). Washington, DC: International City/County Management Association.

59.

Warner

M. E.

Hefetz

(2009). Cooperative competition: Alternative service delivery, 2002–2007. In International City/County Management Association (Ed.), The municipal yearbook 2009 (pp. 11–20). Washington, DC: International City/County Management Association.

60.

Weisbrod

B. A.

Dominguez

N. D.

(1986). Demand for collective goods in private nonprofit markets: Can fundraising expenditures help overcome free-rider behavior? Journal of Public Economics, 30(1), 83–96.

61.

Wing

Hager

M. A.

(2004, November). Who feels pressures to contain overhead costs? Results from a national survey [Paper presentation]. 2004 Annual Conference of the Association for Research on Nonprofit Organizations and Voluntary Action, Los Angeles, CA, United States.

62.

Wong

Ortmann

(2016). Do donors care about the price of giving? A review of the evidence, with some theory to organise it. Voluntas: International Journal of Voluntary and Nonprofit Organizations, 27(2), 958–978.