Alternative Service Delivery

Abstract

We analyze panel data of U.S. states to determine whether nonprofit contribution and program service revenues are correlated with state tax burden. State tax burden is modeled as a function of (a) state tax policy, (b) nontax policy factors that affect state income, and (c) other exogenous factors that are independent of state tax policy and do not directly induce income; regression results reveal correlations with variables in all three categories. Intergovernmental revenue (IGR) paid to local governments, debt burden, tax exporting, a tax revenue limitation, and nonprofit revenue are most consistently correlated with state tax burden. Financial support for nonprofits in the form of contributions helps to reduce state tax burden and does so at a meaningful level. This finding implies nonprofits provide goods and services that are supplementary to government provision. However, the supplementary nature of nonprofit service provision is not universal. Further analysis of contribution and program service revenues for nonprofits in particular service categories finds either no correlation with state tax burden, a reduction in state tax burden, or an increase in tax burden imposed on state residents over time. By controlling for factors influencing demand for service provision and state tax policy changes, the regression results also provide evidence that government acts as a free rider.

Keywords

nonprofit finance state tax burden

Introduction

The United States has a somewhat unique culture of charitable giving compared with the international community, because many other countries have more extensive tax and redistribution systems. As a result, approximately 1 million nonprofit organizations exist in the United States, providing services to improve social welfare within their communities.¹ A central and generally unanswered question related to nonprofits is: what potential influence nonprofit financing (in the form of contribution and/or program service revenues) may have on citizen tax burden? If nonprofits act as alternate providers of services demanded by citizens, then greater financial support for nonprofits might reduce some of the need for government to provide particular services, thereby alleviating the tax burden imposed on residents. However, if nonprofits exist to complement rather than to supplement government provision or to seek the expansion of public financing of public services, then the financing of nonprofit activities may actually increase resident tax burdens. By examining the interconnectedness of the government and nonprofit sectors, the purpose of this article is to determine whether (and to what extent) nonprofit financing correlates with state tax burden.

The next section develops the empirical framework for analyzing the determinants of state tax burden, which is followed by a discussion of the relevant literature on the interaction of the government and nonprofit sectors; this section includes the theoretical basis for analysis. This is followed by an explanation of how nonprofit financing fits into the model as a determinant of state tax burden. Finally, the model estimation, regression results, and implications of the findings are discussed.

Determinants of Tax Burden

Tax burden is fundamentally described as some function of taxes paid and income (Donnahoe, 1947). For practical purposes, this definition translates into the two most common measures of tax burden—(a) taxes per capita and (b) the ratio of taxes to income or wealth (Donnahoe, 1947). Using these and other measures of tax burden, much of the extant literature focuses on assessing the distribution of and trends in federal, state, and local tax burdens, on studying the congruence between who bears tax burdens and who benefits from services financed through taxes, and on the implications of tax burdens with respect to economic activity. Recent analysis of these issues reveals the similarity of state–local tax burdens among the 50 states partly due to the fact that state and local governments provide comparable services (Prante, 2008).

Tax burden was traditionally conceived as synonymous with tax policy because it was thought to directly and almost exclusively correlate with changes in legislation. More recently, scholars have begun to dissect the concept of tax burden in recognition of its complexity and influence by factors other than legislative changes in tax policy. Reed & Rogers (2006) stratify the factors influencing state tax burden into three mutually exclusive categories: (a) state tax policy, (b) nontax policy factors that affect state income, and (c) other exogenous factors that are independent of state tax policy and do not directly induce income. In the first category, changes in tax legislation, tax exporting, and tax competition are considered important determinants of tax burden. Over the past several years, states have made conscious efforts and often garnered public support for increasing tax burdens on nonresidents (Prante, 2008). A state’s ability to export taxes has a direct influence on the state’s ability to meet expenditure needs and consequently may reduce taxes imposed on state residents (Lile & Soule, 1969). In addition, there is evidence to suggest that legislative tax decisions are affected by the tax burdens of neighboring jurisdictions (Case, Rosen, & Hines, 1993; Figlio, Kolpin, & Reid, 1999; Ladd, 1992; among others). In the second category, intergovernmental aid affects the income available to governments to provide goods and services and is therefore an important determinant of taxes paid per capita (Bell & Bowman, 1987). With respect to the third category, state tax burdens are linked to expenditure trends with commitments to frugality and efficiency often resulting in lower tax burdens, and vice versa (Prante, 2008), suggesting demand factors related to government service provision influence tax burdens as well. Political ideology is also influential (Allers, De Haan, & Sterks, 2001).

This conceptualization provides a useful framework for empirically assessing the determinants of state tax burden and is supported by these and other findings from the extant literature. As such, we closely follow the approach of Reed & Rogers (2006) in specifying our econometric model according to these three categories.

Government and Nonprofit Sectors

As noted by Young (2006), government and nonprofit relationships are complex. Specifically, nonprofits may provide goods and services not supplied by government or may partner with government to provide certain public goods (perhaps even financed by government). This view of the nonprofit sector as supplementary or complementary with respect to the government sector suggests that different relationships can exist simultaneously and in different contexts. For our purposes, it also suggests that the provision of goods by government may vary because of this relationship with the nonprofit sector. And varying government service provision (i.e., the basket of public goods financed through taxation rather than voluntarily) leads to different tax burdens incurred by citizens. By extension, then, contributions to and other earned revenues (such as government contracts) by nonprofits may alter tax burden for citizens because of this private financing of goods and services.

The framework we believe most directly informs this notion that financing of nonprofits may ultimately influence citizen tax burden is rooted in theories of welfare economics (based on the “three failures” notion), in which government intervention in an economy is warranted to overcome some market failure. Because of these market failures, governments intervene to provide collective goods, reduce the exclusion of certain populations from public goods (through regulation of private providers or subsidies to certain populations, as two examples), and enforce and regulate contracts to reduce potential fraud in the presence of information asymmetry.

Market Failure

The nonprofit sector fulfills a significant role in society by correcting the “failure” of private enterprise to provide adequate public goods (Hansmann, 1980). Nonprofit organizations offer the kinds of services typically identified with governments, which include assisting the disadvantaged, providing social services, preserving the environment, and funding medical research (Weisbrod, 1997). Schiff & Weisbrod (1991), James (1998), and Weisbrod (1998) all find evidence supporting the notion that nonprofits cross-subsidize their activities to reduce the overexclusion of public goods and services (effectively subsidizing these activities with profits from other activities which may not be mission-critical). In this understanding of government–nonprofit relationships, nonprofits provide particular services in lieu of government provision, suggesting an inverse relationship between the two sectors.

A particular type of market failure is contract failure, in which individuals purchase complex services with difficult to evaluate output or quality. Consumers may feel that the nonprofit “nondistribution constraint”—in which profits may not be distributed for personal benefit—mitigates the profit maximization motivation of a private owner which may lead to reduced quality of the output; in doing so, the consumer has greater confidence that nonprofit organizations compared with for-profit entities. Hence, donors and clients are more likely to trust nonprofits to deliver the goods and services demanded of these difficult to evaluate outputs at the quality expected (Hansmann, 1980). For those seeking to explain nonprofit activities and existence, contract failure remains an important theory (see, for example, Bowman, 2004; Hirth, 1999; Steinberg & Gray, 1993). However, Handy et al. (2010) find evidence that although individuals are more likely to purchase complex services from nonprofits, they are unable to readily differentiate nonprofit from for-profit organizations. Similarly, Van Slyke and Roch (2004) find that citizens frequently misidentify nonprofit providers as government agencies when their own satisfaction is lower, suggesting citizens do not always accurately differentiate nonprofits.

Government Failure

“Government failure” refers to the unmet demand for public goods of those citizens above the median voter (Weisbrod, 1977). This occurs because governments make final decisions about service provision and tax burden based on the preferences of the median voter or dominant political coalition due to the democratic nature of society (Buchanan & Tullock, 1962). As government expenditures increase, individuals’ taxes also increase and their incomes decline. So, as individuals consume more public goods, they consume less private goods because they are constrained by their income level and tax price.

Citizens with preferences for public good provision different from the median voter might be willing to finance nonprofit organizations through donations. In fact, the vast majority of contributions to nonprofits derive from households, not large institutions like foundations or corporations.² Nonprofits have unique abilities to supply the demand from citizens in voting minorities because private contributions are often tax-deductible, thereby lowering the effective price of giving. Therefore, government failure theory suggests an inverse relationship between the government and nonprofit sectors, because the population demands less of government as nonprofits take on more responsibility for the provision of goods.

Despite the importance of the government failure theory in explaining the role of nonprofit activities in certain segments of the economy, empirical results testing this theory have been mixed. For example, Matsunaga & Yamauchi (2004) find support for the government failure theory, whereas Gronbjerg & Paarlberg (2001) do not. Furthermore, Paarlberg & Gen (2009) find that although population heterogeneity may increase demand for nonprofit service provision, population homogeneity may be necessary to sustain and finance such activities over time.

Voluntary Failure

Nonprofits themselves may not adequately fulfill market or government failures, suggesting “voluntary failure” (Salamon, 1987). Most basically, nonprofits are frequently undercapitalized, as they often lack the resources necessary to adequately address social problems. Furthermore, nonprofits are generally unable to sustain these requisite resources over time because they are unable to tax and have do not have access to equity markets.

Anheier (2005) notes that these nonprofit weaknesses are actually governments’ strengths, suggesting a more complementary relationship between them. Governments can provide significant resources to nonprofits over time, give legitimacy to a nonprofit, and include nonprofits in the political process; in return, governments are able to purchase extremely specialized services through nonprofits, respond to new demands more quickly using nonprofit service providers, perhaps reduce programmatic costs by avoiding civil service requirements, and might also reduce costs by requiring nonprofits to augment these public funds with additional donations (for example, through matching requirements) or through other in-kind donations (e.g., administrative overhead).³

In addition, government may even be able to shift responsibility for managing goods and services to nonprofits. For example, although New York City’s Central Park is a public good, it is maintained and financed largely through a nonprofit group (the Central Park Conservancy) with access to additional resources (contributions) that essentially manages all aspects of the park’s operations. Finally, the availability of government grants and contracts might result in the creation of new nonprofits at the behest of government agents; in doing so, the benefits of operating through a nonprofit (lower costs, greater flexibility, etc.) combined with the security and predictability of public resources makes this an attractive option for providing certain goods and services. These publicly created nonprofits may work to influence public policy, expand programs, and grow the responsibilities of the nonprofit sector with government financial support.

An important literature has developed documenting this government–nonprofit complementary relationship (DeHoog & Salamon, 2002; Gronbjerg, 1993; Saidel, 1991; Smith & Lipsky, 1993, as examples). Young (2006) describes the blurring of lines between the government and nonprofit sectors, noting that many nonprofit community development associations have board members appointed by government officials, and public universities and libraries frequently incorporate private nonprofit fund-raising or foundation organizations within their operating structures.

Nonprofit Financing and Tax Burden

When demand for goods and services is homogeneous and to a great enough extent to constitute a majority, citizen preferences and service demands will likely be met by government provision financed through taxation (Ferris, 1998), suggesting little need for voluntary provision. As a rational response, citizens should reduce their own charitable contributions to nonprofit organizations (Bergstrom, Blume, & Varian, 1986; Roberts, 1984; Warr, 1982). In this regard, government provision has a crowding out effect on private charitable giving (Abrams & Schmitz, 1984; Kingma, 1989; Roberts, 1984), suggesting that individuals behave as free riders (Becker & Lindsay, 1994). However, this relationship at the organizational level is not necessarily as straightforward as suggested. Brooks (2000) empirically demonstrates that low levels of government funds may actually crowd in additional contributions, whereas high levels of government funds crowd out. Andreoni & Payne (2003) find that government crowd out of contributions is due in part to reduced fund-raising activities by nonprofits. In addition, Brooks (2003) finds that the average donation to nonprofits declines following the receipt of government subsidies, but the number of donors actually increases.⁴

However, diversity among the population often leaves people dissatisfied with government service provision, which fuels the need for nonprofits to serve as an alternative mechanism for providing services (Weisbrod, 1997). It is this group of citizens who are often willing to provide supplementary levels of goods and services by “mobilizing resources on a voluntary collective basis through the nonprofit sector” (Weisbrod, 1977, p. 151). When these supplementary contributions are given, there is potential for the government to become the free rider and reduce its level of service provision because the median voter (and all other voters) will consider this nonprofit contribution as a bonus (Becker & Lindsay, 1994). As a result, the median voter (and others) will decrease their demand for government provision of the good or service. Similar to an individual acting as a free rider, the government will respond to the alternative service delivery and decrease demand by reducing its level of service provision, which reduces tax burdens. In fact, Becker & Lindsay (1994) empirically find that an increase in private income to public colleges and universities leads to a decrease in state and local government support.

The relationship suggested by voluntary failure (in which nonprofits become a vehicle for providing goods and services financed by the government) suggests a different relationship with respect to tax burden. As government contracts with nonprofits to provide services, savings are either unlikely to develop or unlikely to be passed onto taxpayers in the form of reduced tax burden. Instead, nonprofits become a coalition that seeks expanded government-financed services and that seeks to protect its share of the government’s budget. This suggests a complementary relationship that will lead to increased tax burdens as nonprofit responsibilities and numbers grow.

Model Estimation

In light of this discussion and following Reed & Rogers’ (2006) approach described earlier, the determinants of state tax burden (including the influence of nonprofit financing) are assessed using the econometric model specified in Equation 1.

T B_{i t} = α + T P_{i t} β + I_{i t} β + E_{i t} β + ε_{i t}

In equation 1, tax burden (TB) for state i in year t is estimated as a function of vectors of variables representing state tax policy (TP), nontax policy factors that affect state income (I), and other exogenous factors that neither directly relate to tax policy nor induce state income (E). Tax burden is measured as total state tax revenue divided by total state personal income and is reported in percentage terms.⁵ Complete descriptions of all variables and data sources are provided in Table 1.

Table 1.

Variables and Data Sources.

Variable	Description and data source
Dependent variable
Tax burden	Total state tax revenue divided by total state personal income; Sources: U.S. Census Bureau and U.S. Department of Commerce.
Tax policy factors
Tax policy change	A dichotomous variable coded 1 if the state estimated a tax revenue change resulting from an enacted legislative change to a tax revenue source, and 0 otherwise; Source: National Association of State Budget Officers, The Fiscal Survey of States (various years).
Tax competition	Weighted average (weighted by state population) of tax burdens in states with shared borders; Source: Authors’ calculations.
Tax exporting	Amount of federal spending received per dollar of tax paid; Source: Tax Foundation.
Tax revenue limitation	Dichotomous variable coded 1 if the state has a constitutional or statutory limitation restricting new or increased taxes and/or fees, and 0 otherwise; Source: Mullins & Wallin (2004).
Nontax policy income factors
Federal IGR paid	Per capita amount of IGR received from the federal government; Source: U.S. Census Bureau.
Local IGR paid	Per capita intergovernmental expenditures paid to local governments; Source: U.S. Census Bureau.
Debt burden	Total long-term debt outstanding divided by population; Source: U.S. Census Bureau.
Other exogenous factors
Government ideology	State government ideology; scores indexed on a 100-point scale with increasing values indicating greater liberalism; calculated with a methodological formulation of ideology scores and power share measures for the state governor and major party delegations in each house of the state legislature; final scores calculated as the absolute value of the difference between the state’s government ideology score and the mean government ideology scores for all states; Source: Berry, Ringquist, Fording, & Hanson (1998) revised 1960-2006 governments ideology series.
Poverty rate	Percentage of state residents in poverty; Source: U.S. Census Bureau.
Vehicle miles traveled	Number of vehicle miles traveled on state agency-owned public roads per licensed driver; Source: U.S. Department of Transportation.
University enrollment	Total state enrollment in public degree-granting institutions divided by total U.S. enrollment; Source: National Center for Education Statistics.
Nonprofit financing
Contribution revenue	Per capita dollar amount of revenue from contributions, gifts, and grants (including direct and indirect public support and government contributions and grants) for all 501(c)3 nonprofit organizations within the state filing IRS form 990; measured overall and for seven categories differentiating types of nonprofit organizations; Source: National Center for Charitable Statistics.
Program service revenue	Per capita dollar amount of program service revenue (including government fees and contracts) for all 501(c)3 nonprofit organizations within the state filing IRS form 990; measured overall and for seven categories differentiating types of nonprofit organizations; Source: National Center for Charitable Statistics.

Note: IGR = intergovernmental revenue; IRS = Internal Revenue Service.

Four variables account for the influence of tax policy. First, most states have statutory requirements for estimating the budgetary impacts of proposed and enacted changes in tax and spending legislation. Based on this information, the first variable (tax policy change) is a dichotomous variable that measures whether a state estimated a tax revenue change resulting from newly enacted tax legislation.⁶ We hypothesize tax policy change and tax burden to have a direct correlation. Second, tax competition is measured as the weighted (by population) average of tax burdens in the states that share a border with the state of interest.⁷ Based on prior findings (such as Papke 1987; Wasylenko, 1980), we hypothesize that higher tax burdens among neighboring states will induce a state to reduce its tax burden. Third, a state’s ability to export its tax burden is measured as the amount of federal spending received per dollar of tax paid; dollar values greater than $1.00⁸ represent greater ability to export taxes as a state receives a more than 100% return from the federal government for taxes paid by its citizens. We hypothesize an increase in tax exporting will be associated with a reduction in tax burden. Finally, we include a dichotomous variable coded 1 if the state has a constitutional or statutory limitation restricting new or increased taxes and/or fees, and 0 otherwise, because such limitations inherently restrict the amount of tax burden that states are able to impose on residents. Therefore, the presence of a tax revenue limitation should be associated with a lower state tax burden.

Three variables account for the influence of factors that do not directly represent state tax policy but affect the income available to states for providing goods and services.⁹ The per capita amount of intergovernmental revenue (IGR) received from the federal government and per capita intergovernmental expenditures paid to local governments both capture the influence of intergovernmental fiscal relationships. A greater amount of federal aid increases a state’s income and perhaps alleviates the need to impose taxes on its residents, suggesting a negative correlation between federal IGR and state tax burden. Conversely, a greater amount of state aid paid to local governments reduces a state’s income and perhaps induces greater tax burden on state residents to meet expenditure needs, which implies a positive relationship with state tax burden. Finally, debt burden is measured as the per capita amount of total long-term debt outstanding. Although greater use of debt might imply greater resources available to a state, a higher level of debt outstanding is more indicative of a greater amount of mandatory obligations for debt service payments, which ultimately diminishes state income. The implication is that states might need to impose higher taxes on its residents to meet the requirements for debt repayment and also maintain service levels, suggesting a positive correlation between debt burden and state tax burden.

Four variables capture the influence of exogenous factors that neither relate directly to state tax policy nor affect state income levels, namely, politics, demand for goods and services, and nonprofit financing. A measure developed and later revised by Berry, Ringquist, Fording, & Hanson (1998) measures the ideology of a state’s government representatives. Government ideology is measured on a 100-point scale with increasing values indicating greater liberalism. Based on the findings of Alt & Lowry (1994) and Reed (2006), we hypothesize that increased liberalism among government representatives is associated with higher levels of taxes imposed on state residents.¹⁰ The demand for government service provision is captured by three variables, which correspond to the top spending categories for state governments. Increasing values of (a) the state’s poverty rate, (b) vehicle miles traveled on state agency-owned public roads, and (c) state enrollment in public degree-granting institutions relative to U.S. enrollment all, respectively, contribute to greater demand for services related to welfare, highways, and higher education. These demand factors are likely associated with higher levels of taxes imposed on state residents.

The final two variables measure nonprofit financing and are the variables of greatest interest. First, nonprofit contribution revenue is measured as the state per capita dollar amount of revenue from contributions, gifts, and grants for all public charities that filed a Form 990 in a given year within the state.¹¹ This revenue source includes government contributions and grants received by nonprofits within each state. Second, program service revenue is measured in the same manner and includes revenue generated from government contracts. Because contribution and program service revenues account for all government funding received by nonprofits, these two variables measuring nonprofit financing include all revenue sources resulting from “government contracting.”¹²

Although there is some question of whether the address reported by nonprofits on IRS Form 990 is an accurate reflection of nonprofit location and service provision, existing research indicates that the nonprofit sector is overwhelmingly community-based and locally oriented. Wolpert (1993) finds that locally based nonprofits (rather than national nonprofits) give such organizations comparative advantages and niches to channel local donations intended to enhance civic life and community services. This is especially true even for large well-known nonprofits that are actually federations—such as the United Way, the American Cancer Society, the Red Cross, Planned Parenthood, among others; in these organizations, local communities largely raise their own funds from local sources and spend the money locally (around 85% to 90%), suggesting that local preferences are a top priority (Wolpert, 1993). Downs & Greenstein (1996), Bielefeld & Murdoch (1997, 2004), and DeVita, Manjarraz, & Twombly (1999) also find that nonprofits locate at least in part due to community supply and demand. McPherson (1983) finds that nonprofits locate where potential members or clients may be found. Further, Calabrese (2011) finds that only about 1% of nonprofits actively solicit funds outside their own state of location, suggesting that most nonprofits generate the vast majority of their contributions and program service revenue within their own-state’s borders. All of this evidence suggests that nonprofit activity and financing primarily occur within state borders, making an examination of its influence on state tax burden appropriate.

The literature on the role of the nonprofit sector in service provision and charitable giving almost exclusively focuses on the one-way causal connection between government tax and spending policies and financial outcomes in the nonprofit sector. However, even within this literature, there is evidence to suggest reverse causation.¹³ Becker & Lindsay (1994) were the first scholars to develop a theoretical framework and provide empirical support for a reverse causal connection by showing that financial support for the nonprofit sector influences government decisions in the same vein as government service provision influences a citizen free rider.¹⁴ Becker & Lindsay’s (1994) model offers a valid theoretical expectation for the influence of nonprofit financial support on government financing of goods and services.

Data and Methodology

We analyze panel data of all 50 states during the 1991 to 2001 time period. This time span covers the most recent complete trough-to-trough business cycle for which data are available.¹⁵ The unit of analysis is each state (i) each year it is observed (t), which amounts to a balanced panel of 550 observations (i = 50; t = 11). Nonprofit data (as described earlier) were obtained from the Core Files compiled by the National Center for Charitable Statistics (NCCS). Per capita values were used for standardization due to significant variation among states in the numbers of nonprofit organizations and revenues generated for nonprofit service provision. For example, our data set shows 616 nonprofits in Wyoming in 2001, while 24,776 nonprofits filed IRS Form 990 in California in the same year. All other data were obtained from the Census of Governments Survey conducted by the U.S. Census Bureau, the Highway Statistics Series developed by the Federal Highway Administration of the U.S. Department of Transportation, the Bureau of Economic Analysis of the U.S. Department of Commerce, the National Center for Education Statistics of the U.S. Department of Education, and the Fiscal Survey of States compiled by the National Association of State Budget Officers.

Prior to running the regression analysis, several tests were conducted to determine the most appropriate estimation method for the data.¹⁶ In addition, our hypotheses related to tax competition and nonprofit financing imply an expectation that the error structure is possibly correlated across panels (i.e., states), and the estimated coefficients might be biased as a result of endogeneity. As a result, we offer three sets of regression estimates in Table 2. Model 1 provides estimates of fixed effects regression with robust standard errors clustered by state. Model 2 reports coefficients that are also estimated using fixed effects regression but calculated with Driscoll & Kraay (1998) standard errors, which assumes the error structure is heteroskedastic, autocorrelated up to some lag, and possibly correlated between the groups (i.e., panels; Cameron & Trivedi, 2010).¹⁷ Finally, Model 3 provides fixed effects regression results with robust standard errors clustered by state and incorporates 1-year lags of all nondichotomous potentially endogenous independent variables—tax competition, tax exporting, federal IGR revenue, local IGR paid, debt burden, and nonprofit financing—as suggested by Wooldridge (2006).¹⁸

Table 2.

Descriptive Statistics for All Variables.

Variable	Mean	SD	Minimum	Maximum
Dependent variable
Tax burden	7.10%	1.42%	2.74%	15.90%
Tax policy factors
Tax policy change	0.6964	0.4602	0	1
Tax competition	6.74%	0.77%	2.74%	8.53%
Tax exporting	US$1.11	US$0.28	US$0.65	US$2.08
Tax revenue limitation	0.2709	0.4448	0	1
Nontax policy income factors
Federal IGR paid	US$891.54	US$273.36	US$392.12	US$2,008.65
Local IGR paid	US$909.88	US$319.19	US$88.08	US$2,101.70
Debt burden	US$2,140.46	US$1,579.10	US$160.19	US$11,018.60
Other exogenous factors
Government ideology	47.99	24.76	0.00	97.92
Poverty rate	12.58%	3.77%	4.50%	26.40%
Vehicle miles traveled	14,069.59	2,163.54	8,951.42	22,360.63
University enrollment	1.99%	2.40%	0.17%	16.70%
Nonprofit financing
Contribution revenue	US$422.77	US$271.20	US$28.76	US$2,093.68
Program service revenue	US$1,344.21	US$842.93	US$30.79	US$8,023.34
N = 550 (i = 50; t = 11)

Note: IGR = intergovernmental revenue.

Empirical Results

Table 2 provides descriptive statistics for all variables and indicates that the states maintained an average tax burden of 7.10% during the time period analyzed. However, state tax burden varied considerably with values ranging from 2.74% to 15.90%. Although not shown in Table 2, separate calculations revealed that New Hampshire maintained the lowest levels of tax burden with annual values ranging between 2.74% and 4.57%. Conversely, Alaska, Hawaii, and New Mexico exhibited the highest tax burdens with maximum values of 15.91%, 10.68%, and 10.22%, respectively. The highest average tax burden among all the states occurred in fiscal year 2000 when the mean was 7.26%, whereas the lowest average tax burden of 6.86% occurred in fiscal year 1991.

The independent variables of greatest interest in Table 2 are those related to nonprofit financing. According to Table 3, the mean per capita amount of contribution revenue received by nonprofits within the states was US$422.77. However, these values also varied considerably with a standard deviation of US$271.20. Nevada, Idaho, and Louisiana exhibited the lowest levels of contributions (including government contributions and grants) received by nonprofits with average per capita revenues of US$159.13, US$166.63, and US$196.54, respectively. However, the states of Alabama, Massachusetts, Alaska, and New York displayed the highest levels of contributions to nonprofits with mean per capita revenues of US$1,423.20, US$1,134.32, US$958.07, and US$897.81, respectively. Among all the states, the greatest level of nonprofit contributions occurred in fiscal year 2001 when the mean per capita revenue was US$589.58, while the lowest average per capita contribution revenue of US$316.83 occurred in fiscal year 1992. In terms of program service revenue, which includes government fees and contracts, the mean per capita revenue received by nonprofits within states was much higher than contribution revenue at US$1,344.21. Variation in program service revenue was also much greater as state per capita amounts ranged from US$30.79 to US$8,023.34.

Table 3.

Fixed Effects Regression Results.

Variable	Model 1: Robust SE	Model 2: Driscoll and Kraay SEs	Model 3: Lagged IVs with robust SEs
Tax policy factors
Tax policy change	0.0718* (0.0386)	0.0718 (0.0515)	0.0609 (0.0406)
Tax competition	0.2586* (0.1382)	0.2586*** (0.0590)	0.0378 (0.1899)
Tax exporting	−1.6367*** (0.4744)	−1.6367*** (0.3802)	−1.0412** (0.5057)
Tax revenue limitation	−0.3398** (0.1519)	−0.3398*** (0.0627)	−0.3635** (0.1721)
Nontax policy income factors
Federal IGR paid	−0.0004 (0.0004)	−0.0004 (0.0005)	−0.0007 (0.0005)
Local IGR paid	0.0021*** (0.0003)	0.0021*** (0.0003)	0.0019*** (0.0006)
Debt burden	0.0005*** (0.0002)	0.0005*** (0.0001)	0.0004*** (0.0002)
Other exogenous factors
Government ideology	0.0033* (0.0017)	0.0033** (0.0013)	0.0031* (0.0017)
Poverty rate	−0.0057 (0.0153)	−0.0057 (0.0138)	−0.0090 (0.0146)
Vehicle miles traveled	0.0000 (0.0000)	0.0000 (0.0000)	0.0000 (0.0000)
University enrollment	0.0296 (0.1056)	0.0296 (0.0724)	0.1666 (0.1009)
Nonprofit financing
Contribution revenue	−0.0011** (0.0004)	−0.0011*** (0.0002)	−0.0012* (0.0006)
Program service revenue	−0.0001 (0.0001)	−0.0001** (0.0000)	0.0001 (0.0001)
Constant	4.6267*** (1.1541)	4.6267*** (0.8704)	5.6690*** (1.3297)
F	7.21	234.55	4.87
Prob > F	0.0000	0.0000	0.0000
Within R²	0.3780	0.3780	0.3125

Note: IGR = intergovernmental revenue. IV = independent variable.

p < .10. **p < .05. ***p < .01.

Table 2 provides the three sets of fixed effect regression estimates described earlier. Tax exporting, tax revenue limitation, local IGR paid, debt burden, and nonprofit contribution revenue show the most consistent correlations with state tax burden as each variable is statistically significant in all three models at the 90% confidence level or above and significant at the 95% confidence level or above in at least two of the three models. In addition, tax competition, government ideology, and nonprofit program service revenue achieve statistical significance at the 95% confidence level in at least one of the three models, although each variable is insignificant in one or more models. Tax competition, local IGR paid, and debt burden maintain positive correlations with state tax burden, whereas tax exporting, tax revenue limitation, and both nonprofit financing variables are negatively associated with tax burden. All of these coefficient signs conform to our hypotheses.

Straight interpretation of the regression coefficients in Table 3 is somewhat misleading because a one-unit change in several of the independent variables, which produces the coefficient change illustrated in Table 3, is exceptionally small and therefore unlikely to occur in practice. Therefore, the following discussion interprets the correlation between each statistically significant independent variable and state tax burden that would result, on average, from a one standard deviation change in each independent variable, holding all other independent variables constant at their mean. This interpretation of the regression results is more meaningful because a one standard deviation change in each independent variable is more likely to occur than a one-unit change.

Based on this interpretation, a one standard deviation increase in per capita expenditures paid to local governments, which amounts to US$319.19 as illustrated in Table 2, is associated with an increase in tax burden of between 0.64% (Model 3) and 0.67% (Models 1 and 2) within states over time.¹⁹ This amount may seem small, but tax burden is calculated as the amount of state tax revenue divided by personal income; so this percentage increase would be imposed on all taxpaying residents within a state as an average percentage of income. This potential increase in tax burden is not offset by increases in federal IGR received by a state, as federal IGR fails to achieve statistical significance in all three models. Furthermore, according to Table 3, a one standard deviation increase in a state’s debt burden, which amounts to US$1,579.10 of additional future mandatory expenditures for a state, leads to an increase in tax burden of 0.79% over time. Taken together, these findings illustrate the close association that nontax policy factors influencing state income have with the tax burdens imposed on state residents; factors that reduce state income are associated with higher state tax burden.

As expected, the regression results suggest that state tax policy is also an important factor determining tax burdens imposed on state residents. Based on the results in Table 3, the presence of a tax revenue limitation is associated with a decrease in tax burden of between 0.31% (Models 1 and 2) and 0.32% (Model 3) within states over time. Similarly, a one standard deviation increase in the amount of federal spending received per dollar of tax paid, which amounts to US$0.28, is associated with a decrease in tax burden of between 0.30% (Model 3) and 0.45% (Models 1 and 2) within states over time. Finally, a one standard deviation increase in the weighted average tax burdens of states with shared borders (0.77%) is associated with a 0.21 average increase in state tax burden over time. It should be noted, however, that the correlation between these two variables disappears when neighboring tax burden is lagged 1 year. Although the tax policy change variable is largely not significant in the models (with the exception of significance at the 90% confidence level in Model 1), these related findings suggest that some tax policy factors, particularly the ability to shift tax burden to nonresidents and restrictions on new and/or increased tax levies, influence state tax burden in addition to the nontax policy factors discussed above.

Finally, and of greatest interest for this study, are the results pertaining to the nonprofit financing variables. Based on the results in Table 3, a one standard deviation increase in the per capita aggregated amount of revenue from contributions, gifts, and grants received by 501(c)3 nonprofit organizations filing IRS Form 990 within a state, which amounts to US$271.20, is associated with a decrease in the tax burden imposed on state residents of 0.30% over time. In addition, a one standard deviation increase in the per capita aggregated amount of program service revenue received by nonprofits within a state, which equals US$842.93 on average, is associated with a 0.93% decrease in state tax burden over time. It should be noted, however, that this latter finding is statistically significant at the 95% confidence level in Model 2 but is insignificant in Models 1 and 3. We believe this provides some additional evidence of Boyne’s (1998) findings that savings derived from contracting are generally used to maintain or increase other public spending rather than returned to taxpayers in the form of lower taxes. Overall, though, the findings show that financial support for nonprofit service provision does help to reduce state tax burden and does so at a meaningful level compared with the other correlates in Models 1, 2, and 3.

The implications of these findings are twofold. First, by controlling for factors influencing demand for government service provision and state tax policy changes, the regression results provide support for Becker & Lindsay’s (1994) theoretical and empirical assertion that government acts as a free rider by illustrating that increasing values of nonprofit financing are associated with reductions in tax burdens imposed on state residents. Because tax revenue is the primary revenue source for government service provision, this trend is possibly the result of reduced service levels in response to the alternative delivery of services by nonprofit organizations mainly financed privately by households. Second, because nonprofit financing is measured using two variables that capture all funding received from government, the negative correlation between nonprofit financing and state tax burden provides some evidence of the supplementary viewpoint of the role of nonprofits. The noteworthy reduction in state tax burden suggests that a greater amount of per capita revenue received by nonprofits from both private and government sources is indicative of greater financial support for nonprofits to provide services to supplant government provision.

An important question that remains from this analysis is whether the supplementary nature of nonprofit service provision is generalizable across all nonprofit types. Due to significant variation in the types of services provided by nonprofit organizations, it is plausible to expect the association between the government and nonprofit sectors to vary according to nonprofit type. To probe further into the influence of nonprofit financing on state tax burden and whether the nature of this relationship should be characterized as supplementary or complementary, we conducted additional empirical analysis with differentiation of nonprofit types, similar to Tuckman & Chang (1991): (a) arts, culture, and humanities (Code A), (b) environmental and animal (Codes C and D), (c) education, science, and agriculture (Codes B, K, U, and V), (d) health and medical (Codes E, F, G, and H), (e) crime and public safety (Codes I, M, and Q), (f) human services and development (Codes J, L, N, O, P, X, and Y), and (g) community services and development (Codes R, S, T, and W). These seven categories capture all of the National Taxonomy of Exempt Entities (NTEE) nonprofit classifications.

Table 4 illustrates descriptive statistics for the two nonprofit financing variables divided into the seven categories,²⁰ whereas Table 5 provides reestimated regression results. The only change made to the econometric models was the division of the two nonprofit financing variables into 14 different variables to capture both contribution and program service revenue by nonprofit type; reestimation of the regression models was done using the same estimation methods presented above. Only the results pertaining to the nonprofit financing variables are displayed in Table 5.

Table 4.

Descriptive Statistics for Nonprofit Financing by Nonprofit Category.

Nonprofit category	Mean	SE	Minimum	Maximum
Contribution revenue
Arts, culture, and humanities	US$27.16	US$25.43	US$2.14	US$207.22
Environmental and animal	US$11.67	US$16.43	US$0.39	US$120.89
Education, science, and agriculture	US$117.14	US$97.37	US$1.89	US$853.72
Health and medical	US$79.86	US$67.85	US$2.78	US$775.36
Crime and public safety	US$21.70	US$35.51	US$0.08	US$236.08
Human services and development	US$112.25	US$65.06	US$16.56	US$428.95
Community services and development	US$52.98	US$45.45	US$0.42	US$360.11
Program service revenue
Arts, culture, and humanities	US$14.27	US$11.21	US$1.71	US$64.13
Environmental and animal	US$4.73	US$5.63	US$0.05	US$36.86
Education, science, and agriculture	US$211.84	US$202.22	US$3.65	US$1,069.50
Health and medical	US$956.43	US$683.78	US$19.13	US$7,687.53
Crime and public safety	US$4.54	US$5.78	US$0.03	US$36.02
Human services and development	US$140.81	US$161.40	US$1.97	US$1,176.17
Community services and development	US$11.59	US$11.96	US$0.17	US$78.89
N = 550 (i = 50; t = 11)

Table 5.

Fixed Effects Regression Results for Nonprofit Financing.

Variable	Model 1: Robust SEs	Model 2: Driscoll and Kraay SEs	Model 3: Lagged IVs with robust SEs
Contribution revenue
Arts, culture, and humanities	0.0053 (0.0032)	0.0053 (0.0033)	0.0082 (0.0079)
Environmental and animal	0.0107 (0.0072)	0.0107*** (0.0020)	0.0031 (0.0043)
Education, science, and agriculture	−0.0018** (0.0008)	−0.0018*** (0.0004)	−0.0020** (0.0010)
Health and medical	−0.0018 (0.0017)	−0.0018 (0.0011)	−0.0037*** (0.0012)
Crime and public safety	0.0010 (0.0027)	0.0010 (0.0012)	−0.0022 (0.0031)
Human services and development	−0.0041 (0.0027)	−0.0041* (0.0021)	−0.0062 (0.0042)
Community services and development	−0.0006 (0.0014)	−0.0006 (0.0013)	−0.0001 (0.0013)
Program service revenue
Arts, culture, and humanities	0.0143* (0.0072)	0.0143*** (0.0047)	0.0291*** (0.0075)
Environmental and animal	−0.0334 (0.0217)	−0.0334* (0.0178)	−0.0092 (0.0146)
Education, science, and agriculture	0.0007 (0.0005)	0.0007* (0.0004)	0.0014** (0.0005)
Health and medical	0.0001 (0.0001)	0.0001* (0.0000)	0.0001 (0.0001)
Crime and public safety	0.0005 (0.0151)	0.0005 (0.0147)	0.0167 (0.0224)
Human services and development	0.0002 (0.0005)	0.0002 (0.0003)	0.0006 (0.0007)
Community services and development	−0.0061 (0.0063)	−0.0061* (0.0036)	−0.0061 (0.0066)
Constant	4.9785*** (1.1118)	4.9785*** (0.6532)	6.4449*** (1.2514)
F	42.06	283.46	39.1142
Prob > F	0.0000	0.0000	0.0000
Within R²	0.4169	0.4169	0.3725

Note: The following variables were also statistically significant at the 99% confidence level in either Model 1 or Model 2, or both (coefficients reported in parentheses): tax competition (0.2371), tax exporting (–1.6735), tax revenue limitation (–0.3752), local IGR paid (0.0019), debt burden (0.0005), and government ideology (0.0035). Variables that were statistically significant at the 95% confidence level or above in Model 3 include tax exporting (–1.2971), tax revenue limitation (–0.3854), local IGR paid (0.0018), and debt burden (0.0004). IGR = intergovernmental revenue; IV = independent variable.

p < .10. **p < .05. ***p < .01.

Table 5 indicates that the nonprofit service provision category of education, science and agriculture is the only one to achieve statistical significance in all three models and is significant at the 95% confidence level or above in each model. Based on the results, a one standard deviation increase in per capita contribution revenue received by nonprofits with service provision in education, science, and agriculture (US$97.37 as shown in Table 4) is associated with a reduction in tax burdens imposed on state residents of between 0.19% (Model 3) and 0.18% (Models 1 and 2) over time.

The service provision categories of environmental and animal along with health and medical are statistically significant in one of the three models at the 99% confidence level with the latter category achieving significance in Model 3, which incorporates lagged values of the nonprofit financing variables. A one standard deviation increase in per capita contribution revenue going to nonprofits providing health and medical services, which amounts to US$67.85, is associated with a decrease in state tax burden imposed on residents in the next year of 0.25%. Conversely, however, nonprofits in the service provision category of environmental and animal may be associated with increasing state tax burdens. A one standard deviation increase in per capita contribution revenue in this category (US$16.43) is correlated with an increase of 0.18% in state tax burdens over time. A caveat, however, is that these latter two categories of nonprofit service provision only achieve statistical significance in one of the three models, suggesting the correlation with state tax burden might be rather limited.²¹

The bottom half of Table 5 reports the regression results for per capita program service revenue received by nonprofits in each service provision category. In this context, the service provision category of arts, culture, and humanities stands out as the most correlated with state tax burden, as it is statistically significant at the 99% confidence level in two of the three models and significant at the 90% confidence level in the third model. This category is followed by education, science, and agriculture, which is significant at the 95% confidence level in Model 3 and at the 90% confidence level in Model 2. These regression results show that program service revenue in both categories is positively correlated with state tax burden. Specifically, a one standard deviation increase in per capita program service revenue received by nonprofits providing services in arts, culture, and humanities (which amounts to US$25.43) is associated with an increase in state tax burden of 0.36% (Models 1 and 2) and 0.74% (Model 3) over time. Not only is this correlation positive, this association is the largest in magnitude of all the nonprofit financing correlations. In addition, contrary to the findings of a negative correlation between contribution revenue received by nonprofits in the education, science, and agriculture category and state tax burden, program service revenue received by nonprofits in this category is associated with an increase in state tax burden over time—a 0.14% (Model 3) increase in tax burden with a one standard deviation increase in per capita program service revenue (US$97.37). The magnitude of this correlation, however, does not outweigh the negative correlation of contribution revenue found earlier.²²

Overall, the results in Table 5 suggest that increasing amounts of per capita contribution revenue received by nonprofits providing services in the categories of (a) education, science, and agriculture and (b) health and medical are associated with reduced tax burdens imposed on states’ residents over time. Coincidentally, these categories are among the top three highest levels of per capita contribution revenues and therefore likely reflect categories of great amounts of nonprofit service provision. The negative correlations between these categories of nonprofit financing through contributions and state tax burden reiterate the previous findings that nonprofit provision of some services might be considered supplementary to government provision. These findings also provide some evidence that government might act as a free rider and reduce its tax burden imposed on residents commensurate with greater financial support for nonprofit service provision.

Conversely, Table 5 shows that increasing per capita contribution revenues received by nonprofits offering environmental and animal services, as well as per capita program service revenue generated by nonprofits offering services related to (a) arts, culture, and humanities and (b) education, science, and agriculture are associated with higher levels of state tax burdens over time. These findings suggest that some types of nonprofit activities should be considered complementary to government provision rather than supplementary. However, because our analysis does not allow us to speculate why this might be the case, we invite future research to address this particular question to help further our understanding of the complex relationship between the government and nonprofit sectors.

Conclusion

In this article, we examined the interconnectedness of the government and nonprofit sectors in relation to state tax burden. Specifically, we developed an econometric model to assess the relative influence of several factors, including nonprofit financing, on state tax burden. Overall, the regression results reveal several important findings. First, the findings support Reed & Rogers’ (2006) assertion that state tax burden is affected by (a) state tax policy, (b) nontax policy factors that affect state income, and (c) other exogenous factors that are independent of state tax policy and do not directly induce income. In this analysis, tax exporting, local IGR paid, debt burden, tax revenue limitation, and nonprofit financing show the most consistent correlations with state tax burden. The implication is that myriad factors are associated with state tax burden, and tax burden is more complex than traditional conceptualizations equating it solely with tax policy.

Second, the regression results revealed that financial support through contributions for nonprofit service provision does help to reduce state tax burden and does so at a meaningful level. In addition, nonprofit financing in the form of program service revenue may also have an impact, albeit smaller, on reducing tax burdens imposed on state residents. These findings are especially interesting in light of the way we measured nonprofit financing. The revenue sources we used to develop the two measures of nonprofit financing included all revenue received by nonprofits from governments. By controlling for government funding of nonprofits, the negative correlation with state tax burden implies a positive return for government financial support of nonprofit service provision. The implication of this finding is despite the cost to governments of contracts, grants, and other contributions to nonprofits, the service provision nonprofits offer in return helps to alleviate the tax burden that states must impose on their residents for service provision. Therefore, it appears that nonprofits indeed provide goods and services that are supplementary to government service provision.

Third, further analysis of nonprofit contribution and program service revenue differentiated by seven major categories of nonprofit services revealed that the supplementary nature of nonprofit service provision is not universal but rather might be a reflection of the types of services provided. The implication of these findings is that while some nonprofit service provision supplants government provision of services (such as health), other types of nonprofit activities (such as arts) should be considered complementary to government provision rather than supplementary.

Finally, by controlling for factors influencing demand for government service provision and state tax policy changes, the regression results provide some support for Becker & Lindsay’s (1994) theoretical and empirical assertion that government might act as a free rider by illustrating that increasing values of nonprofit financing in the form of contributions are associated with reductions in tax burdens imposed on state residents. Because tax revenue is the primary financing source for government service provision, this trend possibly results from reduced service levels in response to the alternative delivery of goods and services by nonprofit organizations. Perhaps most importantly in light of the current economic climate is this implication—that greater financial support for nonprofit service provision helps to alleviate some of the burden of government to provide particular services, thereby reducing the tax burden imposed on residents.

Footnotes

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Deborah A. Carroll is an Associate Professor in the Department of Public Administration & Policy at the University of Georgia. Her research focuses in part on the interconnectedness of the public and nonprofit sectors and the implications for tax policy, tax burden, and public service provision.

Thad D. Calabrese is an Assistant Professor in the School of Public Affairs at Baruch College – CUNY, and is also affiliated with the Center for Nonprofit Strategy and Nonprofit. His teaching and research focus on public and nonprofit financial management.

References

Abrams

B. A.

Schmitz

M. D.

(1984). The “crowding-out” effect of governmental transfers on private charitable contributions: Cross-section evidence. National Tax Journal, 37, 563-568.

Allers

De Haan

Sterks

(2001). Partisan influence on the local tax burden in the Netherlands. Public Choice, 106, 351-363.

Alt

J. E.

Lowry

R. C.

(1994). Divided government, fiscal institutions, and budget deficits: Evidence from the states. American Political Science Review, 88, 811-828.

Andreoni

Payne

A. A.

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

Anheier

H. A.

(2005). Nonprofit organizations: Theory, management, policy. New York, NY: Routledge.

Becker

Lindsay

C. M.

(1994). Does the government free ride? Journal of Law and Economics, 37, 277-296.

Bell

M. E.

Bowman

J. H.

(1987). The effect of various intergovernmental aid types on local own-source revenues: The case of property taxes in Minnesota. Public Finance Quarterly, 15, 282-297.

Bergstrom

T. C.

Blume

Varian

(1986). On the private provision of public goods. Journal of Public Economics, 29, 25-49.

Berry

Ringquist

Fording

Hanson

(1998). Measuring citizen and government ideology in the American states, 1960-1993. American Journal of Political Science, 42, 327-348.

10.

Bielefeld

Murdoch

J. C.

(1997). The influence of demographics and distance on nonprofit location. Nonprofit and Voluntary Sector Quarterly, 26, 207-225.

11.

Bielefeld

Murdoch

J. C.

(2004). The locations of nonprofit organizations and their for-profit counterparts: An exploratory analysis. Nonprofit and Voluntary Sector Quarterly, 33, 221-246.

12.

Bowman

(2004). Confidence in charitable institutions and volunteering. Nonprofit and Voluntary Sector Quarterly, 33, 247-270.

13.

Boyne

G. A.

(1998). Bureaucratic theory meets reality: Public choice and service contracting in U. S. local government. Public Administration Review, 58, 474-484.

14.

Bradley

Holden

McClelland

(2005). A robust estimation of the effects of taxation on charitable contributions. Contemporary Economic Policy, 23, 545-554.

15.

Brooks

A. C.

(2000). Public subsidies and charitable giving: Crowding out, crowding in, or both? Journal of Policy Analysis and Management, 19, 451-464.

16.

Brooks

A. C.

(2003). Do government subsidies to nonprofits crowd out donations or donors? Public Finance Review, 31, 166-179.

17.

Brooks

A. C.

(2007). Income tax policy and charitable giving. Journal of Policy Analysis and Management, 26, 599-612.

18.

Brudney

J. L.

Fernandez

Ryu

J. E.

Wright

D. S.

(2005). Exploring and explaining contracting out: Patterns among the American state. Journal of Public Administration Research and Theory, 15, 393-419.

19.

Buchanan

J. M.

Tullock

(1962). The calculus of consent. Ann Arbor: University of Michigan Press.

20.

Calabrese

(2011). Public mandates, market monitoring, and nonprofit financial disclosures. Journal of Accounting and Public Policy, 30(1), 71-88.

21.

Cameron

A. C.

Trivedi

P. K.

(2010). Microeconometrics using Stata, Revised Edition. College Station, TX: StataCorp LP.

22.

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.

23.

Case

A. C.

Rosen

H. S.

Hines

J. R.

(1993). Budget spillovers and fiscal policy interdependence. Journal of Public Economics, 52, 285-307.

24.

DeHoog

R. H.

Salamon

L. M.

(2002). Purchase-of-service. In Salamon

L. M.

(Ed.), A guide to the new governance (pp. 319-339). Oxford, UK: Oxford University Press.

25.

DeVita

C. J.

Manjarraz

Twombly

E. C.

(1999). Organizations and neighborhood networks that strengthen families in the District of Columbia: Final report to the Annie E. Casey foundation. Washington, DC: The Urban Institute.

26.

Donnahoe

A. S.

(1947). Measuring state tax burden. Journal of Political Economy, 3, 234-244.

27.

Downs

T. A.

Greenstein

S. M.

(1996). Understanding the supply decisions of nonprofits: Modeling the location of private schools. RAND Journal of Economics, 27, 365-390.

28.

Driscoll

J. C.

Kraay

A. C.

(1998). Consistent covariance matrix estimation with spatially dependent panel data. Review of Economics and Statistics, 80, 549-560.

29.

Ferris

J. M.

(1998). The role of the nonprofit sector in a self-governing society: A view from the United States. Voluntas, 9(2), 137-151.

30.

Figlio

D. N.

Kolpin

V. W.

Reid

W. E.

(1999). Do states play welfare games? Journal of Urban Economics, 46, 437-454.

31.

Froelich

K. A.

Knoepfle

T. W.

(1996). Internal Revenue Service 990 data: Fact or fiction? Nonprofit and Voluntary Sector Quarterly, 25, 40-52.

32.

Froelich

K. A.

Knoepfle

T. W.

Pollak

T. H.

(2000). Financial measures in nonprofit organization research: Comparing IRS 990 return and audited financial statement data. Nonprofit and Voluntary Sector Quarterly, 29, 232-254.

33.

Giving USA. (2009). Giving USA foundation and the center on philanthropy at Indiana University. Retrieved from http://www.givingusareports.org/

34.

Gronbjerg

K. A.

(1993). Understanding nonprofit funding: Managing revenues in social services and community development organizations. San Francisco, CA: Jossey-Bass.

35.

Gronbjerg

K. A.

Paarlberg

(2001). Community variations in the size and scope of the nonprofit sector: Theory and preliminary findings. Nonprofit and Voluntary Sector Quarterly, 30, 684-706.

36.

Gronbjerg

K. A.

Paarlberg

(2002). Extent and nature of overlap between listings of IRS tax-exempt registration and nonprofit incorporation: The case of Indiana. Nonprofit and Voluntary Sector Quarterly, 31, 565-594.

37.

Handy

Seto

Wakaruk

Mersey

Mejia

Copeland

(2010). The discerning consumer: Is nonprofit status a factor? Nonprofit and Voluntary Sector Quarterly, 39, 866-883.

38.

Hansmann

(1980, April). The role of nonprofit enterprise. Yale Law Journal, 89, 835-901.

39.

Hettich

Winer

(1999). Democratic choice and taxation. New York, NY: Cambridge University Press.

40.

Hirth

R. A.

(1999). Consumer information and competition between nonprofit and for-profit nursing homes. Journal of Health Economics, 18, 219-240.

41.

James

(1998). Commercialism among nonprofits: Objectives, opportunities, and constraints. In Weisbrod

B. A.

(Ed.), To profit or not to profit: The commercial transformation of the nonprofit sector (pp. 271-286). Cambridge, UK: Cambridge University Press.

42.

Kennedy

(1998). A guide to econometrics (4th ed.). Cambridge, MA: MIT Press.

43.

Kingma

B. R.

(1989). An accurate measure of the crowd-out effect, income effect and price effect for charitable contributions. Journal of Political Economy, 97, 1197-1207.

44.

Ladd

H. F.

(1992). Mimicking of local tax burdens among neighboring counties. Public Finance Review, 20, 450-467.

45.

Lile

S. E.

Soule

D. M.

(1969). Interstate differences in family tax burdens. National Tax Journal, 22, 433-445.

46.

Matsunaga

Yamauchi

(2004). Is the government failure theory still relevant? A panel analysis using US state-level data. Annals of Public and Cooperative Economics, 75, 227-263.

47.

McPherson

J. M.

(1983). An ecology of affiliation. American Sociological Review, 65, 791-823.

48.

Mullins

D. R.

Wallin

B. A.

(2004). Tax and expenditure limitations: Introduction and overview. Public Budgeting & Finance, 24, 2-15.

49.

Paarlberg

L. E.

Gen

(2009). Exploring the determinants of nonprofit coproduction of public service delivery: The case of K-12 public education. American Review of Public Administration, 39, 391-408.

50.

Papke

L. E.

(1987). Subnational taxation and capital mobility: Estimates of tax-price elasticities. National Tax Journal, 40, 191-203.

51.

Prante

(2008, August 7). State-local tax burdens dip as income growth outpaces tax growth. Retrieved from http://www.taxfoundation.org/files/sr163.pdf

52.

Reed

R. W.

(2006). Democrats, republicans, and taxes: Evidence that political parties matter. Journal of Public Economics, 90, 725-750.

53.

Reed

W. R.

Rogers

C. L.

(2006). Tax burden and mismeasurement of state tax policy. Public Finance Review, 34, 404-426.

54.

Roberts

(1984). A positive model of private charity and public transfers. Journal of Political Economy, 92(1), 136-148.

55.

Saidel

J. R.

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

56.

Salamon

L. M.

(1987). Of market failure, voluntary failure, and third-party government: Toward a theory of government-nonprofit relations in the modern welfare state. Nonprofit and Voluntary Sector Quarterly, 16(2), 29-49.

57.

Schiff

Weisbrod

B. A.

(1991). Competition between for-profit and nonprofit organizations in commercial markets. Annals of Public and Cooperative Economics, 62, 619-640.

58.

Smith

S. R.

Lipsky

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

59.

Steinberg

Gray

B. H.

(1993). The role of nonprofit enterprise in 1993: Hansmann revisited. Nonprofit and Voluntary Sector Quarterly, 22, 297-316.

60.

Trussel

(2003). Assessing potential accounting manipulation: The financial characteristics of charitable organizations with higher than expected program-spending ratios. Nonprofit and Voluntary Sector Quarterly, 32, 616-634.

61.

Tuckman

Chang

(1991). A methodology for measuring the financial vulnerability charitable nonprofit organizations. Nonprofit and Voluntary Sector Quarterly, 20, 445-460.

62.

Van Slyke

D. M.

Roch

C. H.

(2004). What do they know, and whom do they hold accountable? Citizens in the government-nonprofit contracting relationship. Journal of Public Administration Research and Theory, 14, 191-209.

63.

Warr

P. G.

(1982). Pareto optimal redistribution and private charity. Journal of Public Economics, 19(1), 131-138.

64.

Wasylenko

M. J.

(1980). Evidence of fiscal differentials and intrametropolitan firm relocation. Land Economics, 56, 339-349.

65.

Weisbrod

B. A.

(1977). The voluntary nonprofit sector. Lexington, DC: Heath and Company.

66.

Weisbrod

B. A.

(1997). The future of the nonprofit sector: Its enterwining with private enterprise and government. Journal of Policy Analysis & Management, 16, 541-555.

67.

Weisbrod

B. A.

(1998). Modeling the nonprofit organization as a multiproduct firm: A framework for choice. In Weisbrod

B. A.

(Ed.), To profit or not to profit: The commercial transformation of the nonprofit sector (pp. 47-64). Cambridge, UK: Cambridge University Press.

68.

Wolpert

(1993). Decentralization and equity in public and nonprofit sectors. Nonprofit and Voluntary Sector Quarterly, 22, 281-296.

69.

Wooldridge

J. M.

(2006). Introductory econometrics. Cincinnati, OH: Thomson South-Western.

70.

Young

D. R.

(2006). Complimentary, supplementary, or adversarial? Nonprofit-government relations. In Powell

W. W.

Steinberg

(Eds.), The nonprofit sector: A research handbook (2nd ed., pp. 37-80). New Haven, CT: Yale University.