Intersecting Sectors ? The Connection Between Nonprofit Charities and Government Spending

• Intersecting Sectors? The Connection Between Nonprofit Charities and Government Spending Deborah A. Carroll Thad D. Calabrese • Social Capital and Economic Development: A Neighborhood Perspective Matthew J. Hanka Trent Aaron Engbers • The Truth About Honesty Dominic D. Wells Anthony D. Molina • Revenue Sources and Social Media Engagement Among Environmentally Focused Nonprofits John R. McCaskill James R. Harrington Current Issues in Practice • University Housing Development: A PPP Approach Bruce L. Blair Adam W. Williams

A case in which government intervention in the market economy is deemed necessary is when the public sector acts as a vehicle for overcoming free rider problems. In such a case, private provision of public goods or services would be inefficient and/or socially suboptimal. Therefore, government might directly provide such goods or services or subsidize the private provision of these goods or services to achieve efficiency gains or the socially desirable level of provision. Publicly financed private provision also might take the form of contracting with the charitable nonprofit sector to provide goods or services.
Existing literature has found that private donations to charities generally decrease as a result of government financing; effectively, donors free ride off government provisions by reducing their own contributions (Abrams & Schmitz, 1984;Andreoni & Payne, 2011;Brooks, 2003Brooks, , 2004Gruber & Hungerman, 2007;Kingma, 1989;Roberts, 1984). A caveat, however, is the very important assumption that donors are aware of the amount of government funding their preferred charity receives and reduce their private contributions in response (Horne, Johnson, & Van Slyke, 2005). However, through survey research designed to test this underlying assumption, Horne et al. (2005) found that donors have little knowledge of government subsidies to nonprofits and are unlikely to change their giving behavior in response to government funding. More recently, Thornton (2014) found mixed results with respect to the relationship between government funding of charitable nonprofits and demand for contributions. In addition, econometric approaches used to overcome endogeneity within this literature indicate the likelihood of reverse causation (Bradley, Holden, & McClelland, 2005;Brooks, 2007). Therefore, we believe a void remains in the literature regarding the interconnectedness between the charitable nonprofit and government sectors' activities and financing.
In this paper, we expand upon the median voter model used by Becker and Lindsey (1994) to offer an alternative theoretical explanation of the government-charity free rider problem. We base our explanation upon the well-established theories of rent-seeking behavior of nonprofit charities and budgetary discretion behavior of public agents. In summary, we argue that government tax exemptions at the state and local levels are pivotal for nonprofit charities' capital formation. These charities have incentives to incorporate in higher tax locations and lobby governments for both continued tax benefits and direct financing for services, resulting in increased government spending (Lecy & VanSlyke, 2013;Luksetich, 2008). In doing so, nonprofit charities behave as classic interest groups or rent seekers to advocate for government policies (Buffardi, Pekkanen, & Smith, 2015;Fyall, 2016). Further, because nonprofit charities and governments collaboratively provide public services, government agents might realize increased budgetary discretion through these relationships (Jang & Feiock, 2007). Specifically, governments can rely upon nonprofit charities to provide particular public goods, so they can focus spending elsewhere. Hence, spending that flows through charities is easier to alter than outright cuts in the government provision of public services or altering debt obligations and/or personnel contracts.
In addition to establishing this theoretical framework, we conduct the first large-scale national study that combines two comprehensive data sets -one on government spending and the other on charitable spending -to analyze our hypothesized correlations between the sectors. Our results suggest that charitable spending is associated with higher government spending, all else equal, which is consistent with rent-seeking behavior of nonprofit charities. However, when we break apart government and charitable spending into specific service areas, we see that this theory is supported in some important areas such as public welfare but not in others. And, contrary to existing assumptions in the extant literature, we find little empirical evidence supporting the notion that government agents strategically reduce spending in particular areas in which nonprofit charities are operating.
The implications of our findings suggest that government spending may increase beyond optimal levels. This could lead to tax burdens that are greater than necessary, crowding out of private enterprise, and/or spending patterns that are difficult to alter in light of fiscal shocks. Further, nonprofit charities that become increasingly dependent upon government support may find themselves vulnerable when significant economic contractions require subnational governments to reduce spending to meet balanced budget requirements. However, nonprofit charities provide public goods with significant positive externalities (Bises, 2000). Therefore, it is imperative for policymakers to understand the potential relationship between charitable and government service provision. Our findings suggest that nonprofit charities do not simply implement policies; they may be associated with the resources made available for particular public services.

Nonprofit Charities, Tax Exemptions, and Rent-Seeking
Flowing from the mid-1970s work of Martin Feldstein and his coauthors, early research began an ultimately inconclusive quest into the structure of the U.S. income tax in terms of its treatment of charitable contributions and its association with the amount and distribution of giving and philanthropic activities. By providing evidence from various data samples of the price elasticity of charitable giving due to the federal deductibility of charitable contributions from taxable income, which lowers the price of charitable contributions relative to other goods, Feldstein (1975), Feldstein and Clotfelter (1976), and Feldstein and Taylor (1976) were among the first to associate government tax exemptions with outcomes related to the charitable nonprofit sector. Clotfelter (1980) followed these studies with further evidence that the price of giving is associated with charitable contributions beyond simply the effect of itemization, although short-run income and price elasticities are smaller than the long-run impact on giving. More recent evidence suggests that taxes have both transitory and persistent price and income effects on levels of contributions, and the persistent component of price and income changes is more influential on charitable donations than are the transitory ones (Auten, Sieg, & Clotfelter, 2002).
In addition to the federal deductibility of charitable contributions, the federal government exempts nonprofit charitable organizations that are registered with the Internal Revenue Service from income taxation on annual mission-related profits. Although this exemption is typically thought of as a general public subsidy for charities, Hansmann (1981) suggests this argument is not at all compelling. For example, if nonprofit charities naturally emerge in response to contract failure, it makes little sense to offer incentives for an organic response. 1 Rather, Hansmann (1981) posits that the tax exemption is best justified as compensation for constraints nonprofit charities face on capital formation, especially in industries in which charities are more efficient service providers than for-profits. Capital formation in the nonprofit charitable sector is especially dependent upon retained profits because of the lack of equity markets. Thus, the tax exemption permits charities to increase their retained earnings on a pretax basis.
While Hansmann (1981) focuses on the federal income tax, states and localities levying similar corporate income taxes follow the federal example and also exempt registered nonprofit charities' mission-related profits from taxation. Hence, state and local tax exemptions also encourage nonprofit charities' capital formation. Because much of nonprofit charities' capital takes the form of fixed assets (Calabrese, 2013), this tax exemption is especially valuable at the state and local levels because of the variation in types and effective rates of taxes imposed upon these types of assets. In addition to income tax exemptions, registered nonprofit charities are usually exempt from sales taxes that are levied at the state level, as well as property taxes at the state and/or local level. Of course, such exemptions also assume that earnings, purchases, and property are used for mission-related services. While states and localities with higher taxes are indicative of higher demand for public spending, all else equal, nonprofit charities should be expected to locate in these areas to maximize the value of their tax exemption benefits. In fact, the literature has largely found that the number of nonprofit organizations actually operating in high-stress and high-need areas tends to decrease (Bielefeld, 2000;Grønbjerg & Paarlberg, 2001;Peck, 2008). In addition, Twombly (2003) found that economic need is not associated with nonprofit entry into metropolitan regions.
Critically for our purposes, nonprofit charity dependency on governments for capital formation via tax exemption provides incentive for particular types of charitable behavior. Specifically, representatives might expend economic resources to "lobby" 2 state and local government 1 Smith and Grønbjerg (2006) document the "civil society" framework that explains nonprofit formation as the embodiment of the values necessary for democracy and good government. This is yet another framework that would explain nonprofit charity tax exemption not as a subsidy but as something that keeps a vital sector independent from the government. 2 Incorporated public charities qualifying for federal tax exemption under 501(c) of the Internal Revenue Code are restricted from spending organizational funds on lobbying and campaign activities. However, IRS restrictions do not prevent nonprofit charities from courting government officials as a voting bloc (employees, board members, and also recipients of the charitable service), as a source of campaign funds from individuals employed or volunteering at these nonprofit charities (including board members), or as a source of legitimation for addressing certain social needs. officials for continued tax exemptions (Knauer, 2010). Alternatively, these individuals might advocate for increased government financing for public service areas in which they have a vested interest. This is the essence of rent-seeking behavior (Anderson, 2012). In this regard, nonprofit charities behave as an interest group with preferences for beneficial government tax and expenditure treatment -in the aggregate or in specific service delivery areas (Buffardi et al., 2015;Fyall, 2016). Such behavior is similar to that articulated by Mosley (2012), in which social service providers advocate with government funders to ensure direct public financing for charities continues over time. In fact, several studies have found that government funding of charitable nonprofits leads to greater geographical concentration and/or survival of such organizations (Hager, Galaskiewicz, & Larson, 2004;Lecy & Van Slyke, 2013;Luksetich, 2008;Matsunaga &Yamauchi, 2004;Twombly, 2003).
Overall, therefore, we argue that tax exemptions encourage nonprofit charities' capital formation, and tax exemptions are even more valuable at state and local levels where variations in tax policy occur. We further argue that the dependence of nonprofit charities on these tax exemptions will lead these organizations to advocate for continuation and expansion of tax benefits and/or direct financing in service areas in which charities operate. Such rent-seeking in nonprofit charities articulating self-interested behavior is expected to be associated with greater amounts of public funds directed toward them.

Rent-Seeking and Government Spending
The median voter model describes how individual demands are aggregated under majority rule; it suggests market demand when aggregated by majority rule will equal the demand of the median voter (Holcombe, 1989). As such, the model helps to explain collective preferences in light of the focus of public choice theory on the individual; in certain circumstances, the median voter's preference will emerge as the collective preference in a majority rule election system (Holcombe, 1989). By extension, scholars use the median voter model as the basis for suggesting that government will provide services in response to the median voter's desired level, thereby satiating demand of the median voter and every voter with preferences for public goods below the median voter's desired level. An implication of this outcome is that there will be voters with preferences for public goods provision at a level above that desired by the median voter, and these above-median voters' demands would be left unmet. Weisbrod (1977) explained that, because the government provides public goods based on the preferences of the median voter in a jurisdiction, a minority of voters will desire additional public goods not provided by government, assuming a certain degree of heterogeneity of tastes within the population. This unmet demand for additional goods -termed "government failure" -leads these voters to provide voluntary contributions to the nonprofit charitable sector to meet their demand for additional public goods; these contributions are viewed essentially as voluntary taxes.
However, Tiebout (1956) explains that voters with unmet demands would simply move to a jurisdiction ("vote with their feet") where the median preference was closer to their own. 3 The implication is that, rather than voluntarily contribute to nonprofit charities to provide additional public goods, voters will instead sort themselves into communities with relatively homogeneous demands for such goods. Ferris (1998) indicates this will lead to government (rather than charitable nonprofit) provision. In addition, a federal system of government reduces the need for actual relocation to occur to achieve the outcomes of Tiebout sorting. In particular, voters with unmet demands from one government may have their desired level of public goods supplied by another governmental entity -either a different level of government or an overlapping special district government (Galle, 2011). Therefore, with perfect Tiebout sorting, there will be no unmet demand. Moreover, even with restrictions to Tiebout sorting, a federal system of government with multiple service providers (i.e., federal, state, county, municipal, special district, etc.) will satisfy most preferences because less actual relocation is required due to both vertical and horizontal competition (as well as collaboration) among government service providers.
In addition, Becker and Lindsey (1994) suggest the median voter is indifferent to which sector produces public goods as long as the correct amount of services is provided. In fact, most citizens free ride on the political activities of others or are entirely ignorant of their own stake in a policy outcome (Wilson, 1989). In addition, Handy et al. (2010) found that most consumers cannot distinguish between charitable and government service providers. Therefore, those voters or interest groups with strong preferences for specific outcomes (for specific spending on particular public goods, for example) are likely to dictate how politicians allocate public spending. This is made possible by the public input requirements inherent in government budget processes as well as through referendum requirements many governments now face. In such a case, one implication of rent-seeking is that total government spending may increase because of advocacy.
This complements public choice theory, which posits that elected officials and bureaucrats are self-interested agents and therefore might be persuaded to meet the demands of specialized groups rather than the preferences of the median voter because they can collect rewards from these groups (Olson, 1965). Specifically, "charitable tax subsidies represent the endorsement of a fundamental reallocation of responsibility between the federal government and the charitable community for certain social services. This newly configured charitable community has something to offer both the self-interested and the public-spirited legislator. In fact, reallocation offers the self-interested legislator a golden management opportunity. It permits the legislator to: (i) shift the responsibility for certain social services; (ii) claim credit for encouraging more efficient delivery of needed social services; and (iii) avoid accountability for any unfavorable consequences" (Knauer, 2010, p. 976).
Although public agents may be budget-maximizers (Niskanen, 1971), governments often face institutional rules and restrictions from citizen-imposed tax and expenditure limitations (TELs), balanced budget legislation, and referendum requirements for tax or spending proposals, which constrain the budgetary authority of public agents (Mullins & Wallin, 2004). On the other hand, tax expenditures, which are simply the estimated real cost of tax deductions and exemptions inherent in a government's tax system legislation, are not considered direct spending or taxation. Therefore, they are not subjected to the same rules and restrictions such as TELs and referendum requirements imposed by citizens. In fact, tax expenditures are often invisible to the median voter, even though they present a real cost for government service provision. Although less efficient from an economic standpoint, government agents are not necessarily interested in eliminating tax exemptions for nonprofit charities because tax-supported service provision by charities might provide government officials with greater budgetary flexibility. This is especially relevant when the perceived costs of a policy decision are widely distributed across the tax base but the perceived benefits are concentrated (Wilson & DiIulio, 1995). Current research holds that nonprofit charities and governments often develop collaborative relationships that are interdependent (Gazley & Brudney, 2007;Jang & Feiock, 2007;Salamon & Toepler, 2015). In this context, we expect governments to strategically use the charitable sector as providers of certain public goods and services due to their expertise in an area, perceived cost savings, or other budget maximizing incentives (Knauer, 2010). If a charitable service provider is already providing some public good at a level demanded by the public, government expenditures are not needed. Even if public demand is not being fully satiated, fewer government expenditures are needed to provide the public good than in the absence of charitable provision. In light of this theoretical framework we have explained, our primary research question is whether direct government spending changes in connection with spending by incorporated public charities.

Model Specification and Data
To address this research question, we use a typical cost model (DiPasquale & Wheaton, 1996;O'Sullivan, 2003) to estimate combined state-local government expenditures as a function of vectors of variables measuring the cost of providing public services (C), the division of service responsibility (D) between overlapping governments (in this case state-local vs. federal), and services demanded by citizen voters (S), as shown in equation 1.
Government expenditures (EXP) are measured as aggregate direct expenditures. This measure includes both operating and capital expenditures (but excludes intergovernmental transfers) of the state government and all sub-state general-purpose, special-purpose, and independent school district governments within the state. Total expenditures are then divided by state population to scale the measure on a per capita basis. Each state-local per capita aggregate (i) for each year (t) during 1989-2006 is our unit of analysis. The time period of analysis represents the complete time frame for which all data are available. All financial data were adjusted for inflation using the Consumer Price Index. This measure captures all government spending within a state geography and comprehensively measures total government provision of services within states each year regardless of which government unit is providing it. Data for this variable were obtained from the Census of Governments Survey conducted by the U.S. Census Bureau.
We capture the cost of providing public services (vector C in equation 1) and the division of service responsibility between overlapping governments (vector D in equation 1) with variables measuring net federal receipts to state-local governments and the federal tax burden of citizens within each state-local government, respectively. Net federal receipts measure the cost of receiving federal intergovernmental revenue relative to taxes paid by residents of each state geography. Values greater than $1 indicate a cost for receiving federal income, as residents of those states are subsidizing states in which residents pay less than $1 for an equivalent amount of federal aid received. In the former, it is costlier to expand public service provisions through the use of income from the federal government. Our regression models all also include a series of year-fixed effects to capture variation in general price levels that occur over time and affect the cost of providing public services among state geographies equally but do not vary between state geographies, as does the net federal receipts variable. Federal tax burden is the amount of federal tax revenue derived from residents of a state geography divided by total personal income of residents within that state's geography. A higher federal tax burden presumes greater responsibility for service provision at the federal level versus at the state and local levels of Similar to the Census of Governments survey data, these data are the most comprehensive nationwide source of nonprofit charities' finance information. 5 We measure charitable spending in three ways for the reasons discussed below: 1) nonprofit charity total expenses; 2) nonprofit charity expenses, net of grant income received from government sources; and 3) nonprofit charity expenses, net of grant and contract income from government sources. Each measure is calculated as the aggregate amounts of expenses and/or grant/contract income for all nonreligious nonprofit charities that filed IRS Form 990 each year within each state geography. We also divide these total expenses by the same measure of state population used to calculate per capita values of our dependent variable for purposes of scaling and comparability. To do this, we aggregated organization-level data to the state geography by summing values of relevant financial information based upon the address reported by nonprofit charities on IRS Form 990 and FIPS codes that indicate the state domicile of each organization. Existing research shows the nonprofit charity sector is overwhelmingly community-based and locally oriented, such that both charitable financing and spending for service provision primarily occur within state borders (Bielefeld & Murdoch, 2004;Bielefeld, Murdoch, & Waddell, 1997;Calabrese, 2011;Downs & Greenstein, 1996;DeVita, Manjarraz, & Twombly, 1999). 4 The $25,000 minimum requirement was the rule during the time period of our data; the current minimum gross revenue for required IRS Form 990 filing is $50,000. 5 There are concerns about using the 990 data. Many of these concerns focus on allocations between programs and overhead, how unrealized gains or losses are accounted for, the timeliness of filings, and annual change in net assets (Gordon, Khumawala, Kraut, & Meade, 2007). These concerns are less relevant for the data employed in our empirical strategy because the analyses rely upon aggregate (rather than organizational) data. Thus, the effect from an individual error is minimized. In addition, the patterns of these errors are relatively random. As such, most studies have found these data to be reliable sources of information for nonprofit finance (Carroll & Stater, 2009;Froelich, Knoepfle, & Pollak, 2000).
Because our dependent variable of state-local government expenditures includes all spending by governmental units within a state geography, it is necessary to remove from our independent variable of charitable spending all income derived from grants and contracts awarded by governments. Otherwise, charitable spending likely includes resources accounted for simultaneously in government spending. To do this, we use data from two different data sets provided by the NCCS for constructing the independent variables. Our first measure of charitable spending, which does not subtract government grants and contracts income to nonprofit charities, is derived from the Core file and is simply total annual expenses. The Core file includes all registered nonprofit charities required to file IRS Form 990, but the data include very few variables for analysis. Although this measure of charitable spending is imperfect, it allows us to empirically test data from 1989 through 2006 on nearly all nonprofit charities that exist; therefore, this measure most comprehensively captures spending by the entire nonprofit charity sector, so we use it simply as a benchmark for our other two measures.
To construct our second measure of charitable expenses, which nets out grant income from government sources, 6 we rely upon the Statistics of Income (SOI) file, which is weighted toward larger nonprofit charities. Although we lose observations of smaller nonprofit charities by using this data set, we are still able to capture the majority of aggregate spending by nonprofit charities in the sector. In addition, we are still able to analyze data from 1989 through 2006 by using this data.
Our third measure of charitable expenses, which nets out both government grants and earned income from government contracts, 7 also is calculated using the SOI file. However, the data on government contract revenue is only available beginning in 2000. Therefore, while the third measure of charitable spending is arguably the most proper for our analysis, its use drastically reduces the time period and sample size of our panel. We believe the longer time frame available with our other two measures of charitable spending allows us to provide better estimates of the factors that correlate with government spending within states over time. However, examining the nuances and consistencies of results from all three specifications will enable us to draw better overall conclusions from our findings. Complete descriptions of all variables and data sources can be found in table 1.

Descriptive Statistics
Descriptive statistics for all variables are shown in table 2. Means and standard deviations are shown for the entire time period of this study (1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006) Table 2 also reveals less variation for government expenditures than for charitable expenses as measured by the standard deviations. Finally, table 2 shows patterns of spending for nonprofit charities that suggests at least some connection with government income received by these organizations. For both time periods, as income from government sources is subtracted from total expenses, the mean per capita values consistently decline. If government funding of nonprofit charities has no implications for this sector of charitable spending, we would likely see no such trend. However, our interpretation of these descriptive statistics is that government grants and contracts act as an impetus, at least to some extent, for spending in particular areas of the nonprofit charity sector. Not only does this preliminary evidence present a valid case for future research, it also reiterates our initial assumption of potential endogeneity between government and charitable spending. The next section explains the precautions we have taken for proper model estimation to analyze our primary research question.

Model Estimation
Prior to running any regression, several estimation issues were identified and addressed. First, we conducted Hausman specification tests, which indicated that a fixed effects estimator would provide consistent results for estimating equation 1 using any of the three measures of charitable spending. 8 As a result, we proceeded by testing for heteroskedasticity using the Modified Wald test for group-wise heteroskedasticity in a fixed-effect regression model. Although the presence of heteroskedasticity would not bias our results, it does make our estimators inefficient. These tests indicated the presence of heteroskedasticity in model specifications using all three measures of charitable spending. 9 Therefore, to avoid using an estimator that is not fully efficient and to ensure the robustness of the remaining specification test statistics (Cameron & Trivedi, 2010), the Huber-White sandwich estimator of the variance was used to produce heteroskedasticity-robust standard errors except where cluster-robust standard errors are indicated.
As previously explained, due to extant research, we assume our measures of charitable spending are endogenous or correlated with the error term and therefore might make our estimators inconsistent. While using an instrumental-variables (IV) approach would provide a consistent estimator, it only does so under the assumption that valid instruments exist (Cameron & Trivedi, 2010). Typically, lagged values of endogenous variables are predetermined and are treated as exogenous variables because they are given constants for determination of the current time period's values of the endogenous variables (Kennedy, 1998). Using lagged values as instrumental variables is only valid, however, if the model specification is not correlated over time. Therefore, we used the Wooldridge test for autocorrelation in panel data to test for the presence of serial correlation. These tests yielded F-test statistic values of 2.90 (Prob > F = 0.09) using nonprofit charities' total expenses to measure charitable spending, 3.02 (Prob > F = 0.09) using nonprofit charities' expenses net of grant income received from government sources, and 2.04 (Prob > F = 0.16) using nonprofit charities' expenses net of grant and contract income from government sources. Because we failed to reject the null hypothesis of no first-order autocorrelation in all three specification tests using a 95% confidence level, using one-year lags of each of our three charitable spending variables as instruments should produce consistent estimators for each specification as suggested by Wooldridge (2006). Table 3 provides the regression results of the three model specifications related to each different measure of charitable spending for purposes of comparison. All three specifications use the same dependent variable of per capita direct government expenditures as well as independent variables to control for citizen demand for public services, cost of providing public services, and the division of service responsibility between state-local governments and the federal government. As noted earlier, these categories of results are representative of a typical cost model (DiPasquale & Wheaton, 1996;O'Sullivan, 2003). In addition, we selected one exogenous independent variable to represent citizen demand that directly corresponds to the functional expenditure categories of government spending that are included in our data.

Regression Results
Perhaps most striking from table 3 is the consistency of results both in terms of directional correlation and magnitude of coefficients between the various model specifications. Unsurprisingly, the regression models with the larger sample size that span a longer period of time are a better fit for the data and provide greater explanatory power. Nonetheless, the coefficient signs and results of hypothesis testing are consistent throughout all models. As expected, the results show that factors influencing citizen demand for public services such as education levels and insurance coverage significantly correlate with government spending. In addition, the liberalism of government officials within a state is positively associated with charitable spending, which is not unexpected because Democrats are typically associated with preferences for higher spending than their more conservative Republican counterparts. Finally, the federal portion of tax burden imposed upon residents within state geographies has a relatively large correlation with spending by state-local governments, suggesting the division of service provision responsibility between state-local governments and the federal government is perhaps the most important factor.
Focusing on the primary research question of this study, table 3 shows a consistently positive correlation between government and charitable spending, which is statistically significant at the 99% confidence level or above for all model specifications. According to the results, states with nonprofit charities that spend $1 per capita more than their counterparts located in average states are generally associated with state-local government expenditures that are between an average of 6 and 12 cents higher per capita over time. Using the descriptive statistics in Table 2, the results in Table 3 indicate that a state in which nonprofit charities have aggregate charitable expenses net of grant income from government sources of $12,333,049,720, or $5,426,034 higher 10 than the average state, direct expenditures of state and local governments within this above-average state should generally be $332,073 higher 11 as well. If these above-average expenditures were fully supported by tax revenue, state residents would face a state-local tax burden that is 6 cents 12 higher than for citizens located in average states.
While these amounts may seem small, consider a state in which charitable spending is one standard deviation higher per capita than the average state rather than only $  [1989][1990][1991][1992][1993][1994][1995][1996][1997][1998][1999][2000][2001][2002][2003][2004][2005][2006]. 16 The numbers cited in the remainder of this section were calculated using the same method explained in footnotes 12-17, except using data from tables 2 and 3 pertaining to nonprofit charity expenses, net of grants and contracts, and population values for 2000-2006.  Note: NEC = Not Elsewhere Classified might see a relatedly higher tax burden of $153 (or $612 for a family of four) compared with citizens residing in average states. A direct implication of these findings is a tighter budget constraint for individuals and families residing in states with greater spending by the charitable nonprofit sector, regardless of whether or not they receive such services.
Overall, our results imply that nonprofit charities have a nontrivial positive correlation with government spending. These results are inconsistent with government failure theory, which suggests that nonprofit charities provide services to address unmet service demands of those above the median voter. In such a case, we would expect to see no correlation between nonprofit charities and government spending because nonprofit charities would provide services beyond the government's ability or willingness to do so. Rather, our results lend empirical support to the rent-seeking theory outlined above. The results suggest that as nonprofit charities' activities increase, the sectors become interdependent and nonprofit charities, similar to an interest group, develop strong preferences for increased government spending on goods and services. As a result, we see a positive correlation between the two sectors' spending patterns.

Regression Results by Nonprofit Charity Mission Area
Next, we analyze whether the activities of nonprofit charities among sectors are associated with governments altering their own spending on corresponding functional expenditure areas. To explore this issue, we identified 11 categories of service provision for which spending by both state-local governments and nonprofit charities overlap, as shown in table 4. The left-hand side of table 4 provides the functional expenditure categories used to classify state and local government spending, as reported in the Census of Governments survey. The right-hand side of table 4 provides our matching of these government functional expenditure categories to nonprofit charity mission areas, as reported in IRS Form 990, using IRS National Taxonomy of Exempt Entity (NTEE) codes. The service provision categories we were able to identify as overlapping for both sectors are corrections, education, employment, judicial and legal, health, hospitals, housing and community development, libraries, parks and recreation, police, and public welfare. To analyze the potential correlation between government and charitable spending by service provision category, we disaggregated our dependent variable of state-local per capita direct expenditures and our three independent variables measuring nonprofit charities' expenses (i.e., total expenses; expenses net of government grants; and expenses net of government grants and contracts) into these comparable functional expense categories and mission areas. Spending comparisons between the two sectors by category are shown in table 5.
As can be seen in At the low end of the spending spectrum, state-local government expenditures are lowest, on average, for the largely federal responsibility of employment security administration and libraries at $18 and $20 per capita, respectively. Nonprofit charities, on average, spend the least amounts, net of both grant and contract income from government sources, on the traditional governmental responsibilities of corrections, judicial and legal services, police, and libraries with only $2, $4, $4, and $3 per capita, respectively. Corrections, police, and legal services also represent the largest differentials between the two sectors with nonprofit charities only spending 1.44%, 2.61%, and 4.11%, respectively, on average, of the average amounts spent by state-local governments. The smallest differential between the two sectors is in housing and community development, in which nonprofit charities spend nearly 74%, on average, of what state and local governments expend. Table 6 provides coefficient values for a series of regression analyses of government and charitable spending by service provision area. Each number in table 6 is the product of a unique regression model that utilizes the three charitable spending measures defined earlier. 17 These are indicated by the column headings, for each overlapping service area, for all service types, or for those categories of service provision in which there is no crossover between nonprofit charities' mission areas and government functional expenditure categories reported in our 17 Again, all measures of charitable spending were lagged one year to overcome potential endogeneity.  data. 18 Across the service provision areas, per capita state-local government expenditures pertaining only to those functional expenditure categories were used in calculating the dependent variables. Thus, for example, the row labeled "Legal" reports the results of three different regression models, all of which use state-local aggregate direct expenditures for the judicial and legal functional expenditure category only and are measured on a per capita basis. Along the row, however, each regression utilizes a different measure of charitable expenses, which is consistent with our three measures for the overall analysis, indicated by the column headings. These spending values were calculated by aggregating total expenses for only those nonprofit charities within a given state in a given year that report on their IRS Form 990 that their primary mission area falls within legal services or NTEE code I80 and are also measured on a per capita basis. Thus, all three specifications are different but relevant only to judicial and legal services.
The numbers across each row are the coefficient values with stars indicating the significance levels related to the charitable spending independent variables. The results for all other variables are not included for space and clarity concerns. All regressions were estimated using the exact same procedures and control variables explained above for analyzing total government and charitable spending. It should be noted that all of the regression models that produced the coefficient values illustrated in Table 6 were statistically significant at the 99% confidence level or above. 19 We believe table 6 shows some interesting patterns. It is obvious at first glance that there is a correlation between government and charitable spending for some service provision areas -but not others. Specifically, it appears that the positive and statistically significant correlation we found earlier between government and charitable spending is primarily driven by spending on 18 Nonprofit charity NTEE codes with no corresponding state or local spending function included arts, food and nutrition, international affairs, civil rights and advocacy, philanthropy, science and technology, social science, public and societal benefit, religiously related, and mutual/membership organizations. 19 F-test statistics range from 3.07 to 107.90; Prob > F = 0.00. judicial and legal services, libraries, and public welfare. This finding provides additional support for the rent-seeking theory of government-nonprofit charity relations in these service areas. Further, these results are supported even when all government grants and contracts are excluded from the analysis. Therefore, it is reasonable to conclude that this relationship is not simply reflecting government contracting with nonprofit charities.
According to the results, states in which nonprofit charities spend $1 per capita, which is approximately $5,800,347 more 20 than the average state on legal services, are generally associated with state-local direct expenditures for judicial and legal services that are at least $1.22 per capita or $7,103,685 higher 21 over time. A state with one standard deviation above average charitable spending on legal services, net of grant and contract income from government sources, which amounts to $14,732,881, would 22 be associated with state-local government spending in that state of $18,043,360 higher 23 over time than the average state. The amount has the potential to translate into an additional tax burden of $3.11 for each resident 24 of that state.
States with $1 per capita above-average expenses of nonprofit charities with primary mission areas in library and public welfare services would be associated with per capita spending for those functions that are 18 cents 25 and nearly 70 cents, respectively, higher over time than average states. If nonprofit charities providing library and public welfare services in a state had expenses, net of grant and contract income from government sources, for these services of one standard deviation above their counterparts in average states, which would amount to $29,904,855 and $585,018,938, respectively, residents of those states might expect their state and local governments to expend approximately $5,397,826 and $406,412,656 more over time on these functions, respectively, than that of average states.
To support the additional government expenditure on public welfare through tax revenue, each citizen of an above-average state would need to pay $70 more in taxes. For a family of four residing in such a state, they would be required to subsidize public welfare by $280 more tax burden -and these calculations inherently assume the tax burden would be applied equally to the state's population like a flat tax. However, if the additional tax revenue is generated from progressive state income taxes, wealthier individuals would be subsidizing public welfare through their higher tax burdens to greater extent than residents of the same state who fall into lower income tax brackets or are exempt from state income taxes altogether.
Perhaps most interesting about these findings is that public welfare is one of the largest categories of public service provision for both state-local governments and nonprofit charities as measured by their expenditures and expenses, respectively. It appears, therefore, that 20  -2006). 25 The numbers cited in the remainder of this section were calculated using the same method explained in footnotes 20-24, except using data from tables 5 and 6 pertaining to the service provision areas relevant to the discussion. charitable service provision of public welfare is associated with more government spending on public welfare, all else equal. Our theoretical explanation of nonprofit charities acting as political agents and advocating for increased government spending in this area seems reasonable and supported -perhaps as representatives of communities and citizens. These findings also are consistent with other findings provided by LeRoux (2007), Mosley (2012), and Smith and Pekkanen (2012). However, Young (2000) points out such advocacy may render nonprofit charities and governments as adversaries, further complicating the provision of public services. On this particular question, further research is certainly warranted.
On the other hand, legal services represent one of the smallest spending categories for nonprofit charities and one of the largest differentials in spending between state-local governments and nonprofit charities. Yet, the results show $1 of charitable spending is associated with more than $1 higher government spending. Whether this relationship reflects nonprofit charities successfully advocating for additional resources from the government, or whether nonprofit charities are able to draw significant resources from funders other than government, or even some other explanation, additional analysis on particular charitable subsectors seems warranted. We encourage other scholars to further explore these relationships and plan to do so ourselves in future research.
Aside from these results, what we find equally striking from table 6 is the lack of any statistically significant correlation between government and charitable spending for several service provision categories. Especially because many theories often presume an association between the government and nonprofit charity sectors, we believe these findings show no such correlation between government spending and several large and important nonprofit charity subsectors to be relevant and equally as intriguing as our statistically significant results. We interpret these findings as limited support for government failure theory, in that government and charitable service provision are likely independent.
For example, education is the single largest spending category for both state-local governments and nonprofit charities. Yet, the regression results show a rather weak correlation between the two sectors in providing this service. It is only when grant and contract revenue are both netted out of total expenses for nonprofit charities in this category that there is any statistically significant correlation between charitable expenses and state-local government expendituresand the coefficient is only marginally significant at the 90% confidence level. We take these findings as caution against suggesting charitable spending on education is associated with government spending or vice versa. Perhaps if elementary/secondary education were analyzed separately from higher education, a more definitive pattern in the data could be identified. We certainly believe there is more work to be done in future research on this particular category of service provision. In addition, the categories of health and hospitals, which are also two of the largest spending categories for both the government and nonprofit charity sectors, exhibit no statistically significant correlation. Again, we take the lack of statistical significance to suggest that the services provided by each sector in these categories are not associated with those of the other sector and vice versa.
The only service area that displays a negative relationship is parks and recreation, and this is not consistent or significant across estimations. In this particular case, the results seem to suggest that government officials might strategically pull back government spending on these services if nonprofit charities provide them. Parks and recreation services have seen significant publicprivate partnerships in which private nonprofit charities have assumed operations of public services for governments (Walker, 2004). Therefore, as charitable spending on Parks and Recreation increases, perhaps government spending declines in real terms. 26 More research in this particular area is needed, of course, to provide a definitive conclusion on the implications of public-private partnerships in parks and recreation services.
Overall, the results in table 3 suggest that charitable spending is associated with higher government spending, all else equal -which is consistent with rent-seeking behavior of nonprofit charities. However, when we break apart government and charitable spending into specific service areas, as in table 6, we see that this theory is supported in some important areas, perhaps most importantly in the area of public welfare, but not in others. Further, the results are consistent even when all government funds devoted to nonprofit charities are removed from the analysis. We find little empirical evidence that would support the notion that government agents strategically reduce spending in particular areas in which nonprofit charities are operating. Finally, the lack of relationship between government and charitable spending in many important service areas suggests that the two sectors often work independently of each otherwhich we interpret as empirical support for government failure theory.

Conclusion
In this paper, we articulated that rent-seeking behavior by nonprofit charities and budgetary discretionary behavior by public agents should lead to a positive correlation between nonprofit charity activity and government spending. Using a large national database of government spending that we merged with charitable spending, we empirically tested this question. Overall, we found a positive association between spending by both sectors, which is unequivocal and nontrivial, supporting the rent-seeking theory of nonprofit charities' behavior.
When we examined spending by the sectors by specific areas of service provision to determine public budgetary reallocation, our results indicate positive associations in legal and judicial services, libraries, and public welfare spending -supporting the rent-seeking explanation. We also found no correlation between spending by the two sectors in several important areas of service provision, including education, health, hospitals, and housing. The lack of correlation in these areas might be indicative of government failure theory rather than rent-seeking.
Importantly, the positive association between charitable and government spending suggests that public spending may increase beyond optimal levels -leading potentially to tax burdens that are greater than necessary, crowding out of private enterprise, and spending patterns that are difficult to alter in light of fiscal shocks. Further, nonprofit charities that become increasingly dependent upon government support may find themselves vulnerable when significant economic contractions require subnational governments to reduce spending to meet balanced budget requirements. Finally, we found evidence of direct government funding reductions in parks and recreation, which would support the notion of governments altering public budgets because of nonprofit charity activity.
Our results have important policy and management implications for governments as well as nonprofit charities. Obviously, both sectors respond to common citizen demands by providing public services. Understanding how these demands are met is important for predicting the size of the government sector, resultant tax burdens, and both financial and voluntary contributions to the nonprofit charity sector. Our results, which found a correlation between charitable and government spending in some areas but not others, inform our understanding of how governments and nonprofit charities interact and react to each other. Multiple theories explain this complicated and ever-changing relationship. Here we add to this theoretical tradition by considering nonprofit charity rent-seeking behavior and public choice theory as it relates to government budget changes in light of charitable service provision.
Certainly, we do not consider this study the definitive word on the subject. Our intention was to provide the first large-scale study of spending by state-local governments and nonprofit charities to inform the direction of future research. While our analyses provide new knowledge on the topic at hand, we believe the research also elicits more questions for ourselves and other scholars to study in the future. Future research should try to isolate the relationship at an even more local level, given the focus of most nonprofit charities on communities and locations. Another critical avenue left to explore is whether or not this spending relationship holds equally for capital and operating expenditures. Perhaps separating such spending will motivate future inquiry into the sectors' spending dynamics.

Disclosure Statement
The authors declare that there are no conflicts of interest related to this research, authorship, or publication of this article.
revenue diversification, and urban economic development, as well as the interconnectedness of the public and nonprofit sectors and the implications for tax policy, nonprofit management, and public service provision. Her research has appeared in the Journal of Public Administration Research and Theory, Public Administration Review, Public Budgeting and Finance, American Review of Public Administration, and Urban Affairs Review, among other academic journals.
Thad D. Calabrese teaches and researches in the field of public and nonprofit financial management. He is especially interested in the areas of employee benefits, the financial implications of collaborative governance and contracting, and capital structure decisions in public service organizations. Hanka
Over the past 30 years, there has been an evolving body of literature on social capital and economic development, which demonstrates that strong social networks foster entrepreneurial activity (Kim & Aldrich, 2005), improve job prospects (Erickson, 2001), and support the development of human capital (Croninger & Lee, 2001). There also has been significant developments in the understanding of the effects of social capital on the outcome that matters most for many people: household income. However, this body of literature on social capital's income-enhancing effects has been constrained by concerns about aggregation and the use of methodology with limited ability to demonstrate causation.
This paper looks at the relationship between economic resilience and social capital at the microlevel. This paper builds upon past research by looking at individuals' stock of social capital and its effect on their household income. Individuals are studied in the context of their neighborhood to control for macro-economic effects and to see the relative impact of neighborhood versus extraneighborhood social capital.

Literature Review
Over the past 30 years, there has been exponential growth in the number of studies conducted on social capital's antecedents and effects (Engbers, Thompson, & Slaper, 2016). Because of this growth and the significant amount of conceptual diffusion that has taken place, it is critical to distinguish between different types of social capital and their effects. Putnam (2000) defines social capital as the "connections among individuals-social networks and the norms of reciprocity and trustworthiness that arise from them" (p. 19). Social capital refers to interpersonal resources individuals can access through networks of both strong and weak social relationships (Beaudoin, 2011). Alternatively, Coleman (1988) defines social capital by its function within a social structure and suggests that changes in the relationships among people facilitate certain actions by actors.
At the most basic level, studies of social capital have tended to distinguish between bridging and bonding social capital (Putnam, 2000). Bridging social capital refers to the concept by which individuals are linked to others who fall outside their primary social circle. These relationships need not be substantive and are judged by their quantity and not their quality. Their primary advantage is that they provide access to non-redundant information. In terms of their economic effect, they provide access to customers, producers and suppliers necessary for economic growth (Kim & Aldrich, 2005). For example, Safford's (2009) comparative case analysis of rust belt cities finds that bridging social capital explained Allentown, Pennsylvania's recovery in the face of economic decline. Unlike in Youngstown, Ohio, where social capital was concentrated among the economic elite, Allentown's bridging social capital cuts across ethnic, religious, and economic groups. These cross-cutting ties were sufficient to rebuild the economy in Allentown.
Bonding social capital differs from bridging social capital in that its quality is based on intensity.
Bonding relationships are high in trust and are usually found within homogenous communities or in relationships based in shared history or experience such as familial or long-term friendship and work relationships (Coleman, 1988;Collins, Neal, & Neal, 2014;Putnam, 2001). The literature on bonding social capital also has emphasized the importance of community-based groups (Von Schnurbein, 2014). Nonprofits and other collective identity or service groups provide a place where individuals generate trust and form strong social ties. In recent years, these types of groups also have come to serve an important role in fostering economic development (Borzaga & Defourney, 2001;Stadtler & Probst, 2012).

Economic Development
Economic development refers to the intentional process of increasing the trajectory of a community's economic growth curve (Feldman, Hadjimichael, Kemeny, & Lanahan, 2014). A community's economic development has been linked to a wide range of characteristics that include the community's institutions, geography, economic conditions and social characteristics (Rodrik, Subramanian, & Trebbi, 2004). Community characteristics, such as a sound rule of law, clear state authority and a government with sufficient capacity to provide services and enforce contracts are critical for economic development (Rodrik et al., 2004). The economic development benefits derived from certain communities are based on that community's inherent geographic characteristics. Consider how climate, access to natural and other resources, transportation costs and proximity to knowledge centers shift a community's economic potential. Likewise, economic trajectories are determined by the degree of market integration with a community, its macroeconomic conditions, the occupational and industrial mix of a community, and access to capital (Rodrik et al., 2004). Last, economic development is determined by social conditions. Among these, human capital has received the most attention in the literature (Benhabib & Spiegel, 1994;Machlup, 2014). However, entrepreneurial and business culture (Harrison & Huttington, 2000) and social capital (Engbers, Rubin, & Aubuchon, 2017) also have been shown to be important.
The effects of economic development are numerous, affecting both social and economic conditions. Among economic factors, economic development is tied to job creation, GDP growth, trade, and increases in personal and household income. This study focuses on one aspect of economic development, which is the fostering of household income growth.

Determinants of Personal Income Growth
Like the process of economic development, personal and household income outcomes are associated with a wide range of determinants. These include both individual and social outcomes. Human capital accumulated through schooling and level of work experience are the biggest predictors of increases in income (Heckman, Stixrud, & Urzua, 2006). However, other factors such as personal health (Deaton, 2003), participation in illegal activities (Heckman et al., 2006), conducive family situations (Ellwood & Jencks, 2004), and access to transportation also are significant determinants (Hayaloğlu, 2015).
These individual factors are moderated by the social and economic conditions in which the household resides. In their study of U.S. counties, Rupasingha, Goetz, and Freshwater (2002) find that in addition to the economic characteristic of the individuals, household income is also affected by social characteristics such as higher levels of ethnic diversity, lower levels of income inequality, and higher levels of social capital, the latter of which is particularly important for this study.

Culture
Culture, understood as "shared social practices and the values and beliefs that legitimate them" (Coyle, 1993, p. 20), has been found to be an important foundation for social capital (Fukuyama, 1996) and subsequent economic development. For example, early work by Weber (1905), expounded on by Landes (1998), attribute the culture derived from the Protestant belief system, such as hard work, thriftiness, and "honesty in business" (Trigilia, 2001, p. 429), as particularly important determinants of economic growth and development in Northern Europe and the United States. These particular cultural traits, coupled with the Protestant ethic's application of universal human rights to all individuals (Maridal, 2013), lead to a much broader understanding of the social circle and facilitated the establishment of social networks that have been instrumental in influencing economic growth and development of an area (Triglia, 2001).
A well-cited example that builds on Weber's (1905) thesis of how cultural attitudes and behavior shape economic conditions can be found in Banfield's (1958) analysis of social and economic life in the small village of Chiaromonte in Southern Italy. Banfield observed that the villagers tended to maximize the interests of their immediate family over the interest of the citizenry to such an extent that it led to isolation, nepotism and widespread corruption. Banfield (1958) coined this lack of bridging social capital "amoral familism" and concluded that it prevented the village from addressing limitations on economic development such as extreme poverty, illiteracy, and basic infrastructure.
Using data from the World Values Survey, Marini (2004) and Maridal (2013) further developed and quantified Weber (1905) and Banfield's (1958) work by showing that certain cultural characteristics that stimulate achievement-motivation (see also McClelland, 1961) and develop social capital are necessary for economic growth and development. Achievement motivation is particularly important in the production stage, while social capital is necessary in the exchange stage and serves to enlarge economic markets.

Social Capital and its Economic Effect
While some of the earliest influential work on social capital has had an economic focus to it (Putnam, Leonardi, & Nanetti, 1994), more recent studies have made strides in using econometric analysis to demonstrate quantifiable effects of social capital on economic outcomes (Engbers, Rubin, & Aubuchon, 2017;Oh et al., 2014).
From a theoretical standpoint, there are multiple reasons to suspect an effect of social capital on economic development. These have been most clearly articulated by Woolcock (1998) who suggests that social capital effects job and income creation in four ways, which fit within a twoby-two matrix along the dimensions of embeddedness/autonomy and micro/macro scales. Micro social capital includes integration and intracommunity ties (embeddedness) or the linkage between extra-community networks (autonomy). In contrast, macro social capital features the synergy of state-society relations (embeddedness) or the organizational integrity of institutional capacity (autonomy). Macro embeddedness helps fosters ties that help groups overcome parochial interests necessary for long-term development, such as when procedural requirements like citizen commissions encourage good governance while mobilizing community actors around shared interests. Macro autonomy is concerned with the use of traditional hierarchy for increased performance. Formal organizations reduce collective action problems and facilitate the achievement of more socially optimal outcomes. Thus, communities with effective administrative systems are better able to deliver public services and foster trust in government.
At the micro level, embeddedness leads to access to the necessary economic resources that individuals need for economic advancement. For instance, consider how families, religious and ethnic groups or other tight-knit communities provide access to capital necessary for economic activity. Last, micro autonomy encourages economic actors to reach out beyond their primary community's interests to build linkages with outside groups. This reduces the likelihood that economic resources are not drained by the community of origin (Woolcock, 1998). Empirically, these conclusions are demonstrated in a wide range of contexts. The following sections will begin with an examination of individuals within a context to better understand how social capital affects economic development outcomes, followed by a section on international and regional effects.

Individuals and Context
There is increasing evidence of an empirical relationship between social capital and economic development (Engbers, Rubin, & Aubuchon, 2017;Helliwell & Putnam, 1995;Rupasingha et al., 2002;Oh et al., 2014;Woodhouse, 2006). However, the research presented above is limited in a number of ways. First, the international and regional studies usually do not acknowledge that social capital exists in a context. Consequently, to obtain a fuller understanding of social capital's effects on household income, it is useful to study individuals within a meaningful context. Studies of nations or metropolitan areas may underrepresent effects if the effect size is masked by the unit of analysis. Studies such as the one presented here allow analysis of specific types of social capital relationships. Conversely, the qualitative studies presented above highlight the importance of context in the economic effects of social capital but rarely use a methodology appropriate for demonstrating clear association. This study attempts to rectify these shortcoming by applying quantitative techniques at the neighborhood level in order to better understand the effects of social capital on household income.

International and Regional Studies
One of the earliest studies to establish a link between social capital and economic development came from Putnam, Leonardi, and Nanetti (1994). In their qualitative analysis of Italian regions, they argue that Northern Italy has surpassed the South because of its unique stock of social capital. Unlike the South, where there is an established history of criminal and governmental corruption, Northern Italy exhibits a high degree of grassroots accountability. A tradition of involvement in civil society organizations and healthy governance in the North led to economic and political systems that overcame a history of authoritarianism to lead to more rapid economic growth (Putnam et al., 1994).
The findings from Putnam et al. (1994) since have been supported by statistical analysis. One early econometric approach found that trust and civic cooperation were a major predictor of economic performance and that these social capital effects were particularly prominent in countries with effective government, legal, and political systems. Interestingly, the same study found no economic effect associated with membership in civic organizations (Knack & Keefer, 1997). These results are at odds with those from a study by Dakhli & De Clercq (2004) using a similar survey methodology. This study suggests positive effects of trust and associational membership -but not civic behavior on cross-national innovation. This is further supported by a study that examined international development between 1970 and 1992 in 34 countries and found that social capital as measured by levels of trust has a significant impact on endogenous growth (Whiteley, 2000).
In addition to Safford's regional U.S. comparative study of bridging social capital, as discussed earlier, a regional study of two Australian towns facing similar economic challenges finds important economic effects (Woodhouse, 2006). Using a mix of qualitative and quantitative techniques and considering a wide range of social capital measures, such as informal associations (generalized reciprocity and bonding social capital), bridging social capital, family and work social capital, community engagement, and thin trust, this study finds that the greatest impact on economic development comes from bonding social capital, which results in higher employment rates and a larger proportion of high income households (Woodhouse, 2006).
The quantitative research is equally inconclusive about the type of social capital that affects economic development. One of the first large-scale quantitative studies of social capital and economic outcomes found that bonding social capital, as measured by the number of civic institutions per 10,000 individuals, was statistically and substantively related to growth in per capita income (Rupasingha et al., 2002). Conversely, a recent study of social capital in the 50 largest municipalities in America suggests that bonding social capital has no effect on per capita income at the metropolitan level. However, they do find that social capital has an effect second only to education in predicting job creation. This is most prominent for their measures of bridging social capital (Engbers, Rubin, & Aubuchon, 2017). This suggests that the loose networks associated with bridging social capital foster community-level economic development by reducing transaction costs and by increasing the multiplier effects associated with economic development (Engbers & Rubin, 2016).
The value of bridging social capital for municipal economic development is further supported by evidence that shows that the number of nonprofits but not the volume of volunteering matters. In other words, volunteering intensively for one organization (a measure of bonding social capital) has no statistical effect on a community's per capita income or job creation. However, the presence of nonprofits within a community as a gathering place for cross-cutting interaction creates positive economic effects (Engbers, Rubin, & Aubuchon, 2017). Likewise, Oh et al. (2014) show that social capital in the form of intra-local and inter-local economic development partnerships results in more joint ventures and public/private partnerships and the resulting economic development.

Individual
As with social capital studies at other units of aggregation, the earliest work on individuals was qualitative in nature. This study found that being a member of a tight-knit community such as that found when one lives in an ethnic enclave leads to access to economic and other entrepreneurial resources. Consequently, bonding social capital provides access to resources that might be unavailable through traditional financial systems (Light, Kwoun, & Zhong, 1990). Likewise, a qualitative study of job success found that individuals with high levels of social networks are able to find better jobs and are of more appreciable value to their employers (Erickson, 2001).
The quantitative research that followed has tended to focus on the benefits of bridging social capital that accrue to individuals. Much of this research focuses on the benefits of social capital to entrepreneurial activity, including the fact that those with larger social networks are much more likely to engage in entrepreneurial activity (Kim & Aldrich, 2005) and that the size of an entrepreneur's social network is influential in their financial success (Baron & Markman, 2003). These effects are both direct through the skills associated with social capital (e.g., emotional intelligence, effective communication ability) and indirect through the social network and the benefits it provides in terms of access to resources and customers.

Methods
This study attempts to learn more about how neighborhood and individual characteristics influence economic development and social capital. Past research only examines social capital at the city level or higher, but little is known at the neighborhood level. Methodologically, this study replicates Safford's (2009) findings using an alternative methodology. Unlike Safford's (2009) study, which effectively utilizes case studies and social network analysis, the methodology of this paper uses survey analysis of three Evansville neighborhoods with different socioeconomic status and policy interventions in order to better understand how social capital can serve as a tool for economic development and an explanation for why some communities lag in jobs and growth. This study examines the effect of social capital across neighborhoods in Evansville and shows that those differences are explained by both characteristics of the neighborhood and individual. This study shows the relationship of social capital in multiple forms (e.g., bonding social capital, trust, neighborhood involvement) across neighborhoods in its association with household income.
It is not the intent of the authors to suggest that this method is superior to past qualitative studies but rather to triangulate past qualitative studies with quantitative data, thus presenting an alternative methodological approach (one that mixes quantitative data with analysis of micro regions), which enriches our understanding of the relationship between social capital and economic development.

Study Context
Evansville, Indiana, is a city located on the banks of the Ohio River in the southern part of the state of Indiana. Evansville is the third largest city in Indiana. Evansville is an industrialized city that has experienced population decline and job loss throughout the past 20 years. A study of a mid-size city like Evansville provides a number of benefits. A lot of studies have been published on larger cities, but much can be learned from studies of medium-size and smaller cities that share similar neighborhood dynamics and have faced similar economic challenges (Gilderbloom, Hanka, & Ambrosius, 2012;Hanka, Ambrosius, Gilderbloom, & Wresinski, 2015) Within the City of Evansville, the sample consists of residents within the boundaries of the Dexter, Glenwood, and Mt. Auburn neighborhoods. These neighborhoods were chosen because of their diversity in terms of affluence, city investment, and policy intervention. The boundaries of the cities can be found in figures 1, 2, and 3. There are 177 households in the Dexter neighborhood, 276 households in the Glenwood neighborhood, and 51 households in the Mt. Auburn neighborhood.

Data Collection
Safford's (2009) study of Allentown, Pennsylvania, and Youngstown, Ohio, argues that some rust belt communities have fared better than others. Using social network analysis, he finds that Youngstown failed to rebound from economic decline in the 1980s because the social capital dynamics of the community limited access to resources and expertise necessary for recovery. Safford's study is relevant to this study in understanding the economic effects of social capital on the neighborhood and individual level. For this study, data were collected through a 34-question survey instrument that looked at various social, economic, and demographic indicators. A combination of methods was utilized to increase the response rate. First, a trained graduate student employee from the Master of Public Administration (MPA) at the University of Southern Indiana, a high school intern interested in social research, and two faculty supervisors went doorto-door in the three neighborhoods (Dexter, Glenwood, and Mt. Auburn) from Monday-Saturday in the spring and summer during various times of the day (in the morning between 10 a.m. to 12 noon, early to mid-afternoon and in the evening after 6:30 p.m.) and asked the head of household to complete the survey instrument. This step was designed to increase the number and quality of survey data, and it enabled the development of a definitive list of addresses for the second stage of the study.
During the second stage, residents who were unable to be contacted by door-to-door canvassing were sent a mail survey. As is best practice in survey research to increase response rates, households were sent a postcard priming the respondent to compete the study. One week later, the survey was mailed with a self-addressed stamped envelope (Dillman, Smyth, & Christian, 2009).
The sample contained 504 usable addresses in the three neighborhoods (a number of them were vacant). Completed surveys were received from 121 respondents from the three Evansville neighborhoods (57 in Dexter, 33 in Mt. Auburn, and 31 in Glenwood); 37% of the completed surveys from Dexter, 36% of the completed surveys from Mt. Auburn, and 24% of the completed surveys from Glenwood were completed going door-to-door; 63% of the completed surveys from Dexter, 64% of the completed survey from Mt. Auburn and 76% of the completed surveys from Glenwood were completed by the respondent sending the completed survey back to us by mail. Our survey response rate was 24%. Response rates differed by neighborhood with the most disadvantaged neighborhood, Glenwood, exhibiting an 11% response rate. The middle-income neighborhood, Mt Auburn, had a response rate of 65%, while the more affluent Dexter neighborhood had a response rate of 32%.

Variables
The unit of analysis is the individual, but the model includes neighborhood-level variables to account for contextual effects on social capital. The study emphasizes the neighborhood level because past quantitative studies of the economic effects of social capital have been limited to higher levels of aggregation. In addition to traditional individual-level control variables often used in neighborhood studies, such as homeownership, median income, hours worked, education level, and employment status, the study uses a variety of measurements common in the social capital literature that capture neighborhood-level effects on social capital, such as frequency of socialization, a four-question neighborhood involvement index, a four-question neighborhood attachment index, and the number of neighbors known (Engbers, Thompson, & Slaper, 2017).
An additional advantage of this study is that it controls for macro-economic causes of income because all the participants are spatially concentrated within the same labor market. This study seek to predict household income with bridging social capital variables, such as how many neighbors do you know by first name, how many neighbors do you know by face, number of different organizations respondent volunteered for in the past 12 months, socialization (that reflect a wider range of connections) measured as an average of the following four items: had friends over to your home, been in the home of a friend of a different race or had them in your home, been in the home of someone of a different neighborhood, been in the home of someone you consider to be a community leader or had one in your home. Participation is another variable measured as an average of the seven kinds of participation from this survey: how many times the respondent worked on a community project, donated blood, attended any public meeting in which there was a discussion of local or school affairs, attended a political meeting or rally, attended any club or organizational meeting, given $25 or more to charity, or volunteered.
Our bonding social capital variables are bonding social capital measured as a ratio of hours volunteered per week divided by the number of organizations volunteer over the last 12 months. Generalized trust is measured using the standard general social survey measure of trust. Neighborhood attachment is measured as a mean based on a four-question index ranging from "0" (not answering the question at all) to "4" (answered positively on every attachment question). Neighborhood involvement is measured a mean based on a four-question index ranging from "0" (not answering at all) to "3" (answered positively on every involvement question).
We include other social capital controls, such as the average number of volunteer hours per week; population density measured in population per square mile within the neighborhood's boundaries, and a number of economic controls, such as average hours per week worked; college graduation rate measured in percent of the population with a college degree, median neighborhood income measured in thousands of dollars, and a dummy variable specifying male.

Method of Analysis
To better understand the effects of social capital on household income, a generalized logistic regression model using maximum likelihood estimation techniques was used to estimate the relationship between an ordinal dependent variable and a set of independent variables. For this logistic regression, the dependent variable is a four-point Likert measure of income: 1) less than $25,000 per year; 2) $25,000-$50,000 per year; 3) $50,001-$75,000 per year; and 4) more than $75,000 per year. While the model nests individuals within their neighborhood context, the limited number of neighborhoods leads one away for a hierarchical model. Rather, the model was run using robust standard errors clustered around each of the three neighborhoods. STATA was used to perform the regression using the following specification 1 for the final analysis: y1* = β1*College grad + β2*Frequency of socialization + β3* Different organizations volunteered in past 12 months + β4* Average Number of Hours Volunteered per week + β5*employment + β6* Know neighbors by face + β7*Currently enrolled in school + β8*Neighborhood attachment + β9*Trust + β10*Male + β11*Bonding social capital + β12*Median neighborhood income +εi

Results
The results presented below highlight the important contextual differences in which social capital operates.  Auburn's (18.3%), which is slightly higher than Glenwood's (20.9%). Table 2 provides descriptive statistics for the interval variables, including socialization, average hours volunteered per week, know respondent by face, neighborhood attachment, bonding social capital, and median income. Table 3 provides the frequencies of the variables college graduate, number of different organizations volunteered, employment status, enrolled in school, trust, gender, and household income.  Table 4 presents the results from the generalized logistic regression model predicting household income with the social capital variables but without any controls for human capital or neighborhood characteristics. Approximately 87% of the variation in household income is explained by the included variables in this model. All of the predictors are significant for household income less than $25,000. Number of different organizations volunteered, trust, and bonding social capital are negative, while socialization, average number of hours volunteered per week, know person by face, and neighborhood attachment are positive. The magnitude of the coefficients for these variables is largely determinative with bonding social capital, volunteering for different organizations and trust as having an odds ratio of near 0. The odds ratio for the other coefficients is so high that they have an overwhelming likelihood of falling into a higher income category. For household income between $25,000-$50,000, only socialization is significant at the 0.05 level and is in the negative direction. This translates into an odds ratio of 0.51 of falling into a higher income bracket. When predicting higher household income ($50,001-$75,000), socialization is negative and not significant. Number of different organizations volunteered is positive and significant at the 0.05 level, but trust is negative and significant at the 0.10 level. At this income category ($50,001-$75,000), there is a more moderate effect of the social capital variables with odds ratios ranging from 0.12 for trust to 1.69 for bonding social capital. However, these results should be considered with caution, as the introduction of control variables influences these findings.
Our generalized logistic regression model in table 5 includes the variables from table 4 plus college attainment, employment status, school enrollment, a dummy variable specifying male, and median income of the respondent's neighborhood. The pseudo R 2 of this model is approximately 53%. For household income less than $25,000, number of individuals known by face is not significant, and college grad is significant at the 0.10 level. 2 All other variables are statistically significant with odds ratios ranging from near zero for socialization, number of different organizations volunteered, employment status, neighborhood attachment and bonding social capital to a largely determinative odds ratio for college graduation, average number of hours volunteered per week, trust, male and enrollment in school. The odds ratio for neighborhood median income is 1.02.
Consequently, men, students, and college graduates are more likely to be in a higher income household than women, non-students, and non-college graduate respondents. Likewise, as one increases his or her average number of hours volunteered per week and level of trust, that person's likelihood of being in the next higher income category increases dramatically.
For household income between $25,000-$50,000, all predictors are significant at the 0.001 level.
The number of different organizations volunteered, neighborhood attachment, and bonding social capital are positive predictors, while the other coefficients are negative. The odds ratios range from near zero for college graduation, socialization, average number of hours volunteered per week, employment status, knowing neighbors by face, enrollment of school, trust and male to almost deterministic odds ratios for number of different organizations volunteered, level of trust, and level of bonding social capital. Similar to the lower income category, being male and enrolled in school (though not a college graduate) increases the likelihood of being in a higher income category. Likewise, increasing levels of trust increase the likelihood of being in a higher income bracket. However, unlike those in the under $25,000 bracket, having a higher level of bonding social capital and participating in a greater number of organizations is associated with a dramatic increase in the likelihood of being in a higher income category.
For household income between $50,001-$75,000, being a college graduate and having a higher degree of trust are no longer significant. The odds ratio for neighborhood attachment and neighborhood income are near zero, but enrollment in school, being male and having bonding social capital are quite high. The odds ratios for the other variables are much more moderate: socialization (2.03), number of different organizations volunteered (8.85), average number of hours volunteered per week (0.92), employment status (83.1), and knowing neighbors by face (1.79). This suggests that men, those enrolled in school, and those with a high degree of bonding social capital are much more likely to be in a higher income category. For those individuals with incomes between $50,000 and $75,000, increasing socialization, the number of different organizations volunteered, the average number of hours volunteered per week and the number of neighbors known, will experience an increased likelihood of being in the highest income category. Additionally, those who are employed would also experience a greater likelihood of being in the next income category.

Discussion and Limitations
These results points to three contributions of this study: 1) a surprising finding about differences across income categories; 2) a nuanced understanding of the relative value of bridging and bonding social capital for economic development; and, 3) the importance of neighborhood context in understanding the effect of social capital.
First, the generalized logistical model enables us to see that the impact of social capital is contingent upon the income category of interest. While being male or a student significantly increases the likelihood of being in a higher income for all economic strata, the social capital effects of the other variables of interest depend on the larger economic situation. For those individuals in the lowest income category (household incomes less than $25,000 a year), the range of social capital indicators is more limited than for higher income categories. Only level of trust and average number of hours volunteered per week have a substantive effect. The average number of hours volunteered per week is only statistically and substantively significant for the lowest income category suggesting either a bonding effect of social capital or perhaps more likely a skill-building effect of volunteering where increased volunteering either signals marketable skills or workforce readiness. This is in contrast to those in the $50,000-$75,000 income category, whose probability of being in a higher income category is influenced by a wider range of social capital indicators, including bonding social capital, socialization, the number of different organizations volunteered, and knowing neighbors by face. That said, the odds ratio for these are lower. This suggests that, while a wider range of social capital is influential for higher income individuals, the odds of increasing income categories associated with this social capital is more tempered. For those in the $25,000-$50,000 income category, a moderate number of social capital variables including trust, bonding social capital, and the number of organizations are associated with a change in income status and the change in likelihood is fairly dramatic. Consequently, studies of social capital and economic development that do not differentiate between income categories may mask or amplify the effect of social capital on income.
Second, until recent years, the preponderance of emphasis on developing social capital has been on building bonding social capital as represented by the emphasis on involvement in voluntary organizations found in the work of Putnam and others. However, recently, scholars have tended to focus on the economic benefits of bridging social capital (Engbers, Rubin, & Aubuchon, 2017;  Safford, 2009). This study finds a role for bridging and bonding social capital. The direct measure of bonding social capital has a significant effect in increasing the odds of membership in higher income categories for the two middle income strata investigated in this study. Likewise, trust, which is often classified as a form of bonding social capital, appears to be associated with increasing income for the two lowest income categories. The focus on trust supports one of the traditionally conceived causal mechanisms of social capital: the reduction of transaction costs. By building greater networks of trust, there is less need for formal, and often costly, transaction enforcement mechanisms. This increases the net economic benefit of any relationships. For medium and high-income categories, the effect of bridging social capital as measured by the number of different organizations volunteered and knowing neighbors by face appears to have a greater effect. Consequently, one should be hesitant to espouse blanket support for either bonding or bridging social capital but rather to consider the moderating effect of economic and social network in judging the effect of social capital. This is in many ways consistent with both the weak and strong ties argument (Granovetter, 1973). Furthermore, unlike past research (Engbers, Rubin, & Aubuchon, 2017), which found bridging social capital's effect on job creation but not income, this study finds an effect on income, thus stressing the importance of unit of analysis in social capital research. Aggregation at the MSA level may mask the effect on personal income that can be found when studies examine individual and neighborhood effects.  5 showing that a failure to control for neighborhood-related effects masks the potential impact of social capital. This important finding suggests that, while social capital can be used for economic gain regardless of the neighborhood context, 3 the effect needs to be understood within a cultural context. Thus, groups such as Habitat for Humanity that attempt to build social capital as a means of revitalization are not misdirected in their efforts but must take a more nuanced approach to thinking about social capital's effects.
The consistency with past research and the high levels of statistical significance in light of moderate sample sizes bolster the argument presented above, but some limitations remain. Perhaps the biggest concern includes issues stemming from the lower response rates.
First, the response rate differed among neighborhoods with low income neighborhoods exhibiting lower response rates. This raises the risk of selection bias. This bias is undoubtedly a problem. As seen in table 6, the sample collected from this survey differs from census estimates in significant ways, including more education and more white. Likewise, the low response rate may introduce other forms of bias such as the likelihood that survey respondents differ from non-respondents regarding other unmeasured characteristics such as being more extroverted or having other soft skills.
A second area of concern is perfect multicollinearity of control variables, which limits the number of neighborhood controls included. The model above included median income as a control of neighborhood-level dynamics. 4 To better understand the effects of context, a wider range of neighborhood controls is needed. To do this would require increasing the number of neighborhoods examined, which should be pursued by future scholars.
Third, the use of household income creates some limitations in interpretation of the data. Because the respondent may not be the primary income earner for the household, he or she may be unable to effectively judge household income. Alternatively, and perhaps more of a concern, the connection between social capital and household income is more tenuous. The respondent may have a high degree of social capital, but this capital is unlinked to household income.
Despite these conclusions, this study aids our understanding of social capital and economic development by showing there are personal income benefits associated with social capital and that these benefits accrue to those with bridging and bonding social capital. However, this effect is depending on economic context and current economic status. Moreover, social capital effects appear to exist across neighborhood and that while neighborhood context is important for understanding income, the neighborhood effect does appear to nullify the effect of social capital.

Keywords: Honesty, Public Administration Ethics, Public Service Values
Public administrators must hold themselves to a high standard of honesty. In this vein, Bok (1999) argues for a "principle of veracity" that respects human dignity and autonomy by emphasizing the centrality of honesty and truthfulness as the foundation of human relations. She points out that our choices, if they are to be free and rational, depend on our judgments about what is true and what is false. Because these judgments often rely on information that administrators obtain from others, including public officials, dishonesty has the effect of distorting the information on which they are based, and, consequently, our choices as well. Therefore, honest communication between individuals is essential for maintaining a society that recognizes human dignity and supports the ability of its members to make free and rational choices. Bok (1999) argues that this principle of veracity does not necessarily mean that all lies should be definitively ruled out. However, it does mean that, for dishonesty to be morally justified, administrators must first seek out truthful alternatives, and where none are available, it must clearly provide greater benefit than harm. Even in such rare instances, however, it should be recognized that dishonesty always creates at least some harm. This is because it imposes moral costs on the deceivers themselves and damages the credibility of administrators more broadly when revealed (della Porta & Vannucci, 2012).
There are many cases of lying for personal or agency benefit. In a specific case, a police officer in Sandusky, Ohio, was fired for fabricating reports of disabled vehicle complaints in order to avoid repercussions for arriving to work late. In his termination letter, the Erie County Sheriff wrote about the officer's pattern of excessive tardiness and other violations but stressed, "However, the termination of your employment is not for these violations, but for your dishonest actions" (Astolfi, 2014, p. A7). The sheriff also added that "law enforcement officers are, and should be, held to a high standard of truthfulness" (Astolfi, 2014, p. A7).
In a high-profile case, public school employees in Atlanta were convicted of racketeering charges under the Racketeer Influenced and Corrupt Organizations Act (RICO). Administrators, principals, and teachers were convicted for their role in a cheating conspiracy that involved changing answers on test scores to inflate the performance of the students (McWhirther, 2015).
Though not centrally about lying, this case illustrates how pressures to meet performance metrics and other factors associated with organizational life may lead public administrators to engage in unethical deceptive practices (Molina, 2016). Ariely (2012) cautions that even minor instances of deception or dishonesty can quickly become a slippery slope leading to more serious instances. In effect, once someone has been dishonest about something, it becomes easier to be dishonest in subsequent situations. According to Ariely (2012), because these initial acts create a sort of "what-the-hell" effect, "a single act of dishonesty can change a person's behavior from that point onward" (p. 135). Therefore, "the first dishonest act is the most important one to prevent" (p. 137). Furthermore, acts of dishonesty can assume a contagious nature that induces others to act dishonestly as well. According to Ariely, As long as we see other members of our own social groups behaving in ways that are outside the acceptable range, it's likely that we too will recalibrate our own moral compass and adopt their behavior as a model for our own. And if the member of our in-group happens to be an authority figure -a parent, boss, teacher, or someone else we respect -chances are even higher that we will be dragged along (p. 207).
There are circumstances, however, in which not telling the truth may be viewed as justifiable. The U.S. Supreme Court has ruled in several cases that deception can be used by law enforcement as long as it is unlikely that it will provoke a person to confess to a crime that he or she did not commit or commit a crime that that person would otherwise not have committed. In such cases, the court has ruled that the use of deception is a necessary tool in law enforcement.
An example of such a ruling is U.S. v. Russell, a case that involved undercover law enforcement gaining the confidence of suspects involved in a drug manufacturing operation, and offering them ingredients for manufacturing methamphetamine. Though the defense argued that they were victims of entrapment, the court ruled that the suspects were predisposed, and not pressured by the undercover officers, to commit the crime. Therefore, according to the court, it was not a case of entrapment, and the use of deception was legitimate (Rutledge, 2007).
The U.S. Supreme Court has also ruled that it is acceptable for law enforcement to pretend to have evidence against a suspect, as long as the deception being used is plausible given what is known about the crime and the suspect. For example, in Oregon v. Mathiason a suspect confessed to a crime after being deceived into thinking that officers had found his fingerprints at the scene of the crime. The court ruled that the confession was valid, even though it was obtained through deceptive practices (Rutledge, 2007). That which is legal, however, is not necessarily ethical. Therefore, in using deceptive or dishonest practices, public administrators must look beyond what is lawfully permissible to broader considerations about what can be morally justified.
Fortunately, research has shown that public employees consider honesty to be the among the most important values coming to bear on administrative practice (Molina & McKeown, 2012). On the other hand, it also has been argued that, once taken out of the abstract and faced with the difficulties of concrete situations, commitment to honesty among public administrators can deteriorate (De Vries, 2002). This exploratory research study aims to provide a better understanding of the conditions in which dishonesty and deception (following Bok [1999], the terms are used here synonymously) is an acceptable practice for public administrators. To that end, the views on honesty and deception of public employees working in or retired from police departments, fire departments, or public schools are compared and contrasted to determine when and why deception is considered acceptable in practice. Along these lines, this study employs an analytical framework based on the American Society for Public Administration's (ASPA) Code of Ethics as a way of assessing the justifications for dishonesty and deception described by participants. This article concludes with a discussion of some important implications that these findings have for how ethics training is provided to public officials, how ethics is taught in MPA programs, how public administrators communicate with the public, and how public administrators can better ensure the integrity of our public institutions.

Philosophical Approaches to Honesty and Deception
As Svara (2015) has pointed out, ethical choices concerning right and wrong have traditionally been made through one of three philosophical approaches. These approaches include virtue (character-based), deontological (principle-based), and utilitarian (consequence-based) reasoning. The deontological approach is chiefly associated with the eighteenth-century German philosopher Immanuel Kant (1988) who argues that people should act only in accordance with those maxims that they could rationally will to be a universal rule for all to follow. In the deontological view, lying is never permissible, regardless of the consequences (Svara, 2015).
In contrast, the utilitarian approach associated with philosophers such as Jeremy Bentham (2001) and J. S. Mill (1993) argue that an action is judged to be acceptable depending on the consequences of the action, i.e., it is morally right to promote the greatest good for the greatest number. Thus, from a utilitarian perspective, a lie is justified if the good that comes from it outweighs the negative consequences. Virtue-based ethics, associated primarily with Aristotle (2004), maintains that specific qualities of personal character determine what makes for a "good person." In this tradition, ethical decision-making consists of acting in accordance with virtues, i.e., dispositions toward certain attitudes and behaviors that are essential to living a life of excellence.
Whichever approach is adopted, however, public administrators may ultimately need to draw upon their own moral compass to resolve ethical problems (Cooper, 2012). As Stewart (1984) puts it, "most managers are neither pure deontologists, nor pure Utilitarians, but rather operate according to a kind of ethical pluralism" (p.20). Along these lines, Svara (2015) has proposed the "ethics triangle" framework in which administrators give due consideration to the principles, consequences, and character of their actions while maintaining a focus on their duties as public officials.
On the question of what type of circumstances might justify deception, Bok (1999) argues for a "principle of publicity," which states that a lie is morally justifiable only if it can be defended before a community of reasonable persons. This raises an interesting question regarding who should make up such a community of reasonable persons in a democratic society. Tavaglione and Hurst (2012), for example, argue that, in certain instances, physicians may have a moral obligation to lie for their patients; particularly in "gaming" the system in order to acquire medically necessary treatment for patients who would otherwise be denied by third-party payers. Drawing on MacIntyre's (1984) theory of virtue ethics, they argue that "physicians should obey the internal morality of medical practice, that the cornerstone of this internal morality is the duty of beneficence, and that such a duty, in certain cost-containment circumstances, implies the violation of the reimbursement rules" (Tavaglione & Hurst, 2012, p. 5).
The "principle of publicity" and circumstances where there is a moral obligation to lie contribute to the conditions that allow for the "just lie" theory. Pasquerella and Killilea (2005) argue that the "just lie" theory can be used to achieve middle-ground results that strike a balance between the philosophical approaches to ethics. Rather than provide an opportunity for self-serving excuses, the "just lie" theory allows public administrators to make ethical decisions in an imperfect world by thinking in terms of justifications for deception that would be accepted by a reasonable community (Pasquerella & Killilea, 2005). This suggests that public administrators need to think about the ethical dimensions of honesty and deception from the standpoint of their professional obligations as public servants in a democratic community.

Honesty and Deception in Public Administration
The Oxford English Dictionary defines the term "honest" as "free of deceit; truthful and sincere; morally correct and virtuous." In turn, the terms "dishonesty" or "deception" may be understood simply as the opposite of honesty. To put it in more precise terms, however, these terms involve statements or actions designed "to intentionally cause another person to have or continue to have a false belief that is known or truly believed to be false by bringing about evidence on the basis of which the person has or continues to have the false belief" (Mahon 2007, p. 189-190). Importantly, this study seeks to distinguish between cases involving the legitimate use of dishonesty or deception as appropriate and justified in the context of public administration, and cases involving the illegitimate use of dishonesty or deception for self-serving, inappropriate, and/or nefarious purposes.
This study utilizes the American Society for Public Administration's (ASPA) Code of Ethics as an analytical framework for distinguishing between cases in which deception and dishonesty are ethically justified on the part of public administrators and cases in which it is not. Using the ASPA Code provides for a broadly accepted framework of ethics in public service but also for a more adequate definition of honesty as a public service value. Recently revised in 2013, the Code is comprised of eight core principles that serve as a statement of the ideals and aspirations intended to guide public administrators in carrying out their duties. In addition to these principles, a set of related practices relevant to each of the eight principles provide a set of more concrete guidelines for ethical administrative conduct.
A number of the principles and practices in the Code relate directly to issues involving honesty and deception in public administration. The most obvious of these is Principle 6, which calls upon public administrators to "demonstrate personal integrity" in order to "inspire public confidence and trust in public service." Relatedly, Practice 6b advises, "Maintain truthfulness and honesty and do not compromise them for advancement, honor, or personal gain." Similarly, Principle 5 requires public administrators to "fully inform and advise" key stakeholders by providing "accurate, honest, comprehensive, and timely information and advice." In addition, Principle 3 stresses the importance of democratic participation by advising public administrators to "inform the public and encourage active engagement in governance. Be open, transparent and responsive, and respect and assist all persons in their dealings with public organizations." The full list of ethical principles are as follows: 1 1. Advance the Public Interest. Promote the interests of the public and put service to the public above service to oneself. 2. Uphold the Constitution and the Law. Respect and support government constitutions and laws, while seeking to improve laws and policies to promote the public good. 3. Promote Democratic Participation. Inform the public and encourage active engagement in governance. Be open, transparent, and responsive and respect and assist all persons in their dealings with public organizations.

Strengthen Social
Equity. Treat all persons with fairness, justice, and equality and respect individual differences, rights, and freedoms. Promote affirmative action and other initiatives to reduce unfairness, injustice, and inequality in society. 5. Fully Inform and Advise. Provide accurate, honest, comprehensive, and timely information and advice to elected and appointed officials and governing board members, and to staff members in your organization. 6. Demonstrate Personal Integrity. Adhere to the highest standards of conduct to inspire public confidence and trust in public service. 7. Promote Ethical Organizations. Strive to attain the highest standards of ethics, stewardship, and public service in organizations that serve the public. 8. Advance Professional Excellence. Strengthen personal capabilities to act competently and ethically and encourage the professional development of others.
Crucially, a complete understanding of what honesty means in the context of public administration requires consideration of the full range of values, principles, and practices that are relevant to ethical public service. According to ASPA, The Code and these practices are intended to be used as a whole and in conjunction with one another. An ethical public servant will consider the full range of standards and values that are relevant to handling a specific matter and be committed to upholding both the spirit and the letter of this code (2013, p. 1).
Of course, this is easier said than done because, in some cases at least, ethical principles and practices can come into tension with each other. For example, the Code holds that public administrators must be "open and transparent while protecting privacy rights and security" (Practice 3a). This suggests that honesty in the context of public administration cannot entail disclosing everything one knows in every situation. In some circumstances, non-deceptive responses such as "I cannot comment" or "we do not reveal that information" are appropriate. As Bok (1999) notes, however, communicating false or misleading information is only justified when other alternatives are unavailable, and where the benefit clearly outweighs the harm that might result. In other cases, acting in an honest manner may come into conflict with organizational expectations and interpersonal relations, such as collegiality or obedience to superiors, which are not ethical principles or practices. Public administrators must be prepared to ignore these pressures. As the Code points out, public administrators must be "prepared to provide information and recommendations that may not be popular or preferred by superiors or colleagues" (Practice 5b); they have the responsibility to support accountability "with clear reporting of activities and accomplishments" (Practice 7a) and "seek to correct instances of wrongdoing or report them to superiors" (Practice 7d). Leaders also should seek to reduce possible conflicts by encouraging the "open expression of views by staff members within the organization" by providing "administrative channels for dissent" and protecting the "whistleblowing rights of public employees" (Practice 7c).
In order to negotiate such tensions, administrators need to demonstrate ethical competence (Menzel, 2016). Ethical competence involves being able to assess competing values, understand principled moral reasoning, and recognize when they are confronted with an ethics-related conflict (Bowman, West, Berman, & Van Wart, 2004). Further, the ethically competent administrator must understand the public interest and have the skills needed to allow the public  (2) Fire Chief (1) Teacher (7) Firefighter (1) Coordinator of Advanced Academic Studies (1) Police Chief (3) Speech Pathologist (1) Police Officer (2) Detective (1) Note: 16 participants were men; eight participants were women; 17 participants were 40 years of age or older; seven participants were under the age of 40; all participants were white.
interest to prevail while at the same time respecting the law and their elected bosses (Menzel, 2011). As a way of increasing our understanding of how public administrators balance these considerations, this study aims to answer the following research questions: 1) Why is honesty important for public administrators? 2) What is honesty and dishonesty (deception)? 3) Under what circumstances is the use of deception by public administrators legitimate?

Methodology
To answer the foregoing research questions, semi-structured interviews were conducted with public employees working in or retired from careers in police departments, fire departments, or public schools. Public safety and education were the chosen professions for this analysis for two major reasons. First, the focus on specific public sector professions allowed for data saturation. This allowed for fair comparisons to be made between interviewees. Second, these professions were chosen because the individuals working in them perform different public services. Administrators in education work mainly with children and parents, while administrators in public safety deal with dangerous situations more regularly, many of which involve criminals. Some of the interviewees were street-level bureaucrats, such as teachers and police officers. These perspectives were important to this analysis because these individuals work directly with the public.
The initial recruits were selected through a convenience sample consisting of public employees with an established relationship with one of the researchers because of the potentially sensitive topic of this research. The trust previously established between the interviewer and interviewee made the discussion about deception more likely to be an honest and open discussion. Snowball sampling was used to contact other potential participants for interviews. This is a common sampling method used in qualitative research (Weiss, 1994). At the conclusion of each interview, participants were asked if they knew of anyone else who may be willing to participate. The suggested public employees were then contacted with the same recruitment letter as the original recruits. All participants were public employees in the state of Ohio. Table 1 shows a breakdown of participant occupations.
A total of 59 public employees were contacted, and 24 chose to participate. Of the 24 participants, 15 were employed in education, and nine were employed in public safety occupations. On average, interviews lasted a total of approximately 24 minutes. Interview responses were confidential, and no identifiers appear in this research. Public employees were contacted and interviews were conducted until a point of data saturation was reached. Initial recruits were contacted through a letter inviting them to participate.
The 24 public employees who chose to participate were asked several questions related to honesty and dishonesty within their organizations. First, they were asked to identify their job title and provide a brief description of their occupational role. Second, they were asked about the importance or unimportance of honesty in their work. Third, they were asked to tell at least one story about a time when they felt it was necessary to be dishonest or deceptive in their job. Finally, they were asked to discuss why they believed being dishonest or deceptive was necessary in their case. Stories from public employees can offer narratives that help to expand our knowledge of how meanings and norms are expressed (Hummel, 1991). Pertinent to this research, stories can reveal the moral reasoning of an individual by highlighting his or her understanding of the importance of honesty in public service and the circumstances in which that person considers deception justified (Maynard-Moody & Musheno, 2003). Participants were not provided definitions of honesty or deception prior to or during the interview. Instead, they were given the freedom to conceptualize and define the terms as they understood them. The complete list of interview questions is available in the appendix.
Recordings of the first 15 interviews were transcribed. To be sure that the presence of a recording device was not influencing the responses of participants, the final nine interviews were not recorded but instead were conducted using field notes. Qualitative research software (NVivo 10) was used to analyze the data provided in transcripts and field notes. Themes within the data were then compared and contrasted across occupational settings. An open-coding system was used to analyze the data. Open-coding is a data-driven coding system where codes are created from the text itself. This allows for constant comparisons to be made within the data while limiting the influence of preconceived notions (Gibbs, 2007).

Honesty in Public Administration
The first aim of this study was to explore the reasons public administrators give for why honesty is important in their work. Along these lines, a number of key themes emerged. First, many of the public administrators interviewed for this study equated honesty with integrity, claiming that the two were inseparable. It was perceived that, without honesty, it would be impossible to have strong moral principles and therefore impossible to have integrity. One former school principal began the conversation about honesty by stating, "I can't zero in on honesty without talking about integrity and I believe, I know, I have always concerned myself with my personal integrity, and therefore I try to be as honest as I can" (personal communication, December 23, 2014).
Building respect, trust, and credibility with key stakeholders was a common concern among the public administrators interviewed, and honesty was seen as essential in this regard. Furthermore, participants viewed this as important for doing their jobs effectively. Interviewees working in education, for example, stressed the importance of parents, co-workers, superiors, and students respecting them, trusting them, and believing them to be credible. In this sense, honesty was viewed as an important tool for maintaining good relations with key stakeholders. Interviewees working in safety forces had a similar view. Though not working with the exact same groups of people, the value of maintaining positive relations by establishing respect, trust, and credibility was a common theme. As one police officer put it: They respect me because I tell them straight up. I don't try to trick them into anything and they like that about me. It's important, especially in a courtroom because if you lie on the stand, the judge will never believe you again. All of your credibility will be gone (police officer, personal communication, December 19, 2014).
Similarly, a high school teacher pointed out: We are talking about high schoolers. We want them to respect us. You want to be honest with them, because they can pick up on that. If they don't think you are being honest, they are going to pick up on it. It is what it comes down to. But if we don't gain their respect they are less likely to learn from us, honestly. If they don't respect you and they don't care about you they are going to walk into your class and not care at all. They are not going to try. You have to give them that honesty. You have to gain their trust (high school teacher, personal communication, December 23, 2014).
Building trust, respect, and credibility was seen by public administrators as necessary for maintaining public support more generally as well. This finding echoes Principle 6 of the ASPA Code, cited above, which calls upon public administrators to "demonstrate personal integrity" in order to "inspire public confidence and trust in public service." Participants acknowledged that the public expects police officers, firefighters, and educators to operate at a high level of honesty, integrity, and transparency. These values were not viewed as simply good in and of themselves. Rather, administrators saw honesty as important for performing the duties of their jobs and maintaining public support -not just symbolic support from the public but financial support, too. To that end, administrators stressed the importance of tax levies to their work. According to one police chief: First, let me explain why I think it's so important to be truthful. Especially in my role, it's important. The entire department again-it's funded by taxpayer money. I want our taxpayers to continue to support us through this levy (police chief, personal communication, January 6, 2015).
Likewise, an assistant principal observed: Everyone has to work hard to give the public a good value for their money. It's important to trust people to do their job and not micromanage. It's important to be honest with coworkers because to be honest to the workers means that hopefully in turn they will be honest with you (assistant principal, personal communication, January 14, 2015).
In sum, public administrators viewed honesty as an essential part of their character and important for building respect, trust, and credibility. They argued that this respect, trust, and credibility contributed to their ability to maintain good relationships with superiors, co-workers, and other key stakeholders. Moreover, it also was seen as being crucial for maintaining public support more generally, which, among other things, was linked to continued public financial support of schools, public safety forces, and other important resources. In short, participants consistently spoke of honesty as a sine qua non of effective public service.
The second aim of this study was to explore the circumstances in which the use of deception is legitimate in public administration. Though each of the participants acknowledged that there could be circumstances that would justify deception, these were generally rare and rooted in particular contextual factors. All the same, a few common themes do emerge in the stories provided by participants. These themes fall into three general categories: 1) withholding information; 2) justifiable dishonesty; and 3) unjustifiable dishonesty.

Withholding Information
The necessity of following lawful orders and legal procedures was commonly cited as a justification for using deception. Interestingly, however, this rationale for deception was usually about the withholding of information and not necessarily about practices that would typically be considered dishonest or deceptive. As noted above, participants were asked to provide at least one story about a time when they felt it was necessary to be dishonest or deceptive in their job. It should be acknowledged that this may have pressured them to provide an example that they did not necessarily consider to be a case of deception or dishonesty. In such cases, complying with requirements to withhold information was understood by participants as a legitimate part of their professional responsibility as public administrators. According to a police chief: It's OK that if, from my stand point, I'm directed by a higher authority, some township trustees, to do whatever it takes not to let information out. I have to not fully tell everything I know based on an inquiry. I feel that I'm following an order and I have to do that. Like I said if we are intentionally misleading somebody for a sinister motive, that's 100% wrong (police chief, personal communication, January 6, 2015).
In similar terms, a school principal explained: As far as staff, the best answer I can give you, an example would be knowing that a layoff was coming. I knew I couldn't tell the person. The one person I'm thinking of actually, he got sick here at school. He had to go by ambulance to the hospital. He lives up in [city name] area. He had no family around, so I drove to the hospital. I was with him and we're like two weeks away from telling him because there are certain timelines we have to do for [the layoff] procedure. While we're sitting there he's telling me how much he likes it, it's so much better than his last job, he's glad he's staying at this job. I didn't tell him 'can't wait for next year and see you next year,' but I sort of deflected the conversation. I told him it was great to have him and things like that but I didn't…I didn't look to the future. I didn't comment on about him being here for the next 10, 15, 20 or 30 years. I really wanted to, but I knew that I couldn't tell him. I wasn't lying to him; I just wasn't talking about the future (school principal, personal communication, December 15, 2014).
These cases of deception by withholding information in order to comply with lawful orders and legal procedures appear to be ethically justified. This, of course, presupposes that the withholding of information is not designed to conceal instances of wrongdoing. Furthermore, it also assumes that the public should not have access to it or should be informed for purposes of health or safety. Such commitments are consistent with Principle 2 of the ASPA Code, which calls for public administrators to uphold the Constitution and the law, and to "respect and safeguard protected and confidential information" (Practice 2d). The example provided by the school principal is also supported by Principal 4, which calls on administrators to strengthen social equity by acting with fairness, justice, and equality by providing "equal treatment, protection, and due process to all persons" (Practice 4b). Had the school principal informed the employee of his termination before the standard operating procedure allowed him to do so, the principal would be doing a disservice to any other employees receiving layoff notices without the benefit of advanced notice.

Justifiable Dishonesty
In the stories provided by participants concerning the use of deception, ethically justifiable cases were comparatively rare. One theme that emerged, however, was the use of deception in order to prevent harm to individuals. In order to protect the life of an informant, for example, it may be necessary for undercover police officers to use deception with drug dealers or gang members who are suspicious of the informant. As one detective put it, there can be instances in which it is "way more ethical to lie than to be honest when we have a confidential informant." According to the detective: If confidential informants are found out then they may be killed by criminals, drug dealers. Drug dealers have said [to me] "It was so and so, wasn't it?" I told a bold face lie "I have no idea who that is" (detective, personal communication, June 11, 2015).
Lying is ethically justified in such cases because simply stating, "We don't reveal the names of informants," would only confirm the suspicion of the drug dealer. Narrow instances such as these, where being dishonest prevents serious harm from coming to an individual, and alternatives are unavailable, would be accepted by a reasonable public (Bok, 1999). Additionally, this use of deception is supported by the ASPA Code because it ensures the protection of individuals (Practice 4b).
As a theme emerging in this research, however, the justifiable use of deception to prevent harm to individuals was relatively rare and narrowly circumscribed. On the other hand, a more common, and somewhat more ethically murky theme, involved the use of deception during investigations. In this vein, one police officer spoke of interviewing a suspect while having a blank CD sitting on the table that he claimed had video of the suspect committing a crime. This tactic was used to provoke the suspect into talking with police (police officer, personal communication, December 19, 2014). The reasoning behind this use of deception is that it leads to the successful prosecution of criminals, which, in turn, results in greater public safety.
Although this use of deception is fraught with the potential for abuse, there are circumstances in which its use could be supported by the ASPA Code. Importantly, the police officer did not put the suspect in a position to commit a crime they would otherwise not commit -nor did the police officer fabricate evidence against the suspect. After all, if the suspect did not commit the crime, then he or she would know that the CD used to provoke that person is blank and could call the police officer's bluff. As noted above, the Supreme Court has upheld the constitutionality of using deception in defined circumstances such as these. In order for the deception to be ethically justified, however, the officers involved must ensure that they have recognized and understood the constitutional, legislative, and regulatory framework in which they are discharging their professional responsibilities (Practice 2a). Additionally, they must also ensure that they are promoting principles of equality, fairness, and due process in order to protect the rights of those being investigated (Practice 2b). It is also worth noting the importance of a commitment to using the practice in a manner that is equitable, non-discriminatory, and avoids disparities across different groups are important considerations (Principle 4 and Practice 4c).

Unjustifiable Dishonesty
Assuming the above conditions are met, the use of deception in investigations can be a valuable tool for improving public safety and advancing the public interest (Principle 1). All the same, practices that are permitted lawfully are not necessarily ethically justified, particularly where the potential for abuse is significant. For example, a former fire chief provided the following story: I thought about this and the only one I could come up with is that it was during a fire investigation in a hotel where there were a couple of people injured. One was injured very seriously. She jumped from the third story window, and it was obviously an arson fire that had been set. We had some suspicions about who we thought did it, and we were doing an investigation. It would have been myself, the police department, and the state arson bureau. In our investigation, we determined that we did have sufficient evidence to charge a person, but we were afraid that if we let it be known, this person would run. In the meantime, the newspaper somehow got word that this person was going to be arrested. The newspaper called me and said, "Are you going to arrest so and so?" I said, "No" (fire chief, personal communication, December 17, 2014).
Although the fire chief's motives for lying to the reporter were well-intentioned, doing so was an unnecessary case of dishonesty. Instead, the chief could have explained that it would be inappropriate to reveal the name of a potential suspect. By doing so, the chief could have adhered to the ASPA Code's commitment to "be open and transparent while protecting privacy rights and security" (Practice 3a). Furthermore, lying to the reporter might have damaged the department's relationship with the media once the suspect was ultimately arrested. Among other things, this would have demonstrated a failure to "recognize and support the public's right to know the public's business" (Practice 3b).
Another example that highlights the potential for abuse in using deception during investigations was provided by an assistant principal. In this case, the school deceived students by telling them that a lockdown drill was occurring when, in fact, the purpose of the drill was to conduct a drug search. According to the assistant principal: I honestly tried not to, but now that you mention it, we would have, for example, a lock down drill. What we would do, we had several levels of drills. One of the lock down drills was called a stay put drill. We would tell students that we were having this drill, and the teachers were to continue teaching, to lock the doors, not let anybody in or out for the safety of what may be going on in the halls. We brought drug dogs into the building. Of course, it didn't take long. One dog barking, they figured it out. The purpose was we didn't want them to concentrate on that. We wanted them to concentrate on what they were doing in their classrooms. I suppose that's a way of being deceptive. I wasn't forthcoming with the fact that dogs were coming in. Obviously, because we wanted students to act normally. If they were a person who normally brought drugs to school, we didn't want them to say, "Uh oh.
Better not bring them today." Our drills like that were usually unannounced, and especially if the dogs were coming into the building (assistant principal, personal communication, December 23, 2014).
The assistant principal's dishonest announcement to the students about the true nature of the drills represents a failure to "be open and transparent while protecting privacy rights and security" (Practice 3a). An approach more consistent with the ASPA Code, on the other hand, could have involved announcing to students at the beginning of the school year that periodic locker searches would be conducted by the police using drug dogs while the students were in class. This approach would have advanced the legitimate goal of keeping drugs out of the school in a more honest and transparent fashion. As it happens, the tactic used by the assistant principal seems better suited to catching unsuspecting students than to keeping drugs out of the school.
Interestingly, a number of the stories provided by teachers revolved around the theme of "sugar coating" in order to maintain positive relationships. One teacher explained that a tactic for telling parents something negative about their child was to "sandwich" the bad news by starting and ending the conversation with something positive about the child. The teacher said this was necessary to maintain a good relationship with the community (teacher, personal communication, December 23, 2014). These examples are interesting because the practice is really only deceptive or dishonest if the compliments made are untrue. While sandwiching criticism between positive (and true) comments about a student is a constructive way of providing feedback, cases of sugar coating that involve dishonest compliments are not consistent with the ethical principles contained in the ASPA Code. By failing to be genuine in their remarks, the school teacher fails to "serve all persons with courtesy, respect, and dedication to high standards" (Practice 1e). Additionally, this practice also would violate the duty to fully inform and advise by providing "accurate, honest, comprehensive, and timely information and advice" (Principle 5).
In a few rare instances, participants provided examples of dishonesty that were clearly not ethically justified. One public school teacher described a time when dishonest actions were taken in order to make it appear as if the school was in compliance with state law. The story told by the teacher involved students with individual education plans (IEP). The IEPs are tailored for individual students and, by law, require the student's parent(s)/guardian(s), teachers, and school administrators to meet together to discuss the plan. These meetings typically occur during the school day, but if a substitute is unable to cover the teacher's class, then they miss the meeting. The teacher, however, reported signing a document stating that he or she was present and agreed to the changes made in the student IEP meeting. When asked why the teacher signed the document even though he or she was not present, the teacher replied that state law required a teacher to be present and for the document to be signed (school teacher, personal communication, December 17, 2014). In other words, the teacher felt like signing the document was necessary in order to comply with a legal requirement, even though doing so was a misrepresentation of the facts.
Far from being a case of justified dishonesty, the foregoing story describes a situation wherein the teacher and administrators are complicit in acting dishonestly in order to appear to be following the law. The teacher and administrators essentially chose to lie about being present at the meetings for the sake of administrative expediency. This violates several principles in the ASPA Code. First, the action is a violation of Principle 1 and its commitment to advance the public interest. The action is clearly self-serving and done out of loyalty to administrators in the institution. The teacher therefore fails to "subordinate personal interests and institutional loyalties to the public good" (Practice 1d). Because the action is done to appear to follow the law instead of actually upholding the law, signing the IEP also violates the commitment to upholding the Constitution and the law (Principle 2). By agreeing to lie and sign the IEP, the school teacher also fails to "resist political, organization, and personal pressures to compromise ethical integrity and principles and support others who are subject to these pressure" (Practice 6c). Moreover, the school administrators have failed in this instance to promote an ethical organization (Principle 7) as well as to advance professional excellence (Principle 8).

Discussion
Consistent with previous research, the findings of this study indicate that honesty is indeed a critically important value in public administration (Molina & McKeown, 2012). According to our participants, honesty is essential in maintaining credibility and effective working relationships with colleagues, political superiors, and the public. In other words, honesty is good, not simply for its own sake, but for the role it plays in enabling public administrators to effectively carry out their work. Importantly, however, the stories that participants provided about the use of deception in public administration illustrate the moral complexity of the world in which they operate. Notwithstanding the fact that honesty is of paramount importance, public administrators do sometimes face situations in which they experience pressure to use deception. In some of these instances, the use of deception is appropriate and ethically justified. In other instances, however, its use amounts to nothing more than inappropriate and self-serving dishonesty. Of course, probably most instances fall somewhere in between.
The environment of public administration is characterized by tensions between competing values and ethical principles, each of which may be legitimate in their own right (Molina, 2009;Spicer, 2001). In part, this study has sought to make the case that the ASPA Code of Ethics can be a helpful analytical framework for sorting out these tensions. Public entities, however, also need to take careful steps in adapting the Code to their organizational context, or developing their own codes, in order to promote ethical conduct. Policies and procedures should promote honesty and transparency. Though it is not possible to create an environment entirely immune from unethical conduct, public entities should write policies and procedures that limit the ethical dilemmas that their administrators face.
This study also has important implications regarding the ethics training of public officials and the teaching of ethics in MPA programs. Although ethics training should promote honesty, it also should approach public administration ethics and values from a complex standpoint that reflects the difficult decisions made regularly by public administrators. Public administrators and students in MPA courses need to be prepared to face difficult decisions that require them to balance and compromise different values in service to the public. Along these lines, case-based ethics training is the best approach to ensuring that public administrators understand the role that honesty plays in public administration as well as the circumstances in which deception may be appropriate. In particular, cases should involve scenarios that trainees are likely to encounter in the course of their work. The best cases will be those that highlight the way in which honesty can come into conflict with other important public service values such as effectiveness, efficiency, humaneness, or benevolence. Discussion should focus on identifying the range of values that are in play in a given scenario, and the way in which considerations of principles, consequences, character, and duty can inform ethical decision-making in the public sector.
Nontraditional sources, e.g., popular sitcoms, movies, and novels, can be useful training and classroom tools for demonstrating the ethical dilemmas faced by public administrators. For example, the 2015 film The Martian depicts an astronaut (portrayed by Matt Damon) who is presumed dead and stranded on Mars when his fellow crew members must evacuate for emergency reasons, and the mission undertaken by NASA to rescue him. Set in 2035, NASA, as a public agency, is subject to strict transparency laws that requires the agency to promptly inform the public about the progress of the mission, including embarrassing information about mistakes and errors in judgment. In one scene, NASA administrators debate whether or not Damon's fellow astronauts should be informed that he is still alive, or if that information should be withheld from them in order to help them focus on their mission. Entertaining examples from these types of sources can keep the attention of trainees and students while also teaching a valuable lesson about ethics and serving the public. Additionally, the use of case studies derived from studies such as this one could be a way of presenting students with real-world examples of the types of ethical situations they are likely to face.
The results of this study also suggest a number of directions for future research. For example, the number and range of interviews collected for this study limits its generalizability. It is possible that public administrators in other regions of the United States view and treat the value of honesty and the use of deception differently than those participants included in this research. Further, public administrators outside of public education and public safety may view and treat the value of honesty and the use of deception differently. Public administrators in other professional fields such as human resources, public works, city managers, or budget officers may have different perspectives than those interviewed here. Similarly, public administrators at the state and federal level, particularly those who have less direct contact with the public, may have distinct perspectives as well. Future research on honesty and the use of deception should build upon these preliminary findings by broadening the scope of analysis beyond public education and public safety at the local level by including a broader range of participants.
Future research also may use cases derived from studies such as this one to develop survey instruments that could be presented to respondents drawn from large probability samples. This would help to increase the generalizability of findings across a broader spectrum of professional contexts and levels of government. Along these lines, comparative research would help to increase our understanding of how national and cultural differences influence public administrators' perceptions of the ethics of honesty and deception. Additionally, relatively little is understood about the extent to which public administrators' perceptions about honesty and deception are congruent with the perceptions of citizens. Future research that includes citizens could help us to better understand which uses of deception are likely to meet with public support. Moreover, this type of research would help to connect the public administration ethics literature to other important bodies of literature in the discipline such as scholarship on representative bureaucracy, citizen engagement, and democratic accountability.

Conclusion
This study has examined the ethics of honesty and deception in public administration. In particular, it explored the reasons why honesty is an important public service value as well as some of the circumstances in which the use of deception may or may not be justified. This study utilized the ASPA Code of Ethics as an analytical framework for assessing the cases of deception provided by participants through their storytelling. Careful consideration of the full range of principles and practices found in the Code allowed for a nuanced approach to understanding the ethical tensions and moral complexities of public service. As one participant put it, "We need to work toward being honest 100 percent of the time, except when it's permitted for us to use deception" (police chief, personal communication, January 6, 2015).

Disclosure Statement
The authors declare that there are no conflicts of interest that relate to the research, authorship, or publication of this article.

Keywords: Nonprofit Management, Social Media, Resource Dependency Theory
Social media is a vehicle for nonprofits to simultaneously engage with multiple stakeholders. Simply sharing information with donors is one way to establish interactions; however, dynamic social media posts are more likely to generate user engagement (Carboni & Maxwell, 2015;Cho & Gillespie, 2006). Studies of social media communications relating to nonprofit organizations and their volunteers indicate that higher-quality relationships are being developed through twoway, dialogic communications (Waters & Bortree, 2012). This positive outcome is due to the communications between volunteers and individual donors tending to be more of a mutually communal arrangement versus that of a contractual-symbiotic arrangement expected in a single agency grantor-grantee relationship (Waters & Bortree, 2012). Social media engagement allows many stakeholders to be actively engaged with an organization's mission and purpose while minimizing communication costs for nonprofits that might otherwise need to reach each stakeholder separately (Salamon, 1997;Waters & Bortree, 2012). The emphasis is to establish lasting and dynamic relationships with public stakeholders, based upon mutual respect and trust, which will help further the mission and goals of the organization.
In this paper, we examine the following question: Is stakeholder engagement associated with nonprofit revenue sources? In this study, we examine the relationship between social media efforts for stakeholder engagement and revenue sources in environmentally focused nonprofits. We compare social media engagement of nonprofits receiving government funding and nonprofits that rely entirely on private donations. These revenue sources involve different forms of stakeholder engagement. Receipt of government contracts entails a structured exchange between the funder and recipient. Receipt of private donations requires engagement of private donors. Carboni and Maxwell (2015) demonstrate social media is an efficient medium to simultaneously communicate with and engage with multiple stakeholders. We theorize the receipt of government funds reduces a nonprofit's incentive to engage with social media stakeholders because the recipient has other, more targeted avenues for fund development. To test this premise, we examine differences in social media engagement between environmental nonprofits receiving government funds and those not receiving funds.
To examine this question, we follow in the footsteps of Bortree and Seltzer (2009), who were the first to explore social media engagement among environmental advocacy groups. In their study, the authors found that organizational type is a factor in predicting organizational public relation practices. In particular, they posited that the "foundation" types of organizations would be less responsive than "membership" interest groups in terms of dialogic communication with external stakeholders. Similarly, our study examines the level of organizational engagement with external stakeholders and information seekers in terms of government funding. This paper contributes to the literature because few studies have examined the link between receiving government funds and the amount of social media engagement on Facebook (Clark, Maxwell, & Anestaki, 2016). Particularly, we do not have an understanding of how environmental nonprofits with varying levels of governmental funding will leverage social media tools. This paper finds correlational evidence that environmental nonprofits engage less with social media stakeholders as they receive more governmental funds.
This paper begins with a review of the principles of dialogic communication theory and its application in building trust among stakeholders. From that basis, we discuss social media use among nonprofits and develop hypotheses about differences in social media use among organizations receiving government funds and those not receiving government funds. Hypotheses are rooted in resource dependency theory (RDT) and dialogic communication theory (DCT). We test our hypotheses on data from environmentally focused nonprofits. We conclude with a discussion of the implications of our findings.

Principles of Dialogic Communications Theory
One powerful method for organizations to gain public trust is through dialogic communication. Dialogic communication is defined as an interaction where each participant has an opportunity to express him or herself. It functions as a process for building a relationship, and organizations failing to embrace dialogic strategies risk losing the beneficial stakeholder connection (Bortree & Seltzer, 2009;Kent & Taylor, 2002;Taylor, Kent, & White, 2001). This is a significantly different approach to public relations communication than is the concept of monological communication, where the organization simply functions as a disseminator of information (Kent & Taylor, 1998).
Social media provides a platform for symmetric, two-way communications that can be used for both moral and immoral purposes; dialogic communication implies a commitment to behaving morally. When an organization engages the public via social media, it is forming a relationship that involves honesty, trust, and vulnerability (Kent & Taylor, 2002). The emphasis, then, is to establish lasting and dynamic relationships with public stakeholders, based upon mutual respect and trust, which will help further the mission and goals of the organization. According to Kent, Taylor, and White, (2003), "Dialogic theory suggests that for organizations to create effective organization-public communication channels, they must be willing to interact with publics in honest and ethical ways" (p. 67).
There are five features of dialogue that Kent and Taylor (2002) suggest are necessary for the successful integration of dialogic communication via social media. These five features are: 1) Mutuality: the recognition of the organization's relationships with public stakeholders; 2) Propinquity: the spontaneous engagement with public stakeholders; 3) Empathy: the support of the public stakeholders' goals and interests; 4) Risk: the willingness to interact with the individual public stakeholders on their own terms; and, 5) Commitment: the organization being vested in the process of engagement with stakeholders in the public sphere.
These five features are readily accessible to organizations via social media in general and Facebook specifically, but they also require a commitment of planning and resources. Foremost is the commitment to the process of engagement that the organization must make to effectively engage with the public. The brief history of social media use by organizations is littered with stories of failure when organizations attempt to exploit social media in monological communication. This necessitates social media planning becoming a part of the organization's strategic planning process. The tenet of propinquity requires dedicated personnel resources to achieve, and these employees must be skilled not only in the technical aspects of the social media platform but also skilled in public relations practices such as explaining organizational policies or effectively responding to difficult public concerns (Kent & Taylor, 1998).
The commitment of resources required for an organization to conduct dialogic communication with public stakeholders is significant. Information cannot simply be aggregated and disseminated either intermittently or at a low level of the organization; there must be a process of "negotiated communication" (Kent & Taylor, 1998). This implies that the organization's leaders and decision-makers must devote their attention to this process: dialogic communication is a process as opposed to a product (Kent et al., 2003). This is the nexus of resource dependency and dialogic communication theories. There is also a growing public management literature surrounding this subject, which indicates that this idealized dialogue does not happen when it comes to the public sector (Bryer & Zavattaro, 2011;Mergel, 2013;Zavattaro, French, & Mohanty, 2015;Zavattaro & Sementelli, 2014). While this literature is largely centered on the public sector, research finds public and nonprofit organizations can behave differently due to different types of political or rule constraints (Chen, 2012). Thus, we expect that resource dependency and dialogic communication theories will differ in nonprofits.

Resource Dependency Theory
Resource dependency theory posits that organizations are dependent, to varying degrees, on the external environment for resources. This theory has been used in various contexts such as exports (Hessels & Terjesen, 2010), nursing homes (Banaszak-Holl, Zinn, & Mor, 1996), board of directors (Hillman & Dalziel, 2003), and nonprofits (Guo & Acar, 2005). As Pfeffer and Salancik (1978) note, organizations will attempt to adjust to and manipulate their environment to increase the prospect of organizational survival. If an organization is dependent upon its public stakeholders for the resources it needs to survive, then the investment of resources in dialogic communication is fully rational. If an organization is not as dependent on external public stakeholders for near-term resources, then the discounting of the value of dialogic communication is rational as well.
Organizational survival requires adaptation to the environment because organizations are dependent on the environment for resources. It is the interplay between the deployment of organizational resources and response from the environment that determines organizational organizational resources and response from the environment that determines organizational survival. Organizations pursue strategies to reduce uncertainty about organizational survival and risks for failure. Specifically, organizations attempt to decrease reliance on the environment for resources by controlling the use of resources, access to the resources, or the possession of the resources (Pfeffer & Salancik, 1978;Sheppard, 1995). Organizations pay careful attention to external stakeholders that are key to securing resources, especially when those stakeholders cannot be fully controlled by the organization (Pfeffer & Salancik, 1978). Resource reliance necessitates inter-dependency between organizations and their environments (Anheier, 2005;Pfeffer & Salancik, 1978).

Revenue Sources and Social Media Engagement in Environmental Nonprofits
In 2014, more than 20,000 environmentally focused nonprofits were operating in the United States (Guidestar, 2014). These organizations perform a wide range of activities, including education and training; environmental cleanup; pollution abatement; and energy conservation activities. These nonprofit organizations are a highly visible sector and have been successful in their advocacy efforts (Bortree & Seltzer, 2009;Lu, 2016). Additionally, environmental nonprofits receive a blend of government and private funds. More than half of these nonprofits have annual revenue under $500,000 (Guidestar, 2014), while a fraction of these nonprofits (262) had annual revenue in excess of $5 million. The total amount of the annual revenue of the nonprofits in the study sample was $1.4 billion with an average of $1,130,474 per organization. More than half of these organizations received government funding for a total of $344 million with an average of $1,646,904 in government funds (Guidestar, 2014 For environmental nonprofits, resources include government funds and donor contributions to fund the activities necessary to achieve their mission. Pfeffer and Salanik (2003) note that nonprofit management practices typically emphasize the need to seek a diverse base of funding streams. Nonprofit fund development strategies should be based on the type of funds a nonprofit organization pursues (Foster, Kim, & Christiansen, 2009). Receipt of government grants and contracts typically entails participation in an RFP process where nonprofits respond to government funder requests for proposals. There is minimal need to engage a larger group of stakeholders in pursuit of these funds. Additionally, these funding relationships tend to have an asymmetric power dynamic where the funder controls interactions, providing little room for the fundee to actively manipulate the resource environment (Auger, 2010;Maxwell & Carboni, 2014;Zweibel & Golden, 2007). Conversely, private funding often entails engagement with multiple potential donors. Nonprofits have more opportunity to manipulate the environment to increase the chance of favorable outcomes such as donations. Social media is an efficient medium to communicate with and simultaneously engage with multiple stakeholders (Carboni & Maxwell, 2015;Waters & Bortree, 2012). We predict that, as nonprofits become more reliant on private donations, they will be more likely to engage stakeholders via social media because it is a way to reach many potential donors in the environment with minimal requirements of cost and effort. This relationship is restated in the hypothesis below: H 1: As environmental nonprofits become more reliant on government funds, they will be less likely to engage stakeholders via social media.

Methodology
The study was conducted using the Guidestar (2014) database of nonprofits in the environmental sector. 1 Organizations falling under categories specifically for fundraising purposes were not included in the study. This query generated 26,223 environmentally focused nonprofits operating in the United States in 2014 (Guidestar, 2014). Further refinement of the category generated the 1,471 midrange organizations used in the study with annual revenue between $500,000 and $5 million. Two hundred forty-one of the organizations were eliminated due to sizable portions of their revenue (greater than 40%) coming from investment or service revenue, which made their resource stream significantly different than other organizations in the sample. The elimination left a net sample size of 1,230 organizations represented in the study.
From this group, we randomly selected 36 organizations and examined their social media engagement via Facebook. The 36 organizations were chosen by the random selection of 12 organizations from each income range, $500,000 to $1 million, $1 million to $2.5 million, and $2.5 million to $5 million to ensure we had representation from each revenue range. The selected organizations receiving government funds were compared to the selected group not receiving government funds to ensure their revenue levels were within 1% of each other.
Facebook is the most commonly used social media platform for nonprofits, with 89% of nonprofits having a Facebook presence (M+R & NTEN, 2015). Facebook also readily facilitates dialogic communication with public stakeholders through multiple avenues of engagement. These factors were the primary drivers for our selection of Facebook to conduct our investigation. We also examined the percentage of government funding to the organizations' total revenue. Revenue data was taken from IRS 990 forms.
We collected social media data using Simply Measured, a commercially available program that collects Facebook engagement. Our analysis includes data from a two-week period in December 2014. We follow the data collection practice that is consistent with communication study methodologies (Carboni & Maxwell, 2015;Weare, Loges, & Oztas, 2007). We define social media user engagement as the total number of likes, shares, and comments related to each Facebook post made by the organization. However, it should be noted our user engagement requires little effort and might be only a means to enhance the users' "feel-good factor" (Christensen, 2011). Yet as Christensen notes, evidence suggests there is a link between online activity and offline participation. Thus, it is important to understand even the most basic interactions users have on Facebook with nonprofit organizations.
An organization will typically post information and calls for action on their social media websites such as Facebook. Individuals viewing these posts have the option to "click through" to additional information such as links to other websites, multimedia content, or to "like" a post on the organization's page. Each time an individual intentionally interacts with the organization's page in this manner, it is defined in this study as an engagement. Software embedded in the website tracks these engagements, which allows for analysis of the level of engagement of social media stakeholders.

Findings
The total revenue for 1,230 organizations in the sector was $1,390,483,265 ($1,130,474 average per organization with a standard deviation of $977,403) for 2013. Government funds accounted for 24.72% of the revenue for the organizations in the sample ($343,704,117). In the sample, 639 of the 1,230 organizations received government funds (52%) with average funding of $537,037. The proportion of organizations receiving government funds in the total sample is aligned with the proportion receiving government funds in our subsample.
In our subsample, the average number of user engagements via social media contained several extreme values as we expected, ranging from zero to 27,903 engagements per organization, meaning that some organizations had no one interact with their Facebook sites during the period of study, while others had extremely high levels of interaction with their sites by visitors (see table 2). In the remaining sample, the average number of posts per organization for those receiving government funding was like those not receiving government funding: 12.3 versus 7.
There was an average of 100 engagements for all posts per organization in government-funded nonprofits and 2,457 engagements for all posts per organization in privately funded nonprofits receiving no government funding. Additionally, the receipt of government funds was negatively associated with social media engagement (r=-0.24; see table 3).
In our OLS regression, we examine the impact of government funding percentage, total assets, and total income on total posts and total engagement on Facebook (see table 4). It is essential to control for both the assets and income to account for the size of the organization. The size of the organization is an essential factor that will affect the number of posts and total engagement on their social media platform. While controlling for this, we find that percentage funds from the government is negatively associated with total posts while controlling for assets and income (ß=-0.06). However, this is only marginally significant. Second, and more interestingly, we find that a 1% increase in governmental funds is associated with a decrease of 31% in total social media engagement. Thus, organizations that received government funds had less social media engagement than organizations not receiving government funds. Yet, this is only marginally significant at the 0.15 level, but it should be noted that our small sample size yields less statistical power and increases the burden to find statistical significance at the 0.05 level.
Based on the data, organizations that receive more private contributions were more likely to successfully engage stakeholders via social media. The posts made by privately funded organizations were more likely to generate enough interest from individual stakeholders to cause them to act in such a way that demonstrates a two-way dialogue. In support for our hypothesis, as nonprofits become more reliant on private donations, they are more likely to engage stakeholders via social media in a two-way dialogue, as demonstrated by an increase response of the stakeholders to the Facebook posts of the organization. Conversely, we find that, as nonprofits become more reliant on government funds, they will be less likely to engage stakeholders via social media.
In sum, the increased response by stakeholders to the posts of privately funded nonprofits indicates that these organizations have been more successful in their efforts to generate interest on the part of their stakeholders. This increased level of interest is an important step in establishing resource support for privately funded organizations. On the other hand, the reduced response level of stakeholders to the posts of government-funded nonprofits may indicate a reduced likelihood of establishing resource support within the private sector. Findings lend support to the concept that resource reliance creates inter-dependency between organizations and their environments (Anheier, 2005;Pfeffer & Salancik, 1978).

Implications
Receiving government contracts are associated with fewer stakeholder engagements via social media. The average number of engagements with the non-government-funded organizations' sites is significantly larger for organizations receiving government funding: 2,457 engagements versus 100, respectively. Based on our findings, this paper has two implications.
First, the data show that nonprofits relying on private funds more successfully engage with the public. Following our rationale, this may be an organization's attempt to manipulate the environment in their favor. Increasing stakeholder engagement may be part of a larger strategy of developing reputation and improving the likelihood of private donations. By establishing long-term relationships, organizations can build trust and legitimacy in the minds of their public stakeholders, particularly, and the public at large, in general.
Second, this lower level of successful engagement may affect the long-term viability of the nonprofit if they are not engaged in other ways to attract stakeholders to the organizational mission and purpose. As Anheier (2005) points out, the nonprofit sector has historically served as an intermediary, facilitating the application of social capital into the functioning of civil society. The sector also serves to establish the trust necessary to build social capital.

Conclusion
Distinguishing social media use, and specifically Facebook usage, among nonprofits with private and government funding yields potential contributions to the burgeoning digital government world. First, the study of organizations and their resource dependency is well established, but little attention is paid to how organizations embrace Web 2.0 technologies (Carboni & Maxwell, 2014). Government must concern itself with funded organizations, acting as principals of government to act on government's behalf. Ethical and privacy considerations hover over social media use by nonprofits. To delve into these considerations, we must first understand how organizations use social media and other Web 2.0 technologies.
Within the world of Web 2.0, our government faces numerous challenges, which spill over to nonprofit grantees and contractors, i.e., false emails, fake charities, and website scams in search of unsuspecting individuals and other fraudulent phishing schemes are perpetual concerns for public agencies. Beyond these scams exist ethical and privacy lines that risk overstep by government agents acting on government's behalf. Online social networks (OSNs) turn private information to public in a click (Soma & Rynerson, 2008). Suppose an environmental nonprofit doing legitimate work for the U.S. government is found targeting logging companies with violence or destruction via Facebook. No longer is the content owned only by that nonprofit; rather, the contracting governmental agency must share some responsibility as the funding source of the nonprofit. Such arrangements blur lines between principal and agent farther than before Web 2.0.
This study has several limitations. First, the study has a relatively small sample size. This reduces our statistical power and increases the burden to find statistical significance. Additionally, our social media engagement measurements were limited to only two weeks of data due to the software constraints of Simply Measured. Last, we only examined the number of comments, likes, and shares. These types of social media interactions require little effort and we do not observe deeper and more involved participation.
To overcome this issue, future research should implement more qualitative data analysis to better understand the link between social media engagement and government funding. To conduct a deeper analysis, future research should examine the content of the message. Was the message informational? Did it have an emotional plea? Did it utilize a touching story? It would be useful to examine this across platforms such as Facebook, Twitter, and Instagram. In regards to grant data, it would be useful to collect more detailed data on grants (i.e., contributed income) and contracts (i.e., earned income). Lastly, it would be useful to examine the mission of the organization because this will influence how nonprofits will engage with their stakeholders. This extra level of data might provide more insight into the relationship between social media usage and governmental funds.
Dialogic communication via social media engagement and resource dependency requires further exploration, not only with social media as an adaptation to external environments but also in more nuanced settings that account for restrictions on adaptations. Nonprofits receiving public funding are less likely to engage in social media efforts or at least Facebook to engage stakeholders. Finding creative avenues for engagement in social media efforts without sharing private information could further help nonprofits act as citizens of the Web 2.0 as well as agents of the government.

Disclosure Statement
The authors declare that there are no conflicts of interest that relate to the research, authorship, or publication of this article.
Higher education has joined the growing list of public sector members seeking innovative solutions in an effort to reinvent itself in a time of decreasing state revenue (Government Accountability Office, 2014). Higher education has to now rely more on tuition as a primary revenue source, forcing them to compete with other institutions for admissions. One of the areas in which universities compete with each other is through campus offerings such as new residential halls with amenities similar or better than those at other schools and at home. The University System of Georgia (USG) is one such public sector member embattled with refocusing the mission of a number of its public college and university campuses while also having to account for greater competition amongst its member universities for admissions. This internal competition has left USG with significant debt on its books, leading it to search for a way to entice students while not causing further financial hurdles.
On April 14, 2014, the USG, the public higher education system in the state of Georgia, posted a request for qualified contractors (RFQC) for a first of its kind public-private partnership (PPP). Wishing to move away from its current long-term asset financing plan of utilizing public-private ventures (PPV), and to move much of the bonded debt, the USG issued an RFQC for the construction of campus housing on nine system-member institutions across the state of Georgia and the absorption of at least one housing asset on each campus by a single concessionaire. The agreement also includes a contracting period that lasts between 30 and 60 years, depending on the terms of agreement between the USG and the concessionaire. Hired PPP managers will oversee this one-of-a-kind PPP system at the system and university level. This paper examines this unique PPP system for its practice as a case study with practical implications detailed.
In an effort to evaluate the importance of this contractual arrangement, a case study approach that examines themes within the publicly available contract documents is analyzed. These documents present the economic circumstances and the values of the USG. Additionally, they display the efforts to grow through the development of infrastructure across the various USG campuses. The documents reviewed are only those written by the USG (i.e., not submissions from potential contractors), and the study aims to address three principal questions using those documents and inferences from their findings: 1. What are the rationales associated with the university system's decision to enter into a PPP arrangement for university housing? 2. What roles are identified for university housing and what implications loom from these views on housing within the system? 3. What are the implications on administrative practice associated with large-scale PPP arrangements?
The paper examines the concept of the PPP for its significance to administrative practice while also highlighting important elements of research on university housing for the present context. Furthermore, a brief explanation of the methodological technique undertaken for this study is provided to display the derivation of the information used to address the questions of this study. A discussion of the implications for higher education along with the use of PPP arrangements in new settings such as university housing also is provided. This discussion proposes a series of values that may emerge within other public higher-education systems while also detailing the importance within the context of the USG. Also, the use of this type of PPP arrangement is discussed for its implications to future public contracting endeavors. Using a review of the extant literature, the research is able to identify potential themes that may be present within the documents leading to the analysis and discussion.

Literature
In order to place appropriate context around the contract for university housing from the USG, previous research in two primary areas are identified and discussed. First, the rationale for using public-private partnerships is described. These discussions explain why public agencies consider PPPs as an acceptable option in public contracting. Second, the use of privatization of university housing explores past efforts to utilize contractors for more than just the construction of new housing units. This provides a review for how universities engage private contractors beyond the simple construction of new facilities and develops a basis for exploring longer-term relationships with these contractors for campus development. Prior to these reviews, a brief explanation of what a PPP is and its context in public contracting sets the foundation for remarks on rationale and university housing.
PPPs are more commonly associated with other forms of public infrastructure projects, especially in transportation. A clear definition on this form of contracting is embedded in the description of its makeup. There is not an agreed-upon singular form. Rather, a series of general principals describe how a PPP appears. These principals utilize a long-term contractual agreement, usually 20 or more years (Savas, 2000), to construct a partnership between a private contractor and public agency in an effort to accomplish multiple tasks associated with a public project. This type of arrangement permits the private contractor to take on a greater role within the execution of the project. This comes in the form of completing multiple tasks that are commonly found in different procurements. For example, if a road construction project is needed, a PPP would enable the public agency to contract for the construction, as well as operation and maintenance for a number of years following the completion of construction, to the same contractor. But the rationales behind public agencies utilizing such an agreement are relevant to the contractual situation.

Rationales for Public-Private Partnerships
Specific goals associated with these contractual arrangements tend to vary by agreement but have common rationales. A PPP is a relationally developed agreement built on mutual accountability (Wettenhall, 2005). Such accountability implies a common goal and equality. While the organizational goals are unique -efficiency for the public organization and a more stable return on their investment for the private organization (Runde, Offutt, Selinger, & Bolton, 2010) -for each partner, the contract and corresponding project assume the need for both sectors' expertise. Another unique feature of the PPP is the long-term agreement that may last in excess of 20 years (Savas, 2000). Although these projects have perceived risks and rewards, the focus remains with the benefits to the public organization to highlight the potential reasons an institution of higher education might seek out a PPP arrangement for infrastructure development.
The first of the many benefits perceived by using a PPP is access to the private sector (Brinkerhoff & Brinkerhoff, 2011;De Bettignies & Ross, 2004;Narrod et al., 2009;Wood & Gray, 1991). Access comes in the form of financial, managerial, and technical expertise. This desire to access the private sector is based in the belief that the public sector is incapable on its own to accomplish the level of desired project efficiency (Forrer, Kee, Newcomer, & Boyer, 2010). Thus, the public agency desires an arrangement to account for known shortcomings, and those are believed to be correctable through the partnering with private business. Engagement in private financing, management, and technical knowledge permits a public agency to reduce risks associated with funding, maintenance, and specifications of a project.
Alongside the hope for efficiency comes the sharing of responsibility (Hodge & Greve, 2010;Steyer & Gilbert, 2013). Having access to the private sector is viewed as beneficial, but being able to ease responsibility is a means of mitigating risk and enhancing the long-term capacity of the public agency (Hodge & Greve, 2010;Koppenjan & Enserink, 2009;Steyer & Gilbert, 2013). This creates an opportunity for risk-averse behavior for the public agency through the offloading of project burdens and freeing of resources for the execution of other projects.
Other potential benefits include reducing the cost of contracting (Forrer et al., 2010), financial improvements on borrowing rates (Hodge & Greve, 2010), relationship-building due to the longterm nature of the contract (Brinkerhoff & Brinkerhoff, 2011;Steyer & Gilbert, 2013), and the perception of governmental innovation (Forrer et al., 2010;Link & Link, 2009). These represent financial-and perception-based benefits that aim to cut short-and long-term costs through the collaboration with private business partners and through the use of entrepreneurial financing strategies. Increasing the length of relationship seeks to build trust and enhance communication between the public agent and private contractor. Having a shared mission of project completion fosters this trust. Furthermore, this relationship grants a level of autonomy to the contractor that could enable more innovative strategies for the life of the project while reducing the need to procure additional services associated with different stages of the contract. Thus, the contractor maintains a valid stake in the project and invests in its success.
The place of these benefits in higher education still presents itself in the same manner. Institutions seek arrangements that produce a fiscally favorable outcome and address issues of capacity and risk. This becomes the place of the PPP arrangement whereby any or all of the aforementioned rationales are sought. It is also important to note these are all viewed as positive outcomes in a contractual arrangement that is not without risk, the most relevant of which includes value differences between public organizations and businesses (Brinkerhoff & Brinkerhoff, 2011), communication breakdown (Forrer et al., 2010), and issues of autonomy (Avant, Finnemore, & Sell, 2010).
A discussion on the privatization of university housing follows in an effort to detail the history of contracting with private industry. While not a direct PPP arrangement, a PPP is a mechanism for contracting out, and the context of such behavior in university housing establishes a foundation for engaging in a PPP arrangement.

Privatization of On-Campus Housing
On-campus housing in the United States has existed in one form or another for as long as modern universities have existed in the U.S., though it really began to flourish following World War II (Frederiksen, 1993). Through the G.I. Bill, thousands of soldiers back from the battlefields of Europe and Asia sought to enroll in universities across the U.S. (Lucas, 1994). They were hoping to prepare themselves to compete in the newly developing post-war economy; instead, they found universities and surrounding communities that were unprepared for the increase in enrollment, especially when it came to housing options (Frederiksen, 1993).
Colleges tried to build efficient and affordable housing in a timely manner to match the increase in student population, but, due to the time requirement of planning, funding, and building residence halls, they never could match demand (Olson, 1973). The inadequacies of the university system to offer sufficient on-campus housing options forced students to seek housing options off-campus. These off-campus housing options were owned primarily by out-of-town landlords who took advantage of the ever-increasing demand in housing and the stagnate supply of housing (Bayless et al., 2013). The landowners did not put much investment back into their rental properties and allowed the properties to quickly deteriorate into rather unsightly structures. The areas surrounding many universities became known for what Bayless, Wilhelm, and Wills (2013) described as "student ghettos," or what we would more appropriately call areas of residential decay. These areas were not only unsightly for universities, they created the narrative of the reckless college student that destroys their rental property while failing to pay their bill on time -a narrative that Bayless et al. (2013) proclaims is best viewed in the movie Animal House and one that at first prevented private companies from investing in residential housing on college campuses.
Allen and O'Hara, a Memphis-based contractor more famously known as the primary contractor for Holiday Inn, was the first private company to invest in primarily student housing when it built Granville Towers in Chapel Hill, NC, in the 1960s (EdR, 2017). Following its success with the Granville Towers, the contractor went on to build more college-student-specific housing. Its buildings were the standard "dormitory"-like rooms with twin beds and minimal furnishings, and the contractor soon found it difficult to compete once student housing preferences started to evolve to include more private living options.
In an effort to match shifting student preferences away from traditional housing options to more personal accommodations, including private baths and bedrooms, Polar-BEK and Cardinal Industries began to build off-campus living options in the 1980s (Shearin, 2011). They planned and developed housing similar to suites or apartments with a two bedroom to one bathroom ratio, and students had the privilege of larger beds. Yet, while this marks a significant improvement in off-campus housing options developed solely for students, these private companies had yet to bring the innovative lessons learned from off-campus housing to oncampus housing, which still to this point is primarily dominated by traditional or dormitory-style housing. In other words, student housing was still separated between mostly public offered traditional on-campus housing and more comfortable private options available off-campus.
The lack of on-campus development by private companies changed in 1991 when a private company, Century Development, launched a primarily on-campus housing initiative (Shearin, 2011). The new initiative marked what might be considered the start of the public private partnership movement within on-campus housing. The new initiative deviated from the traditional housing finance model of complete finance by the university and moved to a new finance scheme. Century Development spawned two leaders in early public/private on-campus housing movements: American Campus and Capstone. The preferred financing scheme used by both American Campus and Capstone was 501(c)3 tax-exempt bond financing structures with the developer largely in control of the development, management, and buyout terms (Bayless et al. 2013). This process will be referred to throughout this paper as a public private venture (PPV). Under these PPV agreements: … a qualified not-for-profit organization, such as a university foundation, acts as the borrower and owners of the project through a ground lease from the university. Under this structure, 100% of the project's development cost is then raised through the sale of tax-exempt bonds, and the only collateral for repayment is the revenues generated from the project. The private developer guarantees project completion and is paid a fee for services rendered (Bayless et al., 2013, p. 113).
While this financing scheme was popular amongst universities, Bayless et al. (2013) states that there were some drawbacks, including but not limited to thin debt coverage ratios, highoccupancy break-even points, and no financial investment or risk for the developer beyond building the community. In short, the new housing options were expensive to build and required evolving financial mechanisms to finance the projects. Because of the cost and the fact that repayment is tied solely to revenues generated from the project, the PPV projects were only profitable when enrollments were large enough to guarantee a high-occupancy rate, costs of upkeep were low, and competition between on-and off-campus housing proved minimal. This ideal situation hardly ever took place, and many universities found their selves with significant debt on their books. The PPV system is still the preferred means of financing on-campus growth for many universities, but others, riddled with debt due to their past PPV attempts, have started to search for alternative financing methods.
In order to establish the purposes of university housing, a brief review of the goals associated with services of university housing are detailed. These goals, along with the various rationales of PPPs, produce a series of desired project outcomes associated with the USG case.

Student Housing Goals
In his foundational piece on the educational role of college student housing, Riker (1965) developed the two bedrock assumptions of the field: environment influences behavior and learning is a total process. Environment, for the purposes of student housing, is defined as a safe, inviting, and supportive residential facility (Riker, 1965) as well as a positively reinforcing mentorship program through live-in student and professional staff members (Newcomb, 1962). Learning as a total process is related to the belief that the college experience is transformative in nature and that no one factor has an impact on the totality of learning (Riker & Decoster, 1971).
The environmental and total process of learning assumptions led Riker and Decoster (1971) to develop five general objectives for student housing, which are summarized in table 1. The objectives include the "provision of a satisfactory physical environment through new construction and renovation"; "adequate care and maintenance of the physical facilities"; "establishment of guidelines that provide structure for compatible and cooperative community living"; "development of interpersonal environments that reflect responsible citizenship and a concern for others, as well as an atmosphere conducive to learning"; and "opportunities for individual growth and development" (Riker & Decoster, 1971, p. 6). Riker and Decoster's (1971) objectives of student housing were revisited by Palmer, Broido, and Campbell (2008). Analyzing recent scholarship on the impact of student housing on student retention and resilience, including Pascarella and Terenzini's (2005) synthesis of over 2,600 studies, Palmer et al. (2008) found general support for Riker and Decoster's (1971) two bedrock assumptions and five goals of student housing. Due to their findings and Pascarella and Terenzini's (2005) findings, we will use Riker and Decoster's (1971) five objectives as the general goals of student housing.

USG Case
The present case of the University System of Georgia positioned itself to utilize a PPP as the contracting mechanism on an enormous scale. This presents a clear indication that the belief in the worthwhile nature of the contract against the potential risks provided a more beneficial outcome than not. However, even within the frame of the potential benefits is the question of which benefits are the most critical benefits within each contract justification. These justifications lead to a deeper understanding of the position of the USG as well as its long-term intentions for infrastructure development on its campuses.
Many universities and university systems, such as USG, have stuck with the private-partnership (PPV) model but have moved closer to a pure public-private partnership (PPP) model in which the private party designs, builds, upkeeps, and maintains ownership of the property for a certain period of time, while the public party is paid to operate the building. Examining the PPP system recently introduced by USG is important for it represents an evolution in the role of public private partnerships in on-campus housing. The following section details this new PPP through detailing of relevant contract elements.

Contract Details
USG awarded Corvias Campus-Living-USG, LLC (hereafter referred to as Corvias) the concession for the development and investment in a portfolio of student housing on November 12, 2014. The term of the contract is from the effective date of November 20, 2014, to the closing date of June 30, 2045. At the end of each lease, Corvias is responsible for tearing down the building or turning the building over to USG. It is University System of Georgia's decision of which closing action the Corvias is required to take. USG retains sole power to terminate the agreement early should it choose after notifying the concessionaire ahead of time and paying the fair market value for the buildings.
Corvias' concession not only includes the development and construction of new housing, but the absorption of existing housing along with the debt associated with that existing housing. Each university is required to lease one existing housing unit, but some, such as Abraham Baldwin College, leased all of their existing housing units. Corvias is responsible for the management, operation, maintenance, and repair of each of the existing housing units it leases.
The building of new facilities on all universities associated with the PPP (referred to as new housing), the leasing of existing housing by the concessionaire (referred to as existing housing), and the retaining of housing units by many universities (referred to as retained housing) create a need for each institution within the PPP to work out specific provisions, agreements, and plans with the concessionaire related to the specifics of their campus. In addition, it creates the need to normalize operations across all institution-level housing arrangements. Specific needs addressed include retaining residence life responsibilities for USG in all housing options, making sure that housing fees are similarly determined, and making sure that the different forms of housing do not compete against each other for student contracts.
Each institution that is a part of the agreement is responsible for providing and managing residence life programming for residents as determined in sole and absolute discretion by each institution. The only stipulation regarding the levels of residence life programming is that it must be equal across all three types of housing. The concessionaire is responsible for providing space in both existing and new housing for the planning and administration of resident life programming. In addition, the concessionaire is responsible for renting to each institution administrative spaces, including but not limited to office space, resident assistant options, and live-on staff housing options. Further discussion of resident life and education responsibilities and benchmarks are left absent from the official USG documents.
Housing fees for the first academic year are set in the master concession agreement. Every year thereafter, a management review committee (MRC) shall set the fees for that academic year at the annual MRC meeting, and the fees have to be approved by USG. General rules associated with yearly housing fees stipulate that fees cannot increase by more than 3% a year unless the concessionaire requests USG approval for a greater increase due to extraordinary issues. The concessionaire shall fund, at least every three years, a market study by an independent thirdparty consultant to determine the competitive market for student housing by comparing current housing fees for new, existing, and retained housing with off-campus housing in the surrounding community and on-campus housing at nearby institutions. The study should include benchmark rates for future housing fees, demographic characteristics, and occupancy rates and will be used to determine a recommended residential housing rate for each project. The concessionaire can increase the housing rate beyond the 3% mark after the conclusion of the market study if the need arises. The concessionaire is responsible for coordinating and paying for the cost of the market study.
When it comes to ensuring that the concessionaire does not have to compete with retained housing for occupancy, fees are uniform across campus with relation to type of housing and in marketing. It is the responsibility of the concessionaire and the institution to determine target occupancy levels and to work together to fill all available beds at all projects. Each institution is responsible for marketing and recruitment for the entire campus, including existing and new housing leased by the concessionaire. If the concessionaire wants to market existing or new housing, it must obtain the permission of the home institution of the housing unit, pay for the marketing, and cannot include any special incentives not available institution wide. On the other hand, all marketing efforts by the institution cannot differentiate between new, existing, and retained housing. All forms of housing must be labeled as "on-campus housing." In addition, the institution must include all on-campus housing in the marketing packages and cannot market just retained housing. Last, it is each institution's responsibility and authority to determine the application process, residency requirements, room assignments, wait-list procedures, room occupancy levels, and terms of the resident contract.
Some additional elements of the contract that do not fit neatly into the previous categories are that it is the responsibility of the concessionaire to upkeep not only the new and existing housing but also the grounds that they stand upon. The contract is open as to how the concessionaire can carry out its upkeep responsibilities. It can hire its own company or it can pay existing campus auxiliary services to maintain the property to the same level as retained housing. In addition, it is the responsibility of the concessionaire to hire a representative of the company for each institution to serve as the non-live-on manager of the facility. This person is responsible for carrying out the agreement between Corvias and USG along with each institution's housing/residence life office. Last, and maybe most important to some people, is the fact that this lease agreement is for a period of 30 years, and the concessionaire retains the first-right-of-refusal over future housing expansion on any USG campus. This means that, unless Corvias declines, it will be awarded the concession for any future housing on any USG campus for a period of 30 years. This point is important and is worthy of its own study, but, for the purposes of this paper, it is just another element of the contract.
Details on the technique used to evaluate the USG case are presented in the next section to explain the derivation of themes and implications associated with this case. The methodology details the actions of the researchers and the sources of information pertinent to the study.

Methodology
This study explores the questions surrounding values within the contracting process related to the public-private partnership arrangement involving university housing within the whole of the University System of Georgia signed in 2015. More specifically, the study addresses the following questions: 1. What are the rationales associated with the University System of Georgia's decision to enter into a PPP arrangement for university housing found within the published procurement documents? 2. What roles are identified for university housing within the published procurement documents and what implications loom from these views on housing within the University System of Georgia? 3. What are the implications on administrative practice associated with large-scale PPP arrangements?
In order to provide a discussion relevant to these questions, this case study utilizes a thematic analysis of the documents made publicly available from the University System of Georgia website related to the procurement process for the contract associated with the university housing public-private partnership. A thematic analysis is a form of content analysis focused on code development (Boyatzis, 1998;Braun & Clarke, 2006). An interpretive form of content analysis focuses heavily on manifest codes and storylines within documents (Ginger, 2006;Weber, 1990). This differs from thematic analysis in that thematic analysis also concerns itself with latent codes as well as manifest codes (Boyatzis, 1998). Within the present study, an interpretive approach to theme exploration is necessary for identifying more latent -unwritten -themes, as these themes will not be explicit in public documents, but evidence of their The process for deriving themes begins with coding these data. Creswell (2016) describes the process of coding as conversion of raw data, or text, into a narrative on the basis of conceptual themes that emerge from the various code groupings. For the current study, the process of coding will focus on the use of previously developed themes within the literature as a basis for the researchers' code identification in the text. Hwang, Zhao, and Gay (2013) implement a similar process in exploring public-private partnerships in Singapore to identify risk elements. These themes are those described in the literature pertaining to PPP rationales and goals of university housing. The codes are used to highlight text within only documents made publicly available prior to August 2015. Additionally, only documents written by the University System of Georgia are evaluated, as the qualification and proposal submissions from possible concessionaires do not illustrate themes emergent from the perspective of the University System of Georgia.
Under the steps associated with an acceptable process of coding (Creswell, 2016), this study undertakes a six-stage process prior to developing the discussions and conclusions for this research. First, potential codes are categorized as Student Housing Roles and Public-Private Partnership Rationales (see table 1). These codes are developed from previous research discussed in this paper. By using that literature to develop the codes, it is easier to identify the values and roles portrayed in the documents by USG.
Second, each document is read in its entirety, followed by an initial coding of the documents. Third, codes are evaluated for their overall quantity and presence in order to develop themes. Two separate coders evaluate the documents, and themes are developed by both coders separately in order to ensure more than one evaluator in the interpretation process (Burla et al., 2008;Schreier, 2012).
Fourth, coders assess the coding and themes from the other coder in order to determine the presence of intercoder reliability (Burla et al., 2008;Elo et al., 2014;Vaismoradi, Turunen, & Bondas, 2013). Elo et al. (2014) explain that, while there is not a uniformly accepted method for assuring intercoder reliability, by simply establishing a process of conducting an assessment, the research is able to enhance trustworthiness. Finally, development of the discussion takes place on the basis of the manifest and latent themes agreed upon by the coders following the assessment for reliability in their separate views on the documents. Within the discussion, themes are analyzed on the basis of their categorization and the overarching storyline that develops from them (Ginger, 2006). Additionally, the discussion focuses on how the University System of Georgia views the usefulness of public-private partnerships and their view on student housing as a comparison and contrast to the evident themes present within previous research. This highlights the value structure of the University System of Georgia both administratively and institutionally. These values are discussed for potential implications in the future with university contracting and the roles university housing might play as a bargaining chip for institutions across the country.

Discussion
Within this section, details pertaining to emergent themes and the contract documents are explored to establish findings related to the questions guiding this study. These questions signify the relevance of this case in university housing but also divulge results underlying the innovative practices in public sector procurement. Results relating to the coding for themes -rationales of PPPs and goals of university housing -are presented with details relating to portions of the contract documents as the basis for discussion on the importance of this case.

Code Results
In revealing the results prior to discussing the importance of the findings, each of the categories and their corresponding codes are detailed for their context and frequency within the various documents. The discussion of these will follow in order to provide clarity regarding the importance of the contract within higher education and student housing.
The student housing goals are mentioned first and last when describing the goals of the project.
The USG system identified five project goals: As the documentation proceeds, the primary focus of these goals continues to drive for the upkeep and development of the physical (e.g., the buildings). Table 2 demonstrates that the most prevalent codes are facility maintenance and developing satisfactory physical environments. The documents contextualize this through the description of key building components and the need to ensure that the construction accomplishes the purpose of new housing. A majority of this is found in the operations agreement portion of the procurement documents. This is to be expected, as this part of the contract typically indicates how the project will progress throughout its life. Table 2 also demonstrates the lack of presentation these documents provide pertaining to students and the relationship to the purposes of student housing. Only one of the three goals of student housing related to the key stakeholders (i.e., students) is found in multiple parts of the procurement documents. This is the notion of promoting individual growth and development of the students. However, this is framed in a manner similar to the second half of project goal five: "to enhance the college experience for these students." Therefore, the educational component of campus housing according to official USG documents is tied strictly to the construction of newer facilities. This is evidenced by the lack of presence of the other two goals relating to students: community living and learning environments.
With the lack of presence and context for promoting community living and fostering atmospheres of learning, USG is presenting the attractiveness of housing as the key value relating to student housing. This value, as indicated in the publically available documents, does not follow the total learning process argument of Riker (1965) and places the purpose of housing as an admissions body and not a student-retention body. The lack of a retention focus on residence education is striking considering that USG has signed on to the complete college plan, which places retention and completion values over admission values in order of importance. Finally, the context also fails to truly discuss the students and their function in student housing -beyond that of filling the rooms. Much of this also can be attributed to the rationales associated with entering a public-private partnership arrangement found within the documents.
The PPP arrangement, while new to USG, is not new to public organizations. There are perceived benefits, and those are not missing from these documents. As table 2 indicates, four of the nine PPP rationale codes are not only present but also present throughout. These are heavily focused on accessing the private sector for their expertise, sharing of responsibility, reducing risk, and developing long-term partnerships. Together, these codes identify the major value of this contract: risk aversion. USG is actively pursuing the reduction of risk in the form of infrastructure development and financing. These are most clearly seen in project goals two through four. All three of these goals expressly demonstrate the problem and belief in the solution to mitigate the risk associated with building new housing and taking on greater amounts of debt.
Additionally, the unique nature of the PPP agreement and its cost savings are moderately present, indicating these are important but not as much as engaging the private sector for its expertise. Reducing costs, completing the project more efficiently, and utilizing an innovative strategy all indicate that there is hope that this type of arrangement can be beneficial long-term. The greater presence of the long-term partnership code also alludes to this, but these three codes are more in line with the longevity of the project itself. There is also language indicating these go hand-in-hand with project roles. While responsibility sharing is prevalent, the context of innovation, cost reduction, and efficiency are linked to the ability for USG to maintain a presence in decision-making while limiting the need to constantly modify and develop new contracts. This is due to these three codes being strongly associated with problems of public procurement and how much time and effort are typically necessary for multiple contracts to be executed. Instead, this agreement sets forth an opportunity for USG to execute a single agreement, monitor project progress, and reduce the time and costs associated with future phases of the contract.
Of the codes not present, improving borrowing rates and improvements on capacity are not as present largely due to the scale and magnitude of this agreement. The contract itself is designed to eliminate the need for USG to improve its capacity to borrow and construct. Instead, the intention is to access the private sector with a long-term arrangement that enables those infrastructure developments and financial responsibilities to fall to Corvias. Thus, it would not serve the purpose of the agreement to find these codes present.
In total, the values present within the documents are attractiveness and risk aversion. These values are present to such an extent that the role of student housing, the benefit to students' growth and the growth of the universities are superseded. What this presents is a question on the cost and benefit with respect to the direction of public higher education. Specifically, is cost cutting and campus beautification actually more important than the learning taking place on campus? This is not an argument for discussion in this study, but the values demonstrated within these documents do present a picture of administrative and operational shifts in higher education.
The study does indicate that excessive debt is leading to the use of riskier agreements in place of financial risks associated with campus expansion and development. Although risk aversion is evident throughout, mention of the risk of default is limited. Instead, much of the documentation lends itself to the shift in responsibility for USG campuses in monitoring the developments on their campus. Once more, this is a form of risk aversion for USG by limiting their administrative responsibility in place of the individual campus responsibilities and changes due to the contract terms and conditions. This also highlights the system's structure as a potential factor into the perceived necessity of this contract. It is certainly beneficial to keep student housing affordable, but the question of why housing is becoming unaffordable lends itself to questions within the system as a whole.
Thus, the study is left with as many questions as answers. The values for this contract are present, but implications are to be determined on the basis of the administration of the contract and clarity to the different USG schools directly and indirectly impacted. However, there are a series of implications relating to PPPs, university housing, and general practice that emerge from this study.

Importance and Implications
The importance of contracting in public organizations is often exhibited in infrastructure projects. This particular contract within USG is both a first of its kind and one that invites future agreements connecting student housing to private contractors. Whether the partnership succeeds or not could create a model whereby universities struggling to accommodate the need for expansion will look upon and mimic.
Due to the need for change in higher education infrastructure development, this form of partnership also might become attractive regardless of its perceived success or failure. Universities must meet the needs of their clientele while attempting to minimize costs. In the United States, this is particularly evident, as public universities cannot rely solely on public funds or long-term bonds for infrastructure development. The USG believes it has found a solution to utilize a long-term contract with a single concessionaire and has decided to invest in this partnership without evidence demonstrating its effectiveness in higher education and student housing. Thus, the implications for universities, student housing, and private contractors specializing in student housing and university development are vast.
From the standpoint of student housing, this is a decision that likely will lead to more universities contracting with similar agreements in hopes of reducing costs and the burden of housing students on campus. Public-private partnerships are contracts that take too long to determine actual success and failure for universities to await results. The structure of the contract offers an opportunity to create the housing facilities needed to attract students and also to foster that sense of community within the campus itself. While this type of agreement may provide an option to develop these facilities, it still requires efforts to breed the community that describes one of the primary purposes of housing. It is clear this agreement focuses heavily on the financial benefits without determining the long-term implications. What will become of housing fees? If the concessionaire defaults on the contract, how will the university handle it? These questions exist for all PPP agreements, but those contracts typically do not have a set of stakeholders that could be greatly impacted should the contract fall apart after it is executed. For housing, it is the question of student impact that is left uncertain at this point.
For universities and their development, this type of agreement also might be viewed as an option in other infrastructure projects. Constructing housing facilities often is limited in its purpose. However, with universities becoming more like small communities with dining, shopping, residency, education, and athletic and fitness facilities, the building necessary for growth is often too burdensome. This contract is an opportunity to grow in other areas of the campus as well. PPP agreements show openness with regards to their scope of work. By enabling the private contractor and the university to develop a long-term plan for their agreement, it is possible to incorporate a variety of purposes into new infrastructure. Universities will certainly see the financial benefits through minimizing their upfront costs but also the marketability of their community when attracting students. In the United States, this is particularly important as tuition rates increase and the attractiveness of higher education is beginning to wane. Aspirations of an attractive campus with quality facilities and smaller tuition rates are utopian. While the PPP agreement does not make promises, it provides an option for these campuses at minimal upfront cost to the university.
This could be seen as a door into a new form of investment for contractors with the capacity, experience, and understanding for universities and infrastructure development. The private sector largely sees PPP agreements as the mythical stable return on investment. On university campuses, having these long-term partnerships produces a series of opportunities for these contractors with growth of their organization and also major profits for those capable and qualified to execute the contract and complete the project. Should universities seek to expand this type of contract into other areas on campus, the contractors being awarded contracts for other projects would now have an investment and partnership that could see them being given future opportunities. The biggest hurdle for private contractors is capacity. The nature of PPP agreements is limiting on both the public and private sector sides due to the need for large investment capacity of private contractors to safely enter into these agreements. Thus, only a limited number of primary contractors can undertake the project on their own. However, for the smaller contractor, these do still provide opportunities for subcontracting roles throughout the project.
This agreement with USG and Corvias is unique, new, and critical to understanding the possible future direction of contracting out in higher education. Within this study it is demonstrated that the overwhelming evidence for the purpose of this form of contracting is to mitigate risk, connect to the private sector, and decrease costs while expanding university housing in USG to meet the perceived demand. However, it is also evident that the contract does not place emphasis on the student, and the implications for student housing are more financial rather than a concern for learning environments and community atmospheres. Future research must follow the development of this contract and those similar to it in order to monitor budget impacts along with the further use of this form of contracting. Additionally, the impact on students, their relative costs and perceived sense of community also warrants exploration.
The study is limited largely by its timing. It is speculative with respect to any actual financial impact. Furthermore, the changes in higher education are also limited to conjecture. Due to the focus of the study being on procurement documents, it is also narrowed in its ability to fully identify values of USG in student housing, as these documents focus largely on the financial and construction elements. However, in describing the purpose of the contract, USG attempts to highlight goals that expand beyond the financial benefit and do link to student housing throughout the documents. Also, in tracking any PPP agreement, researchers cannot afford to await results for 30 or more years to discuss implications.
In looking to the future, this contract between USG and Corvias could serve as a milestone in higher education and public contracting. It is already intriguing for the apparent shift in goals of student housing and the willingness for a large university system to reach long-term agreements with the hopes of expanding and privatizing on a public university campus. The completion of the various elements of the project is merely the first step. Whether other universities follow this example and undertake these long-term partnerships should soon be evident. Regardless, this contract creates another bridge for the public and private sectors and provides a new look at contracting out on public university campuses.
Discussions of health care policy and implementation often focus on tripartite factors of access, cost, and quality of care, in many cases with an aim of finding balance among these factors (Thompson, 2013). Not surprisingly, these discussions are often centered on tremendously broad systems-level considerations to the exclusion of more focused topics such as the behavior of health care providers and patients and the interactive aspects of caring processes. These behavioral and relational aspects of health care are crucial, though, and may have a substantive effect on the quality of care provided. Indeed, they constitute the lived experiences of both patients and health care providers and, thus, shape both our understanding of health care and our personal narratives about how well health care is working.
Stephen Buetow's treatment of the interactions between patients and health care providers focuses intently on those micro-level aspects of care with purposeful attention to the consideration of both those receiving care and those providing it. At the center of this discussion is the idea that both patients and caregivers are worthy of consideration as individuals, and that the cognizance of individual values, knowledge, and others will have a substantive effect on improving health care on a broader scale. Buetow divides the book into two sections, with the first outlining the case for a necessary shift in the ways in which we view both health care providers and patients, and the second suggesting fundamental changes to how we think about values in health care and implementing concrete recommendations for "person-centred" care.
In Chapter 1, Buetow provides an introduction to concepts of personhood in health care by describing a case of his own medical care in which he felt that clinicians were treating a condition rather than a whole person. Buetow notes that "[r]eports like mine -of patients feeling dehumanized and a burden -are ubiquitous in health care" (p. 1). The standardization of health care process and reduction of the caring process to a routinized activity are at odds with the fundamental nature of health care as focused on individuals. Despite persistent calls for patient-centered care, Buetow notes that the process "… reflects a pervasive cultural malaise: the 'depersonalization' of patients and clinicians, in its de-humanizing and de-individuating senses" (p. 3).
Chapter 2 discusses health care processes from the perspectives of clinicians. This includes what has become our standard perspective of viewing patient care activities in health care -one in which the clinician's care for the patient and the perspectives of the caregiver are brought to the fore -but also the difficulties of the standardization and commodification of health care. Buetow provides a substantive overview of the historical and educational roots of the perspectives of health care professionals as they engage in caring processes, and highlights the concept of the patient-centered perspective as "… an ideal that expects modern clinicians to focus attention on, and care for, patients rather than themselves" (p. 19). This chapter also outlines both a principle-based view of clinicians, emphasizing patient welfare, patient autonomy, and social justice (p. 22), while also presenting a number of challenges facing this patient-centered view of the clinician-patient experience.
Chapters 3 and 4 both emphasize the need for "self-care" on the part of both clinicians and patients. Chapter 3 discusses the emergence of patient self-care and societal changes in which "patients have increased health consciousness, autonomy, and independence" (p. 45). These change emerged both from cultural shifts -increased societal appreciation for health-improving activities -and also from the impact of the health care market and considerations of individual responsibility and cost. This patient self-care, according to Buetow, must acknowledge the interrelationships among individual capabilities and health-related knowledge, patient values, and the potential for dissonance between an individual's self-caring activities and the impact of those activities on their feelings of being healthy.
Chapter 4 extends this discussion of self-care to health care providers, something that is often ignored given that the "… bright light of patient-centred care casts a vast shadow over clinician welfare" (p. 61). Buetow highlights the sometimes extreme stressors at work in a health care setting, including increased patient loads and working hours, occupational hazards, patient criticism, and challenges to health care provider authority and expertise produced by increased patient access to health information, all potentially contributing to provider burnout and feelings of dissatisfaction. Health care providers' responses to the challenges vary and include coping methods such as increasing self-identification with the caring process, adopting a relentlessly positive attitude toward their profession, focusing on individual development, and attempting to strike a genuine work-life balance. Other coping mechanisms are less positive such as attempting to maximize profit through extended working hours or avoiding pressures or conflict. Important here is this continued theme of the importance of recognizing the personhood of clinicians, their understanding of the work context and their roles and the ways in which their values and knowledge shape that understanding.
Chapter 5 continues this theme, introducing a rather unique perspective of the caring process, focusing on the patient's care of the clinician. Beutow posits that all individuals need care, including those who are enacting a role of the professional caregiver in a work setting. Clinicians, "[t]hrough self-abnegation, … forego the kindness that, as human beings and vulnerable clinicians, they need" (p. 81). These health care providers shun feelings of individual need while engaged in the health care process both given the professional emphasis on this as a value and to create detachment from patients. Changing this, as Buetow notes, requires a shift in the roles of both patient and clinicians and in the feelings and values related to each; patients must be open to considering how they can engage in caring and health care providers must be open to patient expressions of care and gratitude. For clinicians, this may be the most professionally and culturally challenging aspect of this book.
Chapters 6 and 7 provide additional depth for Buetow's central argument that a shift from "patient-centred" to "person-centred" health care is important, and that individuals involved in caring processes must embrace necessary changes in values related to this change. Chapter 6 provides Buetow's conceptualization of personhood and relationships that recognize this personhood, as well as personalized medicine and more comprehensive humanistic models of human caring. Perhaps most importantly, Buteow suggests that those involved in health careand those studying these professions -must recognize concepts of personhood and the need for a shift to person-centered care as an explicit part of what they do. Chapter 7 outlines the necessary values and virtues that go along with this shift. Buetow notes that "[k]ey virtues include justice and caring -in order to respect through joyful, reciprocated caring the value of the moral equality of all persons -in good faith and inter dependently" (p. 137). This moral equality is enacted through mutual trust and reciprocal caring and attention to individual authenticity in enacting roles.
Finally, Chapter 8 provides a practically focused discussion of how implementing personcentered care would work. Buetow's treatment of the necessary changes to individuals and systems here is comprehensive, and includes individual moral development from childhood, changes to medical education, improved understanding of character, and an improved understanding of how new technologies have an impact on these in a health care setting. Implementing person-centered care, then, "… requires producing social conditions conductive to clinicians and patients developing and exercising virtues and good character" (p. 161).
Buetow's treatment of the concept of person-centered care is conceptually well developed, engaging, and highlights a necessarily personalized view of health care. Losing sight of the people-centered nature of these services does a disservice to both patients and clinicians. The individual, relational, and cultural changes necessary to support this shift are, however, truly impressive and are likely to be overshadowed by broader systems-level discussions that are more concrete and less ethereal. These difficulties, though, should not dissuade patients or clinicians from engaging in activities that might support an enhanced view of personhood in health care. The impacts of these efforts are likely to be felt and appreciated.

Disclosure Statement
The author declares there are no conflicts of interest that relate to the research, authorship, or publication of this article.
One of the most popular interventions to address perceived deficiencies in bureaucratic functioning in recent decades has been the reinvention movement. Beginning with exhortations by Osborne and Gaebler (1992) to transform public management through market-based processes and "steering rather than rowing," the reinvention movement gained worldwide attention. In the United States, Vice President Al Gore led a reinvention-based effort to change the way federal agencies were managed. Beginning as the National Performance Review and later titled the National Partnership for Reinventing Government, the Clinton Administration released the group's first report, From Red Tape to Results: Creating a Government That Works Better and Costs Less in late 1993 (Gore 1993). Consistent with Osborne and Gaebler's recommendations, the report recast the role of citizens into customers and suggested ways that government agencies could manage better with fewer resources by adopting market-based practices.
While many academics failed to embrace reinvention as a serious public management concept, others did. Observing the adoption of reinvention principles in Western democracies throughout the world, scholars who took reinvention seriously developed its academic face, the New Public Management. As with the original reinvention movement, the New Public Management relied on market-based principles and the concept of citizen as customer (Kettl 2000).
Fast forward to 2017 and the Trump Administration: Calls to reform government agencies and make public management more business-like are being made with new vigor. This vigor has already resulted in the creation of a new White House Office of American Innovation. Only time will tell what kind of impact will be generated by the current administration's bid to change the federal bureaucracy.
Into the often-heated atmosphere of dissatisfaction with American bureaucracy enters long-time public management scholar Donald Kettl with Escaping Jurassic Government: How to Recover America's Lost Commitment to Competence, a cogent plea for government to adapt or die. He warns that, without adaptation through the return of bi-partisan commitment to competence, democratic government itself will be as extinct as the dinosaurs.
In making the case for renewed commitment to competence, Kettl describes the basic paradox that exists between harsh views of government and demands for government action. Despite repeated calls for a smaller federal government, when people want government help, they not only want it, they want it immediately. He provides numerous examples throughout the book, ranging from federal forest fire fighting to collaboration among federal, state, and local governments working to solve the mystery of food-borne illnesses quickly so that widespread health problems will be avoided. The examples bring into sharp focus the need for competence in managing the public's business.

Competence and the Progressive Era
Kettl begins by describing the Progressive Era (1890s through 1920s) and its reform efforts. Reformers from both Republican and Democratic parties successfully advocated for more professionalized management of government, moving away from the patronage system epitomized by the corrupt Tamany Hall in New York City. While members of both parties may have disagreed over what government should do, they agreed that whatever was done should be done in a competent and professional manner. The bipartisan, Progressive commitment to competence made it possible for the United States to successfully weather economic challenges such as the Great Depression and emerge as a world leader in World War II.
The bipartisan compact gradually began to unravel, as Republicans and Democrats increasingly differed with regard to the size and role of government. Confidence in the Progressives' reform strategy and a professionalized civil service and executive budgeting dissipated over time. As policies came to increasingly rely on non-governmental actors for implementation, bureaucratic boundaries blurred. Government by proxy has made it far more difficult to retain accountability, leading to growing distrust in government. By the twenty-first century, Americans were (and are) plagued by highly polarized political parties and gridlock. Kettl makes the case that American government is in need of a re-boot to become Government 2.0.

Kettl's Six Basic Truths
Kettl introduces what he terms "six basic truths," asserting that they "must become self-evident and provide the backbone for genuine leadership" (pp. 15-17). These truths guide the book's focus and are: Much of what constrains the ability to reduce the size of government is demographic. For instance, the graying of America driven by the aging baby boomer generation makes it difficult for conservatives to reduce the size of government. At the same time, the number and complexity of citizen's demands on government is increasing. We expect our government to protect us from threats ranging from Ebola to terrorism. Thus, while the mix of services provided by government may change over time, its overall size will not. Indeed, the size of government as a share of GDP has changed little over the last 50 years.

Government Does Hard Things
Escaping Jurassic Government is filled with examples of just how difficult the work of government is. The author describes how public health workers at all levels of U.S. government traced a 2008 "attack of the killer tomatoes" -when tomatoes were seemingly causing severe stomach distress across America -to a single jalapeno at a food-processing plant in McAllen, Texas.
In another example, Kettl describes the frustration that can result from reliance on privatesector partners in times of crisis. When wildfires started in the area of Bastrop, Texas, in 2011, Rep. Michael McCaul (R-Tex.) demanded that the U.S. Forest Service deploy a DC-10 firefighting plane. Once it arrived in Texas, the plane sat on the runway for 48 hours. The catch was that the DC-10 was not owned by the U.S. government but rather by a private contractor whose crews had logged so many hours that they had to rest for safety reasons before flying again. Other planes (P-3 Orion tankers) also sat idly by while the wildfires burned because the plans were owned by another private company that had not yet completed the required safety training. Further complicating the situation was the fact that 99.9% of the land for which the congressman was demanding assistance was outside the purview of the U.S. Forest Service. Ultimately, the U.S. Forest Service helped because they were asked by the Texas Forest Service. Addressing this public emergency eventually involved federal, state, and local government, as well as private-sector contractors, amply demonstrating the complexity of government action and responsibility.
Other "hard things" we expect of our government include ensuring bridge safety, safe drinking water, airport screening, and investor protection -which is just a sample of what we expect from our government every day. Managing complex government responsibilities with networks that often include different levels of government and nongovernmental actors translates into a profound need to recruit and retain the right people to work in government. Despite the need for a skilled civil service, investment in the civil service has declined since the 1980s. There are some interesting paradoxes in civil service trends, too. For instance, Ronald Reagan ran for president on a platform that included reducing the size of the federal bureaucracy, but it actually grew by 4% during his administration owing to the growth in defense spending, which created a need for more civil servants to manage those contracts. Clinton reduced the size of the federal workforce by more than 350,000, but it grew again under George W. Bush in part because of the federalization of airport screening workers after 9/11. One of the things that did change in the federal workforce as a result of Reagan's approach to government was its composition. Since the 1980s, the federal civil service has moved away from blue-collar employees through privatization but increased its white-collar numbers due to the need for more contract managers throughout government.

Interweaving
In tales of government fraud, waste, and mismanagement, it is easy to lose sight of the fact that much of what government does comes about through interwoven processes involving various levels of government and nongovernment organizations. Kettl highlights the need to understand that three important threads run through these stories.
First, any government problem is much more visible than problems in the private sector. Citizens expect transparency of government, including public meetings, open records, and access to emails. In contrast, the private sector expects (and often gets) the precise opposite.
Second, well-run private-sector organizations have strategies for how much waste they will accept. Sometimes these strategies are about maintaining quality. For instance, McDonald's will throw away any unused fries after seven minutes because, after that, they no longer taste like McDonald's fries. Sometimes private-sector waste occurs as a result of a mistake: The glass covers for several floors' light fixtures for the World Trade Center were thrown out when it was built in the 1970s because they did not fit. Because the private sector is far less transparent than the public sector, it is simply impossible to compare waste or mismanagement between the two sectors.
Third -and perhaps most important -is that government will do things that the private sector cannot or will not do. The complexity of the issues government addresses through public policies and programs makes effective management more difficult. Americans often take for granted the work of the Social Security Administration, which issues more than 59 million checks each month with an error rate so low it barely registers. Highway fatality rates have dropped significantly over the last three decades due to government work at all levels, including that of the National Highway Traffic Safety Administration. Fire deaths also have greatly declined due to the work of local firefighters and stronger building codes.

Combined Truths
Given that government not only does hard things but also increasingly does them through proxies, it is much more difficult to deliver what is expected by citizens. Conservatives favor government-by-proxy, e.g., increasing reliance on third-party actors for policy implementation, because it provides a mechanism through which government can continue to address public demands without increasing the size of government. Liberals favor providing services through third-party actors as a way of expanding government's reach without increasing the size of government.
These combined truths mean that the civil service itself has changed to one with more contract management specialists. This translates into a move away from a civil service with a large proportion of lower-paid government employees to a cadre of well-educated professionals overseeing billions of dollars' in work delegated to third-party actors each year. The move away from direct government provision of services means that it is more difficult to achieve the accountability on which trust in government rests.

Starving the Beast Undermines Performance
Conservative efforts to "starve the beast" began with the Reagan Administration and are predicated on denying public agencies the financial resources required to carry out their mission. This effort to indirectly reduce the size of government by constraining or eliminating the capacity of agencies to get work done has not had the effect desired by those who wish for smaller government. Instead, it has eroded public trust in government by reinforcing public perceptions about the ability of government to get the job done.
This strategy is another departure from the Progressive Era bipartisan commitment to competence. Instead of policy action based on debate of what government does, the emphasis is on the how. Efforts to starve the beast do not lead to less government action but rather undermine government effectiveness.

Failure to Perform: Bad Policy, Bad Politics, and Bad Democracy
Kettl tells us that undermining government's ability to perform does not benefit anyone. He provides examples from recent presidential administrations to bring the point home. Federal efforts during the George W. Bush Administration in addressing the aftermath of Hurricane Katrina will long serve as an example of government failure. Another failure of recent memory, the Obamacare website rollout, provides more evidence that government competence is a key factor in public trust.
While failures to fulfill citizen expectations for government performance can spell political disaster for elected officials -for instance, George W. Bush's approval ratings never recovered from Katrina -the more important issue is that public trust in government may be irretrievably eroded if the commitment to competence is not restored. By limiting government's ability to adapt in an ever-changing social, political, and economic environment, the move away from bipartisan commitment to competence threatens democracy itself. In Kettl's view, this puts democracy on the same road to extinction as the Jurassic dinosaurs.

Recovering Lost Commitment
The answer to the challenge of Jurassic government, Kettl tells us, is in recovering America's lost commitment to bipartisan competence in public management. Using Federalist No. 70 as a touchstone, the author quotes Hamilton: "a government ill executed, whatever it may be in theory, must be, in practice, a bad government." Further, Kettl states, "Making policy decisions but undermining the ability of them is, quite simply, dishonest government" (p. 177).
The biggest challenges of governance require adapting government institutions and processes to new problems, without cutting ties with a nation's values. In the case of American democracy, this means improving the government's ability to perform in response to citizens' expectations without excessive constraint of basic freedoms.

Conclusion
Kettl closes by citing three lessons. First, government adaptation has already begun to take place through efforts to manage roles and responsibilities interwoven with third-party actors by effective use of evidence and data. Second, there is no need to simply take it on faith that necessary adaptation is taking place. Skilled leaders inside and outside government have already demonstrated how to make the interwoven activities of governmental and non-governmental organizations work. Third, successful government is possible. This will require government leaders with the instinct to lead, i.e., leaders who will build a strong public service and strong information systems to manage an increasingly interwoven public sphere.

Disclosure Statement
The author declares there are no conflicts of interest that relate to the research, authorship, or publication of this article.