Citizen Attitudes Towards Public–Private Partnerships

Abstract

This study examines the factors that influence public attitudes toward public–private partnerships (PPPs) through an analysis of public opinion data collected in 2014. Although previous literature has examined public attitudes toward government contracting and asset privatization, there is little understanding of how the public feels about more collaborative forms of public–private interaction. Counter to previous studies that suggest that support for free enterprise and a disdain for government increases support for private involvement in public services, we find that attitudes toward PPPs are nuanced: Respondents favor them not only when they have positive feelings toward the business sector but also when they also report trust in government. PPPs are thus perceived not as a replacement to public administration, but as a delivery model that demands competence and trust of both public and private partners. The results also explain a previously unstudied relationship between respondent familiarity with PPPs and their attitudes toward them. Counter to expectations, we find that the more familiarity that respondents have with PPPs, the more likely they are to view them favorably. We also identify factors that predict public opinions of PPPs which can inform public outreach and public involvement programs involved with PPPs.

Keywords

public–private partnerships public opinion state and local government

Introduction

Public–private partnerships (PPPs) are part of government trend toward engaging the private sector in hybrid, collaborative partnerships between public and private partners, marking an important deviation from government contracting and full-scale privatization (Geddes & Wagner, 2013; Klijn & Koppenjan, 2016). Although prior research has examined public attitudes toward government contracting and asset privatization (Battaglio & Legge, 2009; Thompson & Elling, 2000), we have little understanding of how the public feels about emerging modes of public–private partnership. The issue deserves attention given increasing government interest in PPPs, and the consequences associated with failing to adequately account for public perceptions of public and private roles in PPPs.

The study of PPPs and public opinion brings attention to the ways that public leaders gauge public preferences in policy formation (Monroe, 1979; Stimson, Mackuen, & Erikson, 1995), and how they link public opinion to public policy (Barabas, 2006; Page & Shapiro, 1983). Over the past 30 years, the public sector has often failed to adequately understand and respond to public opinions of PPPs. Within the United States, and in international contexts ranging from China to the United Kingdom, there are examples of PPPs that were created by governments only to be canceled as a result of public opposition and resistance (Chen, Hubbard, & Liao, 2013; Ni, 2012; Zhang & Ali Soomro, 2016). Central to these failures were questions of how well government leaders understood the interests of their constituents, and how well they identified segments of their communities who found fault with PPPs. An improved understanding of public attitudes toward PPPs and the factors that influence these views are thus essential for aligning the development of PPPs with local constituent interests.

The aim of this research is to integrate the literatures of public opinion, public administration, and PPPs to understand three issues: (a) public support for and opposition to government’s use of PPPs, (b) the factors that shape public attitudes about PPPs, and (c) the extent to which perceived familiarity with PPPs influences one’s support for their use. We analyzed public opinion data from 2014 to achieve these goals. The results suggest that attitudes toward PPPs are nuanced: Respondents favor them not only when they have positive feelings toward the business sector but also when they also report trust in government. The results also explain a previously unstudied relationship between respondent familiarity with PPPs and their attitudes toward them. Counter to expectations, we find that the more familiarity that respondents have with PPPs, the more likely they are to view them favorably. We also identify factors for predicting public opinions of PPPs which can inform public outreach and public involvement programs involved with PPPs. We first review the literature to date on public opinion and PPPs. We then describe the methods and analysis, present and interpret the findings, and conclude with a discussion of the study’s implications.

Defining PPPs

Around the world, government leaders are engaging the private sector in more collaborative and hybrid forms of coproduction (O’Leary & Vij, 2012; Poocharoen & Ting, 2015). PPPs are reflective of this trend, offering the potential to integrate the private sector beyond arms-length transactions, to develop public projects where both the public and private sectors have a stake in their success (Boyer, Van Slyke, & Rogers, 2016; Lohmann & Rötzel, 2014). The term “public–private partnership” has been applied to wide range of government collaborations with the private sector ranging from urban renewal and economic development to public policy networks (Hodge & Greve, 2010). It is in the area of infrastructure services where the most complex, high-cost, and integrated partnerships have emerged. For the purposes of this study, we focus on PPPs utilized for the development of physical infrastructure, a definition Hodge and Greve (2007) refer to as PPPs as long-term infrastructure contracts and which is often referred to as “P3” within the North American context (Federal Highway Administration, 2017; Vining & Boardman, 2008).

The defining characteristic of infrastructure PPPs is the integrated risk-sharing agreements that hold both public and private partners to account for the project’s success. Typical approaches to infrastructure development limit the roles of a private partner, assigning them to the development or renovation of a facility that the government owns and operates. Much of the U.S. highway system, for example, is built and renovated through construction contracts. This approach is largely transactional, and the public sector bears the primary responsibility and risk for the delivery of services on the constructed facility. PPPs are policy tools in which government delineates a larger share of risks to a private partner—beyond construction—into areas such as financing, design, and operations for the development of a public asset.¹ These transfers of risk represent important differences from the more controlled approach of government contracting. PPPs therefore occupy a spectrum (Ke, Wang, Chan, & Cheung, 2009; Li, Akintoye, Edwards, & Hardcastle, 2005) or continuum (Garvin, 2010; Hodge & Greve, 2007) of structural arrangements, and degrees of risk delegation.

PPPs are also different from full-scale infrastructure privatization. In privatization, ownership of a facility is transferred to a private partner, and the government’s role is reduced to that of regulation. Public ownership is retained in PPPs and the public sector maintains some fundamental responsibilities and risks, which, at minimum, involve their remuneration and/or monitoring of partner performance. The privatization of infrastructure has taken place in a host of industrial areas following regime changes in the Soviet Union and Eastern Europe, and following large-scale government reforms in the United Kingdom and Australia (Savas, 2000).

Public Attitudes and PPPs

The distinctions between PPPs, contracting, and privatization frame our investigation of the predictors of public attitudes toward PPPs. Although prior research has examined public opinion of government contracting (Thompson & Elling, 2000) and asset privatization (Battaglio & Legge, 2009; Legge & Rainey, 2003), there is little understanding of how the public feels about PPPs, or how they come to these conclusions. We draw from literature on public opinion, government contracting, and privatization to identify factors that have predicted public opinion of related market-based reforms, and we examine their influence within the context of PPPs.

The public is likely to be more critical of PPPs than they are of contracting or privatization. Citizens are typically comfortable with private contracts that are “simple,” or easy to specify and easy to observe and measure, such as those for landscape services or refuse collection (Thompson & Elling, 2000). Citizen enthusiasm can wane when considering private roles in more complex services, such as the development and operation of physical infrastructure (highway systems, bridges, subways, light rail mass transit, etc.). Given the fact that any private sector role in government services can be complicated for the average person to understand (because of questions about cost, quality, accountability, and so on; Battaglio, 2009; Legge & Rainey, 2003), there is good reason to expect that as products or services provided by a private partner become more complex, public support for the initiative will decrease. There is also increasing evidence of public opposition of PPPs (Zhang & Ali Soomro, 2016), within the United States and in other parts of the world (Chen et al., 2013). We can expect that public opinions of PPPs reflect this trend, and that large portions of the population will be skeptical of them.

In this study, we consider public attitudes toward three issues related to PPPs: (a) the potential of PPPs to benefit their communities, (b) the expertise of the public sector to manage PPPs, and (c) the perceived responsiveness of the public sector in a PPP.² These issues address public attitudes toward increasing private involvement in policy areas typically administered by government, as well as attitudes toward the public sector when public and private roles are shared. These attitudes address tensions in the literature about whether PPPs can benefit a community more than what the government could conduct on its own (see, for example, Savas, 2000), whether or not the public sector has the “capacity” (or expertise) needed to manage a complex public–private transaction such as a PPP (Van Slyke, 2003), and whether the public sector maintains a level of responsiveness to citizens when services are delivered by an outside provider (Bingham & O’Leary, 2008).

Predicting Public Attitudes of PPPs

The second goal of this study is to explain how people form attitudes about PPPs. These factors are outlined to frame our exploratory study of public opinion of PPPs. First, public opinion research indicates that women often take more liberal political positions (Wilcox, 1991) and they are often more “pro-government” (Schlesinger & Heldman, 2001). Stronger support for liberal and government positions may lead them to be more resistant to the government transfer of responsibilities to the private sector. Older adults are also more reliant on pensions and related government support, and may therefore be skeptical of the private sector’s role in infrastructure delivery. Persons with higher levels of education have more access to media and information (Alvarez & Brehm, 1998), which could lead them to better understand the deleaneated role of a private partner. This improved understanding could position them to be more accepting of private roles in governemnt services, such as PPPs. Persons of lower incomes are likely to be skeptical of the private sector involvement in public services, as they may be more vulnerable to decreases in public services (Battaglio, 2009; Durant & Legge, 2002).

Prior literature also suggests that opinions toward the private sector can be influenced by respondent predispositions toward political parties and their perceptions of business and government. Political party identification has been found to influence citizen views of infrastructure privatization (Dubin & Navarro, 1988; Rudolph & Popp, 2009), and self-identified Republicans tend to be more supportive of private roles in government services as they typically have the strongest connections with the business community. Another factor is approval for one’s governor, given the high publicity of PPPs and the evidence that citizens often align with political leaders they respect (Dunleavy & Husbands, 1985). Governors are often the most recognized political champion for this type of state or local government-led project.³

The extent that respondents trust civil servants may also have an independent effect on perceptions of PPPs. Multiple studies indicate a decrease in public trust of government in the United States and around the world (Newell, Reeher, & Ronayne, 2008; Nye, Zelikow, & King, 1997). There is some evidence that public trust in government reduces support for other market-based reforms (Battaglio & Legge, 2009). The way that PPPs are typically promoted is based on assumptions about government failures in administration (monopolized services, bureaucratic inefficiency, etc.), and the potentials for the private sector to correct these inefficiencies through market discipline and innovation (Garvin, 2010; Grimsey & Lewis, 2007; Savas, 2000). These messages are similar to related efforts to reform public management through market principles (Hood, 1991; Osborne & Gaebler, 1993), and involve underlying assumptions about bureaucratic administration. The more distrust that people have in government, the more amenable they are likely to be to a private partner assuming public responsibilities.

Personal views about the appropriate role of the government in the economy can also influence how one feels about private roles in public services (Feldman, 1988; Heath & Gifford, 2006). Support for free enterprise, for example, generally demonstrates faith in market principles, and a preference for business over government. Citizen views toward the federal deficit can also influence support for government reform. A common argument in favor of PPPs is the potential for leveraging private equity and finance for public works projects, which transfers greater borrowing risk to private partners (Garvin, 2010). Fiscal stress has in fact been a predictor of government policy decisions in favor of privatization (Bel & Fageda, 2008; Lopez de Silanes, Shleifer, & Vishny, 1997) and, citizens, similarly may be more inclined toward alternative forms of delivering infrastructure projects when they are more concerned about government deficits.

Perceived Familiarity With PPPs

The third goal of this study is to examine respondent familiarity with PPPs, and the extent that their familiarity influences their opinions of PPPs. This involves an examination of the knowledge respondents have about local policies (Clawson & Oxley, 2013; Goldenberg, Butani, & Phipps, 1993), which provides some indication of how well-informed they are about the subject of the study. Questioning respondent familiarity with PPPs complements attitudinal and demographic information about respondents and allows for comparison against other sources of data (Ansolabehere & Rivers, 2013; Lee, Benoit-Bryan, & Johnson, 2012; Swidorski, 1980). The premise that a PPP will be more effective and efficient than public sector delivery is a landscape with contradictory and inconclusive evidence (Hodge & Greve, 2007). Respondent familiarity with PPPs can provide some indication of whether their knowledge of PPPs improves or diminishes their attitudes toward them. This provides us with some understanding of how the uneven track record of PPPs in the United States affects public perceptions about them.

Research Questions

This review of literature informs our development of a three-part research question:

Research Question 1: How do U.S. citizens feel about PPPs for infrastructure services (PPPs)?

Research Question 2: How do they arrive at these points of view?

Research Question 3: Are their views shaped by their perceived familiarity with PPPs?

Given our knowledge of PPPs and this review of research on related market reforms, we have expectations about the factors that could influence attitudes toward PPPs. For one, we expect that men with higher levels of education and higher incomes will be more supportive of PPPs. Second, we expect that political party id will influence attitudes, such that persons who identify as Republicans will be more supportive of PPPs. Third, we expect that persons who view PPPs favorably care about financial issues (they are concerned about the deficit and free enterprise) and that they have negative views toward government (they have little trust in civil servants and low approval of their governors). Finally, we expect that, given the uneven track record of PPPs within the United States and around the globe, the more familiarity that respondents have with them, the more negatively they will view PPPs. We specify these expectations to frame our analysis, but given the fact that we are studying a context that has not been studied before, we take a largely exploratory approach to these research questions.

Data and Method

The data for this study were collected in the fall of 2014 through the Cooperative Congressional Election Study (CCES) at Harvard University (N = 1,000). The survey questions analyzed here were purchased by our research team, and were included in a larger survey that asked respondents about issues reflecting legislative priorities of the U.S. Congress. YouGov/Polimetrix administered the survey to an Internet-based sample. The respondents were drawn from a proprietary opt-in survey panel of U.S. residents who have agreed to participate in YouGov’s web surveys. The survey was administered to approximately 20% more persons than analyzed here. The sample was reduced to 1,000 respondents through the process of matching by gender, age, race, education, partisanship, ideology, and political interest in respect to a frame constructed by the Census Bureau’s 2008 American Community Survey. This sample closely approximates the demographic makeup of the known general population of the United States, and studies of CCES show that the method yields a sample representative of the national public on a variety of observable characteristics (Ansolabehere & Rivers, 2013). Prior research on market-based reforms examined public opinion data from a single U.S. state (Thompson & Elling, 2000) or from populations from outside the United States (Battaglio, 2009; Battaglio & Legge, 2009; Durant & Legge, 2001, 2002; Legge & Rainey, 2003). Although informative, this prior research could be built upon through analyses of a more representative sample of U.S. residents.

The focus of this study is on perceptual data, which has been shown to be important in advancing theory development (Kalleberg, 1969; Ketokivi & Schroeder, 2004; Lincoln & Zeitz, 1980). Respondents were asked 120 questions about their political preferences, and their views toward market and government relations in the United States and their opinions of a range of public policy issues. A section of the survey also asked respondents about their views toward PPPs and their familiarity with them. The survey questions were developed from literature reviews regarding PPPs, public opinion, and public administration. The survey language is provided in Appendix A. We also analyzed a measure of state deficit to examine environmental factors that may have influenced respondent attitudes.

The survey language explaining PPPs was crafted to direct respondents to projects related to infrastructure development (as opposed to PPPs in other areas, like urban regeneration), and to explain infrastructure PPPs in ways that their local representatives were likely to communicate them to relevant constituencies. The preamble was pilot-tested with practitioners to align it with definitions of infrastructure PPPs used in the field (Dillman, Smyth, & Christian, 2008).

Results

Public Attitudes Toward PPPs

The first stage of the analysis inventoried public opinions of PPPs. Figure 1 shows respondent attitudes toward PPPs, with substantial variation across the three measures (full descriptives in Appendix B). Before controlling for any demographic factors, predispositions, or perceived familiarity, we can see that respondents were more inclined to support the first statement, then the third, and then the second. In other words, respondents were generally most optimistic about the potentials for PPPs to benefit their communities, with 54.46% of the respondents agreeing with this statement, and only 6.5% disagreeing with it. Respondents were also rather optimistic about the potential for the public sector to maintain responsiveness to the public when a large-scale public service is delivered by a private partner. In all, 43.56% of respondents agreed that public officials are receptive in PPPs, and only 13.96% of respondents disagreed with this statement. This result suggests that the U.S. public is less concerned about their direct service provider being a private entity rather than a public entity—at least in respect to government responsiveness.

Figure 1.

Public attitudes toward PPPs.

The respondents were far more critical of the public sector’s expertise or capacity for managing a PPP. Only 28.03% of respondents agreed that the government has the skills necessary to manage this kind of a contract with the private sector, and nearly the same amount, 21.17% disagreed that the public sector had the know-how for this type of work. Given the mounting evidence of the requirements for the public sector to develop adequate expertise for managing PPPs (Robinson, Carrillo, Anumba, & Patel, 2010), it is important to note that respondents expressed concerns in this area.

These results demonstrate that U.S. residents do in fact have significant opinions about PPPs (more than half of the respondents either agreed or disagreed with the statements for the three PPP attitudes), and they generally viewed the potential benefits of PPPs more highly than the public sector’s potential for managing them. A large share of the respondents, however, only expressed opinions of “indifferent” to the three attitudinal questions asked of them. Despite the mounting evidence of public opposition to PPPs (Chen et al., 2013; Ruiz, Elliott, & Ratcliffe, 2009), we can see that public concern with PPPs is rather limited (40% or more of respondents expressed indifference toward the three attitudes toward PPPs), and the negative opinions of them were limited to issues related to government expertise and responsiveness.

Respondent Familiarity With PPPs

We also asked respondents about their familiarity with PPPs conducted in their respective states (Table 1). These survey questions involved quantitative measures of respondent knowledge of local policies, and allow us to compare their responses against other points of data (Ansolabehere & Hersh, 2012; Lee et al., 2012). To compare their responses with PPPs that were conducted in their home states, we consulted three sources of information. Data were first collected through our subscription to the P3 Major Projects Database compiled by the Public Works Financing Newsletter (Public Works Financing, 2017). This includes 3,850 P3 projects that we sorted by year, state, infrastructure sector, and risk transfer configuration. We then limited the count of projects to those either in development or in operation at the time of the survey (fall of 2014),⁴ involving the private sector in development or renovation, and allocating significant risks across public and private partners. This information was collected at the state level, and by each of the five infrastructure types considered in this study (transportation, water, housing, energy, and telecom). These data were also triangulated with two open-access databases on U.S.-based PPPs during this timeframe (InfraPPP, 2017; P3Bulletin, 2017).

Table 1.

Respondent Reported Familiarity by Total Number of PPP Types.

	Perceived familiarity		Validated familiarity		Misreport rate
	n	%	n	%	n	%
No. of PPP types by person
0	163	16.3	245	24.5	693	69.3
1	227	22.7	262	26.2	182	18.2
2	194	19.4	198	19.8	80	8
3	164	16.4	138	13.8	33	3.3
4	92	9.2	76	7.6	7	0.7
5	160	16	81	8.1	5	0.5

Note. PPP = public–private partnership.

Comparing perceived familiarity with the collected data allows us to draw two conclusions about respondents: the extent that they were informed (that their perceived familiarity is validated by projects in their states), and the extent that they were misinformed (that their perceived familiarity was not validated by projects in their states; Clawson & Oxley, 2013).⁵ In all, 83.7% of respondents indicated familiarity with at least one type of PPP in their home state, and 75.5% of respondents indicated familiarity that was validated for at least one of the infrastructure sectors (Table 2). In other words, approximately three fourth’s of our sample reported familiarity with a PPP of an infrastructure type that was in development or operation in their state at the time of the survey. In all, 30.7% of the respondents misreported at least one type of PPP in their home state.

Table 2.

Accuracy of Reporting by Infrastructure Types.

	Transport (%)	Water (%)	Housing (%)	Energy (%)	Telecom (%)
Pr (valid/perceived familiairty)	89	89	77	62	67
Pr (misreported/perceived familiarity)	11	11	23	38	33

We also considered these figures by infrastructure type to determine the extent of misreporting. Respondents were most informed about PPPs in the areas of transportation and water services (89% for each infrastructure type), which can be explained in part by the federal requirements of public involvement and outreach for these forms of public planning. Respondents were less informed about PPPs for energy projects, telecommunications projects, and housing projects (62%, 67%, and 77%, respectively). Respondents were also mostly likely to misreport PPPs for energy projects, telecommunications projects, and housing projects (38%, 33%, and 23%, respectively). The extent that respondents were misinformed about projects in these areas may be explained at least in part by the lower levels of public communications and community outreach that are likely to be conducted in these infrastructure sectors. The validated PPPs for housing projects were also most prominent with the military, an area that affects a lower spectrum of the U.S. population.

Predicting Support for PPPs

The next stage of our analysis examined the demographic characteristics and predispositions of respondents in predicting their levels of support or opposition for PPPs. Our focus was on understanding what leads respondents to “agree” or “disagree” with the three attitudinal questions regarding PPPs. Given that the three outcome variables are limited dependent variables, we estimated probability models to determine the likelihood of respondents choosing among the answer choices (Eliason, 1993; Long, 1997). We are most concerned with the factors that can lead an individual to deviate from a state (or answer) of “indifference” to one of “agreement” or “disagreement” on PPP-related attitudes. We are interested in the factors that lead respondents to form one opinion or another. Therefore, we analyzed the probability of respondents choosing the response “agree” or “disagree” given the option to remain “indifferent” through multinomial logistic regression models (MNLM; see Note 5).

A test for multicollinearity among the independent variables was acceptable, with all variables demonstrating tolerance levels of at least .37 and a mean variance inflation factor (VIF) of 1.7. A test for independence of irrelevant alternatives (IIA), a Small-Hsiao test, was also acceptable for all models, failing to reject the hypothesis that the outcomes are independent of relevant alternatives. Regression diagnostics indicated that the models were well specified and meet the assumptions of multinomial logistic regression. In each of the models, the likelihood ratio chi-square had a p value < .05, giving us confidence that the model as a whole fits the data significantly better than a model with no predictors. Independent Wald and Loglikehood tests were conducted on individual predictor variables, which confirmed the significant values identified in the full models. The Harman Single Factor test indicated that common method bias was not a problem; it explained only 16% variance among survey items, far below the 50% threshold.

PPPs can benefit my community

Table 3 shows the factors that influenced respondent attitudes about the statement PPPs can benefit my community. We estimated two regressions, considering how perceived familiarity and validated familiarity affected respondent attitudes.⁶ The primary predictors included gender, age, trust in government, governor approval, political party id (Republican and Independent), support for free enterprise, views of the deficit, and perceived familiarity with PPPs.

Table 3.

MNLM: PPPs Can Benefit My Community.

	Model 1		Model 2
	Disagree	Agree	Disagree	Agree
Gender	−0.745*	−0.203	−0.758*	−0.214
	(0.328)	(0.159)	(0.328)	(0.159)
Age	0.016	0.017*	0.017	0.018*
	(0.010)	(0.005)	(0.010)	(0.005)
Education	0.054	0.084	0.054	0.084
	(0.116)	(0.057)	(0.116)	(0.057)
Family income (low)	−0.804	0.449	−0.776	0.479
	(0.482)	(0.269)	(0.482)	(0.269)
Family income (high)	−0.157	0.472	−0.135	0.497
	(0.448)	(0.266)	(0.448)	(0.266)
Trust in civil servants	−0.631*	0.142	−0.616*	0.155
	(0.210)	(0.104)	(0.208)	(0.104)
Governor approval	−0.299*	0.040	−0.298*	0.040
	(0.149)	(0.073)	(0.149)	(0.073)
Democrat	−0.001	0.422	0.003	0.412
	(0.498)	(0.278)	(0.498)	(0.278)
Republican	−0.224	0.647*	−0.226	0.632*
	(0.555)	(0.286)	(0.555)	(0.286)
Independent	−0.396	0.616*	−0.389	0.611*
	(0.512)	(0.275)	(0.512)	(0.275)
Support free enterprise	−0.272*	0.186*	−0.271*	0.189*
	(0.130)	(0.068)	(0.131)	(0.068)
Worry about deficit	0.071	0.166*	0.070	0.165*
	(0.156)	(0.079)	(0.157)	(0.079)
Want to expand deficit	−0.003	−0.134*	−0.007	−0.139*
	(0.103)	(0.051)	(0.103)	(0.051)
State deficit per capita	0.007	−0.076	−0.042	−0.141
	(0.565)	(0.279)	(0.567)	(0.281)
Perceived Familiarity	0.139	0.156*
	(0.098)	(0.049)
Validated familiarity			0.130	0.163*
			(0.102)	(0.053)
_cons	0.609	−3.063*	0.615	−3.068*
	(1.342)	(0.728)	(1.341)	(0.728)
Final log likelihood		−656.4		−656.7
χ²(30)		110.5*		109.8*
N		821		821
Pseudo R²		.15		.15

Note. MNLM = multinomial logistic regression models; PPP = public–private partnership.

p < .05.

The negative coefficient for gender indicates that men were more likely to disagree with the potential benefits of PPPs (women were less likely to be in this disagree category). The positive coefficient for age indicates that older persons were more likely to agree with the benefits of PPPs. The negative coefficients for trust in civil servants and governor approval in the disagree column indicate that as government trust and governor approval decrease, respondents were more likely to view PPPs negatively. Persons identifying as Republicans or Independents were more likely to agree with the potential benefits of PPPs, and support for free enterprise predicts both disagree and agree answer choices. Perceptions about the deficit also predict answer choices; the more worried respondents were about the deficit, the more likely they were to agree with the benefits of PPPs. The less they wanted to expand the deficit, the more likely they were to agree with PPP benefits. We also note that perceived familiarity with PPPs positively influenced respondent attitudes. The more validated familiarity respondents had with PPPs, the more likely they were to view them positively.

Our first observation is that the greater the trust that an individual had in government and the greater one’s approval of one’s governor, the lower the likelihood of disagreeing with the notion that PPPs could benefit their communities. Support for free enterprise positively influenced the likelihood of agreeing that PPPs can benefit one’s community. This is expected as preferences for private sector involvement in public services can be influenced, in part, by one’s faith in free enterprise. The independent effects of perceptions of government and free enterprise in this context suggest that it is not only positive views toward the private sector that influence this outcome but also perceptions about government.⁷

As the coefficients in the probability estimation are difficult to interpret on their own (Eliason, 1993; Long, 1997), we also estimated the predicted probabilities of a one unit change in each predictor variable, when holding all other continuous variables at their means. This is the estimation of first differences (Table 6). These estimates explain the extent that a one standard deviation change in the value of an independent variable affects the probability of agreement or disagreement with the outcome, all else equal.

For example, when a respondent’s political party id is Republican or Independent, the probability of agreeing that PPPs can benefit one’s community increases by 15%, all else equal. A change in a respondent’s view of support for free enterprise from “neither agree nor disagree” to “agree strongly” increases the probability of agreeing that PPPs can benefit their community by 10%, all else equal.

Government has PPP expertise

The results of the second MNLM (Table 4) show the factors that influenced respondent attitudes toward the government’s experience and expertise for managing PPPs. We can see that trust in government, approval of one’s governor, political party id (Democrat), perceptions of the deficit, and familiarity with PPPs influenced respondent feelings about this issue (Table 4). Age and education are also significant across the models, but due to their influence on both outcomes, we conclude that older and more educated persons were more likely to have extreme opinions on this topic (older and more educated persons were not “indifferent”).

Table 4.

MNLM: Government Has Expertise for PPP.

	Model 1		Model 2
	Disagree	Agree	Disagree	Agree
Gender	−0.392	−0.174	−0.392	−0.189
	(0.202)	(0.176)	(0.202)	(0.175)
Age	0.032*	0.010	0.032*	0.011*
	(0.006)	(0.005)	(0.006)	(0.005)
Education	0.182*	0.168*	0.181*	0.168*
	(0.071)	(0.063)	(0.071)	(0.063)
Family income (low)	−0.256	0.476	−0.250	0.503
	(0.317)	(0.320)	(0.317)	(0.319)
Family income (high)	0.006	0.223	0.010	0.249
	(0.309)	(0.316)	(0.308)	(0.315)
Trust in civil servants	−0.547*	0.197	−0.548*	0.209
	(0.131)	(0.118)	(0.130)	(0.118)
Governor approval	−0.337*	−0.028	−0.338*	−0.030
	(0.092)	(0.080)	(0.092)	(0.080)
Democrat	−0.607	0.905*	−0.614	0.892*
	(0.350)	(0.341)	(0.350)	(0.341)
Republican	−0.058	0.511	−0.063	0.496
	(0.328)	(0.356)	(0.328)	(0.356)
Independent	0.179	0.337	0.174	0.330
	(0.310)	(0.352)	(0.310)	(0.352)
Support free enterprise	0.105	0.098	0.104	0.100
	(0.086)	(0.074)	(0.086)	(0.074)
Worry about deficit	0.125	0.198*	0.125	0.199*
	(0.099)	(0.086)	(0.099)	(0.086)
Want to expand deficit	−0.035	−0.110	−0.035	−0.115*
	(0.064)	(0.057)	(0.064)	(0.057)
State deficit per capita	0.012	0.066	0.006	0.016
	(0.335)	(0.310)	(0.336)	(0.309)
Perceived familiarity	0.014	0.145*
	(0.061)	(0.054)
Validated familiarity			0.024	0.138*
			(0.066)	(0.057)
_cons	−1.199	−3.815*	−1.204	−3.796*
	(0.867)	(0.830)	(0.867)	(0.830)
Final log likelihood		−767.6		−768.4
χ²(30)		173*		171.5*
N		812		812
Pseudo R²		.22		.22

Note. MNLM = multinomial logistic regression models; PPP = public–private partnership.

p < .05.

Government trust, again, is found to influence attitudes toward PPPs. In this case, it is negatively associated with disagreeing that government has expertise to manage PPPs. Similarly, governor approval is negatively correlated with the “disagree” answer choice; the less respondents approve of their governors the more likely they are to find fault with government expertise in PPPs. These relationships indicate that confidence in government abilities and expertise is likely to be influenced by one’s trust of and approval of them. We also see that persons identifying with the Democratic party were often more supportive of the public sector, and this relationship may be explained in part by the confidence that persons identifying as Democrat have in the public sector.

We also note that perceptions about the deficit influenced opinions about PPPs. The more respondents were concerned about the deficit, the more likely they were to agree that government has expertise to manage PPPs. This finding is similar to the regressions performed in Table 3. Interestingly, we note that only the perceptual measures related to deficit predict these attitudes; the rates of state deficit have no impact on this outcome. Validated familiarity with PPPs is also positively associated with “agreeing” that government can serve this role.

A review of the first-difference predicted probabilities explains the extent that these factors influence public attitudes (Table 6). Changing one’s trust in government from “some” to “a great deal,” decreases the likelihood of a respondent disagreeing with this outcome by 9%, all else equal. Reporting approval for one’s governor from “somewhat approve” to “strongly approve” decreases the likelihood of disagreeing by 6%, all else equal. Identifying as a Democrat decreases the likelihood of disagreeing by 12%, and increases the likelihood of “agreeing” by 18%, all else equal.

Public officials are receptive on PPPs

The results of the third MNLM (Table 5) show the factors that influenced respondent attitudes toward the government’s potential for maintaining responsiveness in a PPP. We can see that trust in civil servants, political party id (Democrat), and perceived familiarity with PPPs influenced respondent feelings about this issue. We also note some influence of education, such that persons with higher levels of education are more like to agree with this statement.

Table 5.

MNLM: Government is Receptive in PPP.

	Model 1		Model 2
	Disagree	Agree	Disagree	Agree
Gender	−0.201	−0.130	−0.204	−0.153
	(0.231)	(0.164)	(0.231)	(0.163)
Age	0.008	0.000	0.009	0.001
	(0.007)	(0.005)	(0.007)	(0.005)
Education	0.134	0.128*	0.132	0.129*
	(0.082)	(0.058)	(0.082)	(0.058)
Family income (low)	0.291	0.401	0.301	0.439
	(0.359)	(0.278)	(0.359)	(0.278)
Family income (high)	0.045	0.456	0.048	0.490
	(0.363)	(0.273)	(0.363)	(0.272)
Trust in civil servants	−0.517*	0.332*	−0.520*	0.346*
	(0.146)	(0.109)	(0.146)	(0.109)
Governor approval	−0.124	0.124	−0.124	0.124
	(0.105)	(0.074)	(0.105)	(0.074)
Democrat	−0.485	0.649*	−0.498	0.635*
	(0.378)	(0.304)	(0.378)	(0.304)
Republican	−0.099	0.614	−0.116	0.597
	(0.363)	(0.313)	(0.363)	(0.313)
Independent	0.078	0.540	0.067	0.534
	(0.337)	(0.306)	(0.337)	(0.306)
Support free enterprise	0.014	−0.081	0.014	−0.078
	(0.101)	(0.069)	(0.100)	(0.069)
Worry about deficit	−0.121	0.044	−0.124	0.044
	(0.112)	(0.081)	(0.112)	(0.081)
Want to expand deficit	0.040	−0.006	0.039	−0.014
	(0.071)	(0.052)	(0.071)	(0.052)
State deficit per capita	0.092	−0.110	0.050	−0.189
	(0.392)	(0.288)	(0.392)	(0.290)
Perceived familiarity	0.050	0.194*
	(0.070)	(0.050)
Validated familiarity			0.087	0.185*
			(0.076)	(0.053)
_cons	−0.179	−2.851*	−0.197	−2.819*
	(0.973)	(0.750)	(0.973)	(0.748)
Final log likelihood		−758.9		−760.6
χ²(30)		126*		122.7*
N		820		820
Pseudo R²		.17		.16

Note. MNLM = multinomial logistic regression models; PPP = public–private partnership.

p < .05.

Government trust, again, is found to influence attitudes toward PPPs. In this case, the more trust that a respondent reported, the lower their likelihood of disagreeing with this statement, and the higher the likelihood of agreeing with the statement. The positive association between Democrats and the “agree” answer choice indicates that when persons identified as Democrats, they were more likely to agree that government is responsive in PPPs. This finding is similar to findings in the previous models (Table 4), and indicates that persons that identifying as Democrats are more likely to support government roles in PPPs. The positive relationship between validated familiarity with PPPs and the “agree” answer choice indicates that increased familiarity with PPPs improves their opinions of government responsiveness in this context.

Consulting Table 6, we can see the effect of these perceptions on attitudes related to government responsiveness in PPPs. When a respondent’s view toward civil servant trust changes from “some” to “a great deal,” the likelihood of disagreeing with this statement decreases by 7%, and the likelihood of agreeing with this statement increases by 9%, all else equal. When a respondent identifies as a Democrat, it decreases the probability of “disagreeing” with government responsiveness in PPPs by 8% and it increases the probability of “agreeing” by 16%, all else equal.

Table 6.

Predicted Probabilities From First Differences.

	PPPs can benefit my community		Government has PPP expertise		Public officials are receptive
	Disagree	Agree	Disagree	Agree	Disagree	Agree
Base probability^a	9%	31%	24%	15%	16%	25%
Gender
Female > male	8%	1%	7%	1%	2%	2%
Age
49 > 66	2%	6%	10%	0%	2%	0%
Education
2-year college > postgraduate	0%	2%	3%	1%	1%	2%
Family income (low)
>less than US$50k	−5%	13%	−6%	8%	2%	9%
Family income (high)
>more than US$50k	−2%	12%	−1%	3%	−2%	10%
Civil servant trust
Some > a great deal^b	−4%	5%	−9%	5%	−7%	9%
Governor approval
Somewhat approve > strongly approve	−2%	2%	−6%	0%	−2%	3%
Not sure > Democrat	−1%	9%	−12%	18%	−8%	16%
Not sure > Republican	−3%	15%	−3%	7%	−4%	13%
Not sure > Independent	−4%	15%	2%	4%	−2%	11%
Support free enterprise
Neither agree nor disagree > agree strongly	−4%	10%	3%	2%	1%	−3%
Worry about deficit
Neither more nor less > somewhat more	0%	7%	3%	4%	−4%	3%
Want to shrink the deficit
Shrink by about 1% > expand by about 1%	1%	6%	−1%	−3%	1%	−1%
Deficit per capita US$–2 > US$285	0%	−1%	0%	0%	0%	−1%
Validated familiarity
Two PPP Types > four PPP Types	1%	3%	0%	1%	0%	3%

Note. Estimations of first differences predict the change in probability of each outcome when the value of a single independent variable is increased or decreased by one standard deviation from its mean, holding all other continuous variables at their mean values. PPP = public–private partnership.

The base probability was estimated for a female, 49 years old, with a 2-year college education, no reported household income, “some” trust in civil servants, who “somewhat approves” of his or her governor, identified as “not sure” for political party affiliation, “neither agreeing nor disagreeing” in supporting free enterprise, “neither more nor less” worried about the deficit, a preference to “shrink the deficit by 1%,” living in a state with an average of –US$2k per capita deficit, and validated familiarity with two PPPs. We only perform these estimations of first differences in respect to validated respondent familiarity with PPPs.

We also estimated the first difference for a change in civil servant trust from 1 (not at all) to 3 (somewhat approve) in the models. For the model estimating that PPPs Can Benefit My Community, this changes the probability that a person will select “disagree” by –19% and the probability that a person will select “agree” by 11%. For the model estimating that Government has Expertise for Managing PPP, this changes the probability that a person will select “disagree” by –26% and the probability that a person will select “agree” by 8%. For the model estimating that Government is Responsive in PPP, this changes the probability that a person will select “disagree” by –22% and the probability that a person will select “agree” by 14%.

Discussion and Conclusion

The first contribution of this study is a more informed understanding of how U.S. residents feel about PPPs. The literature suggests that as the private sector takes on responsibility for more complicated and integrated products and services the public is likely to be more skeptical of their role (Brown, Potoski, & Van Slyke, 2016; Thompson & Elling, 2000). The numerous examples of political controversy, public opposition, and negative public perceptions that have led to PPP cancelations (Chen et al., 2013; Zhang & Ali Soomro, 2016) also leads us to expect that large portions of the U.S. public have negative perceptions of PPPs. Surprisingly, we find that public attitudes toward PPPs are not that negative across a majority of participants. A large share of them have “indifferent” opinions toward PPPs. When respondents do express concerns with PPPs, they are more critical of the role that the public sector plays in the PPP rather than the potential benefit of the project itself. Their greatest concern about the government’s role in PPPs is related to expertise for managing the partnership, echoing a concern long-raised by academics (Robinson et al., 2010).

The second contribution of this study is an informed understanding of the factors that predict public attitudes toward PPPs. Counter to expectations, we find that attitudes toward PPPs are influenced by not only support for free enterprise, but also trust in government. We also find that higher levels of approval for one’s governor, positively affects attitudes toward PPPs. In other words, support for PPPs is not driven solely by a preference for business over government, but with a concern for economic issues and trust in government. The more trustworthy civil servants are, the more accurate the information they provide to the public, and the greater reliability that citizens can have in the services they provide. Support for PPPs appears to be driven less by a “make or buy decision,” (seeking an alternative to government) than by a preference to improve the benefits of markets while ensuring a level of government competency. The results suggest a preference for an active and trustworthy presence of the public sector and a high level of approval of local government officials. We also note that persons identifying as Democrats are more likely to support public sector roles in PPPs, and persons identifying as Republicans and Independents are more likely to support the potential of PPPs to benefit their communities. This indicates that feelings toward the public sector’s role in a PPP may be particularly important for persons who do not identify as Democrats.

The third contribution of this study is an informed understanding of how one’s familiarity with PPPs influences their attitudes toward them. Despite the uneven track record of PPPs around the world (Hodge & Greve, 2007; Jamali, 2004; Siemiatycki, 2010; Siemiatycki, 2011), and the negative press that follows problems associated with PPP disruptions and failures (Laris, 2015; Ruiz et al., 2009), we find that the more PPPs that respondents report familiarity with, the more likely they are to have positive views toward them. On one hand, this implies that the track record of PPPs in the United States, though uneven, has not dissuaded people from considering PPPs as a viable option. On the other hand, these results indicate that persons who are less informed about PPPs are the ones who are less likely to view them positively.

The results from this study provide important recommendations for practice. Much of the promotion of PPPs as a delivery model has echoed arguments of new public management (Barzelay, 2002; Hood, 1991; Osborne & Gaebler, 1993), by extolling the potential for increased private involvement through PPPs to reform the work of public administration (Delmon, 2011; Grimsey & Lewis, 2007; Savas, 2000). This logic informs much of the promotion of PPPs, and the communication strategies surrounding marketing campaigns to “sell” the idea of the delivery model to policy makers and the general public. The results here suggest that highlighting the private sector’s potential contributions to PPP is just one part of the messaging needed to garner public support. The reputation of one’s governor as well as the trust in government are unrecognized, though important, elements of messaging PPPs to U.S. communities.

In other words, public entities interested in developing PPPs should consider public outreach and involvement programs that not only profile the potential benefits of a private partner’s involvement in PPP, but the abilities of public servants and the reputation of public officials in carrying them out. These results also indicate that this messaging and communication of the public sector’s role is particularly important for persons who have little familiarity with PPPs. This is an issue that is likely to be more prominent in jurisdictions that have less experience with PPPs.

One of the limitations of this study is our approach to understanding respondent familiarity with PPP in this study. There are inherent challenges with determining public knowledge of particular policies accurately (Ansolabehere & Hersh, 2012) and the average U.S. resident is often uninformed of even the most basic political knowledge (Delli Carpini & Keeter, 1996; Kohut, Doherty, Dimock, & Keeter, 2012). Measuring respondent familiarity with PPPs captures only a surface-level understanding that respondents have of PPPs. It does not consider the depth of their understanding of this delivery model or the depth of their firsthand involvement with it. Qualitative research could better examine the connections between individual understanding of PPPs and attitude formation by examining respondent views toward specific projects in their communities, and the depth of their understanding of the project model. This is an important area for future research.⁸

This study also only examines attitudes toward PPPs in respect to their potential benefits to communities and the potential roles that the public sector plays in PPPs (expertise and responsiveness). This approach places more emphasis on the public sector’s role in PPPs, and further research could better address public opinions of the private partner’s roles in PPPs. Further research could explore public attitudes toward specific aspects of the private sector’s role in PPPs, such as financing, foreign company involvement, or tolling. Despite these limitations, this study is the first of its kind to measure public opinions of PPPs across a broad sample of U.S. residents, and it is the first to consider how respondent familiarity with PPPs could influence these views.

The practice of PPPs has long called into question the need for identifying the scale of citizen support for PPPs, and the types of persons who support or oppose them (Box, Marshall, Reed, & Reed, 2001; Chen et al., 2013; deLeon & Denhardt, 2000; Nabatchi, 2012). The research provides a step in this direction; highlighting the areas of public support for PPPs and their concerns, and identifying the influence of particular factors in formulating those attitudes. This research can also inform government and its private partners in learning more about citizen expectations, preferences, and understanding of PPP projects, to reduce opposition, and to manage projects that meet requirements, cost, and schedule parameters.

Footnotes

Appendix A

Appendix B

Descriptive Statistics.

Variable	Observation	M	SD	Minimum	Maximum
PPPs can benefit community	917	2.46	0.63	1.00	3.00
Government expertise for PPP	908	2.06	0.71	1.00	3.00
Government receptive in PPP	918	2.28	0.71	1.00	3.00
Gender	1,000	0.54	0.50	0.00	1.00
Age	1,000	49.47	16.75	18.00	88.00
Education	1,000	3.58	1.47	1.00	6.00
Family income (low)	1,000	0.43	0.50	0.00	1.00
Family income (high)	1,000	0.45	0.50	0.00	1.00
Trust in civil servants	996	2.70	0.83	1.00	4.00
Governor approval	909	2.7	0.83	1.00	4.00
Democrat	1,000	0.36	0.48	0.00	1.00
Republican	1,000	0.24	0.43	0.00	1.00
Independent	1,000	0.27	0.45	0.00	1.00
Support free enterprise	996	3.45	1.37	1.00	5.00
Worry about deficit	997	4.47	1.14	1.00	7.00
Want to expand deficit	977	5.47	1.75	1.00	9.00
State deficit per capita	997	0.002	0.28	−2.61	2.68
Perceived familiarity	1,000	2.28	1.67	0	5
Validated familiarity	1,000	1.78	1.54	0	5
Misreport rate	1,000	0.49	0.89	0	5

Note. PPP = public–private partnership.

Acknowledgements

We would like to thank the editor and anonymous reviewers for their useful suggestions. We would also like to thank the following people for feedback on earlier versions of the analysis: Daniel Scheller, Aleksey Kolpakov, and Wendy Francis. All errors are our own.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Eric J. Boyer is an assistant professor at the University of Texas at El Paso. He teaches on Public Management, Nonprofit Management & Social Entrepreneurship, Public and Nonprofit Program Evaluation, and Contract Management.

David M. Van Slyke is Dean of the Maxwell School of Citizenship and Public Affairs at Syracuse University and the Louis A. Bantle Chair in Business-Government Policy.

References

Alvarez

R. M.

Brehm

(1998). Speaking in two voices: American equivocation about the internal revenue service. American Journal of Political Science, 42, 418-452.

Ansolabehere

Hersh

(2012). Validation: What big data reveal about survey misreporting and the real electorate. Political Analysis, 20, 437-459.

Ansolabehere

Rivers

(2013). Cooperative survey research. Annual Review of Political Science, 16, 307-329.

Barabas

(2006). Rational exuberance: The stock market and public support for social security privatization. Journal of Politics, 68, 50-61.

Barzelay

(2002). Origins of the new public management: An international view from administration/political science. In McLaughlin

Osborne

S. P.

Ferlie

(Eds.), New public management: Current trends and future prospects (pp. 15-33). New York, NY: Routledge.

Battaglio

R. P.

(2009). Privatization and citizen preferences: A cross-national analysis of demand for private versus public provision of services in three industries. Administration & Society, 41, 38-66.

Battaglio

R. P.

Legge

J. S.

(2009). Self-interest, ideological/symbolic politics, and citizen characteristics: A cross-national analysis of support for privatization. Public Administration Review, 69, 697-709.

Bel

Fageda

(2008). Reforming the local public sector: Economics and politics in privatization of water and solid waste. Journal of Economic Policy Reform, 11, 45-65.

Bingham

L. B.

O’Leary

(2008). Big ideas in collaborative public management. Armonk, NY: M.E. Sharpe.

10.

Box

R. C.

Marshall

G. S.

Reed

B. J.

Reed

C. M.

(2001). New Public Management and Substantive Democracy. Public Administration Review, 61(5), 608-619.

11.

Boyer

E. J.

Van Slyke

D. M.

Rogers

J. D.

(2016). An Empirical Examination of Public Involvement in Public-Private Partnerships: Qualifying the Benefits of Public Involvement in PPPs. Journal of Public Administration Research and Theory, 26(1), 45-61.

12.

Brown

T. L.

Potoski

Van Slyke

(2016). Managing complex contracts: A theoretical approach. Journal of Public Administration Research and Theory, 26, 294-308.

13.

Chen

Hubbard

Liao

C.-S.

(2013). When public-private partnerships fail: Analysing citizen engagement in public–private partnerships–cases from Taiwan and China. Public Management Review, 15, 839-857.

14.

Clawson

R. A.

Oxley

Z. M.

(2013). Public opinion: Democratic ideals, democratic practice. Thousand Oaks, CA: Sage.

15.

deLeon

Denhardt

R. B.

(2000). The Political Theory of Reinvention. Public Administration Review, 60(2), 89-97.

16.

Delli Carpini

M. X.

Keeter

. (1996). What Americans know about politics and why it matters. New Haven, CT: Yale University Press.

17.

Delmon

(2011). Public-private partnerships in infrastructure: An essential guide for policymakers. New York, NY: Cambridge University Press.

18.

Dillman

D. A.

Smyth

J. D.

Christian

(2008). Internet, mail, and mixed-mode surveys: The tailored design method. New York, NY: John Wiley.

19.

Dubin

J. A.

Navarro

(1988). How markets for impure public goods organize: The case of household refuse collection. Journal of Law, Economics, & Organization, 4, 217-241.

20.

Dunleavy

Husbands

C. T.

(1985). British democracy at the crossroads: Voting and party competition in the 1980s. London, England: Allen & Unwin.

21.

Durant

R. F.

Legge

J. S.

Jr. (2001). Politics, public opinion, and privatization: A test of competing theories in Great Britain. Public Organization Review, 1, 75-95.

22.

Durant

R. F.

Legge

J. S.

Jr. (2002). Politics, public opinion, and privatization in France: Assessing the calculus of consent for market reforms. Public Administration Review, 62, 307-323.

23.

Eliason

S. R.

(1993). Maximum likelihood estimation: Logic and practice. Newbury Park, CA: Sage.

24.

Federal Highway Administration. (2017). P3 Defined. Retrieved from https://www.fhwa.dot.gov/ipd/p3/defined/

25.

Feldman

(1988). Structure and consistency in public opinion: The role of core beliefs and values. American Journal of Political Science, 32, 416-440.

26.

Garvin

M. J.

(2010). Enabling development of the transportation public-private partnership market in the United States. Journal of Construction Engineering and Management, 136, 402-411.

27.

Geddes

R. R.

Wagner

B. L.

(2013). Why do US states adopt public-private partnership enabling legislation? Journal of Urban Economics, 78, 30-41.

28.

Goldenberg

K. L.

Butani

Phipps

P. A.

(1993). Response analysis surveys for assessing response errors in establishment surveys. Retrieved from https://www.bls.gov/ore/abstract/st/st930240.htm

29.

Grimsey

Lewis

M. K.

(2007). Public private partnerships: The worldwide revolution in infrastructure provision and project finance. Northampton, UK: Edward Elgar.

30.

Heath

Gifford

(2006). Free-market ideology and environmental degradation. Environment & Behavior, 38, 48-71.

31.

Hodge

G. A.

Greve

(2007). Public-private partnerships: An international performance review. Public Administration Review, 67, 545-558.

32.

Hodge

G. A.

Greve

(2010). Public-private partnerships: Governance scheme or language game? Australian Journal of Public Administration, 69, S8-S22.

33.

Hood

(1991). A public management for all seasons? Public Administration, 69, 3-19.

34.

InfraPPP. (2017). Project tracker: Database of PPP projects. Retrieved from http://www.infrapppworld.com/project-tracker

35.

Jamali

(2004). Success and failure mechanisms of public-private partnerships (PPPs) in developing countries. The International Journal of Public Sector Management, 17, 414-430.

36.

Kalleberg

A. L.

(1969). Concept formation in normative and empirical studies: Toward reconciliation in political theory. The American Political Science Review, 63, 26-39.

37.

Y. J.

Wang

S. Q.

Chan

A. P. C.

Cheung

(2009). Research trend of public-private partnership in construction journals. Journal of Construction Engineering and Management, 135, 1076-1086.

38.

Ketokivi

M. A.

Schroeder

R. G.

(2004). Perceptual measures of performance: Fact or fiction? Journal of Operations Management, 22, 247-264.

39.

Klijn

E. H.

Koppenjan

(2016). The impact of contract characteristics on the performance of public–private partnerships (PPPs). Public Money & Management, 36, 455-462.

40.

Kohut

Doherty

Dimock

Keeter

(2012). What voters know about campaign 2012. Washington DC: The Pew Research Center for the People & the Press. Retrieved from http://www.people-press.org/2012/08/10/what-voters-know-about-campaign-2012/

41.

Laris

(2015, October 17). Agreement for new submerged tunnel in Norfolk leaves Virginia underwater. The Washington Post. Retrieved from https://www.washingtonpost.com/local/trafficandcommuting/agreement-for-new-submerged-tunnel-in-norfolk-leaves-virginia-underwater/2015/10/17/f03b68f4-566b-11e5-8bb1-b488d231bba2_story.html

42.

Lee

Benoit-Bryan

Johnson

T. P.

(2012). Survey research in public administration: Assessing mainstream journals with a total survey error framework. Public Administration Review, 72, 87-97.

43.

Legge

J. S.

Rainey

H. G.

(2003). Privatization and public opinion in Germany. Public Organization Review, 3, 127-149.

44.

Akintoye

Edwards

P. J.

Hardcastle

(2005). Critical success factors for PPP/PFI projects in the UK construction industry. Construction Management and Economics, 23, 459-471.

45.

Lincoln

J. R.

Zeitz

(1980). Organizational properties from aggregate data: Separating individual and structural effects. American Sociological Review, 45, 391-408.

46.

Lohmann

Rötzel

P. G.

(2014). Opportunistic behavior in renegotiations between public-private partnerships and government institutions: Data on public-private partnerships of the German armed forces. International Public Management Journal, 17, 387-410.

47.

Long

J. S.

(1997). Regression models for categorical and limited dependent variables. Thousand Oaks, CA: Sage.

48.

Lopez de Silanes

Shleifer

Vishny

R. W.

(1997). Privatization in the United States. Rand Journal of Economics, 28, 447-471.

49.

Monroe

A. D.

(1979). Consistency between public preferences and national policy decisions. American Politics Quarterly, 7, 3-19.

50.

Newell

Reeher

Ronayne

(2008). The trusted leader: Building relationships that work. Washington, DC: CQ Press.

51.

A. Y.

(2012). The risk-averting game of transport public-private partnership: Lessons from the adventure of California’s state route 91 express lanes. Public Performance & Management Review, 36, 253-274.

52.

Nye

J. S.

Jr. Zelikow

P. D.

King

D. C.

(1997). Why people don’t trust government. Cambridge, MA: Harvard University Press.

53.

O’Leary

Vij

(2012). Collaborative public management: Where have we been and where are we going? American Review of Public Administration, 42, 507-522.

54.

Osborne

Gaebler

(1993). Reinventing government. New York, NY: Penguin Group.

55.

Page

B. I.

Shapiro

R. Y.

(1983). Effects of public opinion on policy. The American Political Science Review, 77, 175-190.

56.

Poocharoen

O. O.

Ting

(2015). Collaboration, co-production, networks: Convergence of theories. Public Management Review, 17, 587-614.

57.

P3Bulletin. (2017). Project tracker. Retrieved from http://www.p3bulletin.com

58.

Public Works Financing. (2017). Public Works Financing’s major projects database. Retrieved from http://pwfinance.net/projects-database/

59.

Robinson

Carrillo

Anumba

C. J.

Patel

(2010). Governance & knowledge management for public private partnerships. Oxford, UK: Wiley-Blackwell.

60.

Rudolph

T. J.

Popp

(2009). Bridging the ideological divide: Trust and support for social security privatization. Political Behavior, 31, 331-351.

61.

Ruiz

Elliott

Ratcliffe

R. G.

(2009). Trans-Texas corridor plans dropped after public outcry. Houston Chronicle. Retrieved from http://www.chron.com/neighborhood/cyfair-news/article/Trans-Texas-Corridor-plans-dropped-after-public-1745411.php

62.

Savas

E. S.

(2000). Privatization and public-private partnerships. London, England: Chatham House.

63.

Schlesinger

Heldman

(2001). Gender gap or gender gaps? New perspectives on support for government action and policies. Journal of Politics, 63, 59-92.

64.

Shadish

W. R.

Cook

T. D.

Campbell

D. T.

(2002). Experimental and quasi-experimental designs. Boston, MA: Houghton Mifflin.

65.

Siemiatycki

(2010). Delivering transportation infrastructure through public-private partnerships: Planning concerns. Journal of the American Planning Association, 76, 43-58.

66.

Siemiatycki

(2011). Urban transportation public-private partnerships: Drivers of uneven development? Environment and Planning A, 43, 1707-1722.

67.

Stimson

J. A.

Mackuen

M. B.

Erikson

R. S.

(1995). Dynamic representation. The American Political Science Review, 89, 543-565.

68.

Swidorski

(1980). Sample surveys: Help for the “out-of-house” evaluator. Public Administration Review, 40, 67-71.

69.

Thompson

Elling

R. C.

(2000). Mapping patterns of support for privatization in the mass public: The case of Michigan. Public Administration Review, 60, 338-348.

70.

U.S. Census Bureau. (2017a). State population totals tables: 2010-2016. Retrieved from https://www.census.gov/data/tables/2016/demo/popest/state-total.html

71.

U.S. Census Bureau. (2017b). 2014 state and local summary table by level of government and by state. Retrieved from https://www.census.gov//govs/local/historical_data_2014.html

72.

Van Slyke

D. M

. (2003). The mythology of privatization in contracting for social services. Public Administration Review, 63, 296-315.

73.

Vining

A. R.

Boardman

A. E.

(2008). Public-private partnerships in Canada: Theory and evidence. Canadian Public Administration/Administration Publique du Canada, 51, 9-44.

74.

Wilcox

(1991). The causes and consequences of feminist consciousness among Western European women. Comparative Political Studies, 23, 519-545.

75.

Zhang

Ali Soomro

(2016). Failure path analysis with respect to private sector partners in transportation public-private partnerships. Journal of Management in Engineering, 32(1), 1-10.