What Factors Affect Management Quality? State Infrastructure Management and the Government Performance Project

Abstract

Whether the US$131 billion set aside for infrastructure projects under the American Recovery and Reinvestment Act of 2009 can make a lasting contribution to improving the nation’s public infrastructure will depend, in part, on the quality of infrastructure management systems and practices in the states. In this article, we examine the factors that influence how well state governments plan for and manage public infrastructure using results from the 2005 and 2008 Government Performance Project. The pooled Tobit regression analysis shows that capital management quality is affected by political variables such as divided legislatures and legislative term limits, fiscal institutions including tax and expenditure limits, and environmental demand factors, specifically the extent of urbanization.

Keywords

infrastructure management determinants of management quality government performance project

Introduction

The American Recovery and Reinvestment Act of 2009 (PL 111-5) was designed to infuse the national economy with US$787 billion in expenditures and tax credits for the purposes of (a) stimulating job creation and retention and promoting economic recovery, (b) assisting those most impacted by the recession, (c) spurring technology in science and health, (d) investing in transportation, environmental protection, and other infrastructure, and (e) stabilizing state and local government budgets. Nearly US$131 billion of that amount was set aside for infrastructure projects, and the majority of those funds were to support “shovel ready” projects (defined as projects that could be started within 180 days), which meant in most cases repair, replacement, and renovation activities. The size and scope of this extraordinary federal investment in infrastructure in such a short period of time reminds of the large public works projects of the 1930s. Massive public capital investment in the nation’s basic infrastructure was expected to prepare a broad foundation for the purposes of promoting economic activity and enhancing the quality of life. Economists have long been debating the private capital formation impacts of public investment without reaching any consensus (and, indeed, there is a debate about the direction of the impact arrow). Yet, they do agree that economic activity depends on an adequate level of infrastructure investment (see Munnell, 1992), provided principally by the public sector.

Although it is widely recognized that the absence of a supportive infrastructure for private economic activity would have a deleterious effect on the nation’s economy, it should be understood that the absence of a sound system of managing such infrastructure can also cause economic disruptions, dislocations, and decline. Bridge collapses, transit derailments, the “Big Dig” cost overruns, potholes that snap truck axles, bridges to nowhere, water main breaks that flood residences and buildings, and other manifestations of poor infrastructure management should encourage us to examine more than just the dollar amount of infrastructure investment. How effectively infrastructure is managed should also be considered. Ineffective, poorly planned, and obsolete infrastructure would potentially harm private economic activity and the quality of life of citizens. Infrastructure management that is based on comprehensive capital planning, effective project oversight, and adequate asset preservation can benefit the economy and society.

Clearly, whether the US$131 billion set aside for infrastructure projects under the stimulus program will make a lasting contribution to improving the nation’s public infrastructure will depend, in part, on the quality of infrastructure management systems and practices in the states. But what factors influence how well state governments plan for and manage their fixed assets? In this article, we test an exploratory model that examines the impact of political actors and institutions, fiscal rules, government fiscal condition, and environmental demand factors on the quality of capital management systems in the states. Management quality is measured using the grades from the 2005 and 2008 rounds of the Government Performance Project (GPP). Brudney, O’Toole, and Rainey (2000) argue that one advantage of the GPP grades is that they facilitate systematic analysis of public management systems. The GPP grades have been used to explore the links between pubic management and human resource outcomes (Donahue, Selden, & Ingraham, 2000), the policy priorities of state officials (Coggburn & Schneider, 2003), and state fiscal condition (Rubin & Willoughby, 2009).

This study is organized as follows: In the next section, we briefly describe the GPP methodology, and present and contrast the infrastructure management grades in 2005 and 2008. We then develop and test a model explaining the variation in the quality of infrastructure management across the states. We summarize the implications of key findings in the concluding section.

Grading Infrastructure Management in the States

GPP was initiated in 1997 to evaluate state governments’ performance in managing money, people, infrastructure, and information, and to inform state governments’ future management decisions (Ingraham, Joyce, & Donahue, 2003). For infrastructure management, five major criteria were used to assess the quality of state management systems: capital planning, project monitoring, maintenance, internal cooperation, and intergovernmental coordination. These criteria were selected based on the standards employed by GPP in earlier years, further refinement to those standards, and advice from scholars and state governments (Ingraham, 2007; Pagano, 2008). Box 1 provides the details for each criterion.

Box 1.

Infrastructure Management Criteria

Criterion 1—Capital planning: Constructing a fixed asset should be an activity that derives from the state’s assessment of its current and future needs. The state must plan for the fixed asset’s long-term usability. A thorough and comprehensive capital improvement plan must include the state’s best estimate of population growth, demographics, changes in the underlying economic base, transportation growth, technological changes, and the needs and demands of the citizens. These activities are elements of a comprehensive approach to effective capital planning (Dowall, 2001; McNeil, Tischer, & DeBlasio, 2005; Neumann & Markow, 2003; Stich & Eagle, 2005).

To assure a rational and systematic building of the state’s infrastructure, the capital plan should be clearly and logically linked to the state’s capital budget. Rational planning of infrastructure also requires the state to estimate the attendant operating and maintenance (O&M) costs of new or rehabilitated capital assets. Those O&M costs ought to be linked directly to the state’s operating budget. Likewise, expanding capital facilities has consequences for the state’s operating budget, an impact that should be incorporated in a state’s capital planning process.

Criterion 2—Project monitoring: Allowing an adequate time frame for the project, keeping track of the progress of the project, and addressing problems head-on are key components of project oversight. For example, the time frame of the project does not start with groundbreaking or construction. Projects that are identified in the multiyear capital improvement plan must work their way into the capital budget, which is effectively “Year 1” of the capital improvement plan that is authorized by the legislature. Once authorized, the project requires state appropriation of funding; only then does the state begin the design and construction work. Similarly, monitoring the project once its construction has begun is critical. During the construction phase, the state is required to be vigilant in overseeing the project, and the state must have the legal and administrative capacity to intervene when problems are detected. Cost overruns, unwarranted delays, threats to worker and citizens’ safety, and other problems need to be detected and corrected in short order so that the project can be completed according to plan (Harder, Pedersen, Warne, & Martin, 2005).

Criterion 3—Maintenance: After project completion, facilities are “consumed” or used for their designated purposes. Trucks and cars travel on roads and bridges, government employees and petitioners use office buildings, state parks provide recreational facilities to the public, education facilities are used by students. Once a state decides to construct a facility, it is incumbent on the state to assume the related responsibility of ensuring that the expected life span of the facility is reached. This, in turn, requires the state to maintain the facility in proper condition. The rate of deterioration depends on use, quality of the asset, the normal life cycle of the asset, and regular repair and maintenance of the facility. Underfunding maintenance and repair activities can, over the long term, reduce the efficiency of the fixed asset and hasten its demise. When that happens, the social compact between generations is broken (Baker & Lambert, 2001; Maxwell School of Citizenship and Public Affairs, 2003; Neumann, Markow, & Lambert, 2003; Petersen, 2004; Transportation Research Board, 2004).

Criterion 4—Internal coordination: The fourth criterion of effective infrastructure management requires cooperation and coordination across the various agencies of the state that have infrastructure responsibilities. Statewide priorities and a statewide review board can ensure that public facility construction is undertaken to further the collective interests of the state. When the state’s parks department plans for expansion, is the department of transportation part of the deliberation process? Do the state’s leaders view the construction and integration of the state’s infrastructure from a systemic perspective?Box 1. (continued)Coordination among state agencies in planning and building the state’s infrastructure can improve the efficiency of state operations (Groden, 2000; Lufkin, Desai, & Janke, 2005; Neumann, Markow, & Lambert, 2003; Vogt, 2004).

Criterion 5—Intergovernmental coordination: The fifth criterion speaks to another form of cooperation and coordination, but from the intergovernmental perspective. Local governments, as creatures of the states, often receive infrastructure construction and operating support from the state. Does the state communicate needs, priorities, regulations (from both the state and the federal government) to the local governments, or require performance standards to be met, or actively plan infrastructure projects with local governments? State advisory commissions on intergovernmental relationships are representative of the kinds of institutions that are designed to promote intergovernmental cooperation and coordination (Foster & Franklin, 2000).

GPP Methodology

GPP employed a multimethod data gathering effort to evaluate the management systems and practices in the states (Donahue, Selden, & Ingraham, 2000; Ingraham, Joyce, & Donahue, 2003). Data were collected from a survey of all states, interviews with state officials, and documents and reports on infrastructure management. Each state’s infrastructure management capabilities were evaluated on the basis of the extent to which they met the conditions established in each criterion. The evaluation scheme employed by GPP ranked states’ performance for each criterion by a rating scheme that included “weakness,” “midlevel,” and “strength.”

The overall infrastructure management grade were determined by differentially weighting the criteria scores, ensuring that grade differences represented real distinctions among the states and ensuring that like cases were treated in a like manner—that is, that two states with the same criteria scores got the same overall grades. The weighting scheme, which was only a guideline and not a rigid algorithm, was heavily influenced by the maintenance criterion. Maintenance of existing infrastructure is indeed the primus inter pares among the five criteria and weighted approximately 30% of the final grade. Capital planning and project monitoring are each weighted approximately 25% of the final grade, and internal coordination and intergovernmental coordination are each weighted approximately 10% of the final grade. Based on the performance of the states in the five interrelated areas of infrastructure management, GPP assigned letter grades to each state. Possible grades ranged from A to F.

Comparing the Results of the 2005 and 2008 Rounds

Tables 1 and 2 present the 2005 and 2008 GPP grades for infrastructure management. Note that because of the changes implemented in the assessment criteria in 2004, only the GPP grades in the last two rounds are directly comparable (Barrett & Greene, 2005). In the 2005 and 2008 GPP, a majority of the states received letter grades in the B range (B+, B, and B–). States which received grades within the A range increased from only 2 in 2005 (Utah and Florida) to 4 in 2008 (Florida, Kentucky, Michigan, and Utah). In 2005, Alabama received the lowest grade of D followed by New Mexico with a D+. In 2008, New Hampshire and Massachusetts had the lowest grade of D+. The discussion below focuses on the evaluation of states’ performance in each of the five infrastructure management criteria.

Table 1

2005 GPP Infrastructure Management Grades

State	Overall grade	Capital planning	Project monitoring	Maintenance	Internal coordination	Intergovernmental
UT	A	s	s	s	s	s
MD	A–	s	s	m	s	s
OH	A–	s	s	s	m	s
VA	A–	s	m	s	s	m
DE	B+	s	s	m	m	m
FL	B+	m	m	s	m	s
KY	B+	s	m	m	s	s
MI	B+	s	m	m	m	s
NE	B+	s	m	m	m	s
NV	B+	m	m	m	s	s
NY	B+	s	m	m	s	s
PA	B+	s	s	m	s	s
IA	B	m	w	s	m	m
ME	B	m	s	w	s	m
MN	B	s	s	m	s	s
OR	B	m	m	m	s	s
SD	B	m	m	m	w	m
WA	B	m	m	m	m	s
AZ	B–	m	s	w	m	m
IN	B–	m	m	w	m	s
KS	B–	m	m	m	m	m
MO	B–	m	m	w	m	s
MT	B–	m	m	w	s	m
NJ	B–	s	m	w	s	s
ND	B–	m	m	w	m	s
RI	B–	m	m	m	w	s
TN	B–	m	m	w	s	m
TX	B–	w	m	s	m	m
VT	B–	s	m	w	s	s
AK	C+	m	m	w	m	m
AR	C+	w	m	w	m	m
CO	C+	m	m	w	m	m
CT	C+	w	m	m	m	m
GA	C+	m	w	w	m	s
ID	C+	w	m	w	m	m
IL	C+	m	w	m	s	m
LA	C+	w	m	w	m	m
MS	C+	m	m	w	m	m
NH	C+	m	m	w	m	m
NC	C+	w	w	m	m	m
SC	C+	m	m	w	m	s
CA	C	m	m	w	w	s
WV	C	w	m	w	w	m
WI	C	w	m	w	m	s
WY	C	w	m	w	m	m
HI	C–	w	w	m	w	m
MA	C–	w	w	m	m	m
OK	C–	w	m	w	m	m
NM	D+	w	w	w	m	m
AL	D	w	w	w	w	w

Note: w = weakness, m = mid-level, s = strength.

Table 2

2008 GPP Infrastructure Management Grades

State	Overall grade	Capital planning	Project monitoring	Maintenance	Internal coordination	Intergovernmental
UT	A	s	s	s	s	s
FL	A–	m	s	s	m	s
KY	A–	s	s	m	s	s
MI	A–	s	s	m	m	s
DE	B+	m	s	m	s	s
IN	B+	m	m	s	s	s
MD	B+	s	m	m	s	s
MO	B+	s	m	m	m	s
NE	B+	m	m	s	s	s
NV	B+	m	w	m	s	s
VT	B+	m	m	s	m	s
VA	B+	s	m	m	m	m
WA	B+	s	m	m	m	s
GA	B	m	s	m	m	s
SD	B	m	m	m	m	m
TN	B	m	m	m	s	s
WY	B	m	s	m	m	m
AZ	B–	m	m	m	m	s
ID	B–	w	m	m	m	m
NY	B–	s	m	w	m	s
NC	B–	m	m	w	m	m
OH	B–	m	m	m	m	s
PA	B–	m	s	m	m	s
TX	B–	m	m	s	m	m
WI	B–	w	m	m	m	s
AL	C+	m	s	w	m	m
AR	C+	w	m	m	m	m
CO	C+	m	m	w	m	m
CT	C+	w	m	m	m	m
IA	C+	m	m	w	m	m
KS	C+	m	m	w	m	m
LA	C+	w	m	w	m	m
ME	C+	m	m	w	s	m
MN	C+	m	m	w	m	s
MS	C+	m	m	w	m	m
MT	C+	m	m	w	m	m
NJ	C+	m	m	w	s	s
OR	C+	m	m	w	m	s
RI	C+	m	m	w	m	s
CA	C	s	s	w	w	s
HI	C	w	m	w	m	m
IL	C	w	m	w	m	m
NM	C	m	m	w	m	m
ND	C	m	m	w	m	s
AK	C–	w	m	w	w	m
OK	C–	w	m	w	m	m
SC	C–	w	m	w	m	s
WV	C–	w	m	w	w	m
MA	D+	w	m	w	m	m
NH	D+	w	w	w	m	m

Note: GPP = Government Performance Project. w = weakness, m = mid-level, s = strength.

Criterion 1: Capital planning

In the 2008 round, a majority of the states received favorable assessments for capital planning. Federal highway planning requirements are the primary reasons for the generally higher rating on the capital planning criterion. All states receive federal-aid money for transportation. As a condition of aid, a state must create a Statewide Transportation Improvement Program (STIP) that lists all federal projects ordered by priority and by funding sources. Federally supported transportation projects in metropolitan areas must be approved by the local Metropolitan Planning Organizations, which also must conform to State Implementation Plans for attaining federal air quality standards. In effect, the federal grant requires a minimum level of transportation planning for every state or what might be thought of as the “floor” for infrastructure planning in any state.

Comparing the ratings in the 2005 and 2008 rounds, there has been a general improvement in states’ performance in capital planning. The number of weak states decreased from 14 in 2005 to 13 in 2008. In addition, states which received a midlevel grade jumped from only 23 in 2005 to 28 in 2008, whereas the number of strong states decreased from 13 to 9.

In the online survey, a total of 37 states listed the reasons for failing to undertake a comprehensive assessment of needs and to do a transparent selection of projects during the capital planning process. Competing priorities (15 states) and lack of funding (12) were the most frequently cited obstacles, followed by high cost of the process (5), political processes (3), lack of staff or skilled personnel (2), and high population growth (1).

Criterion 2: Project monitoring

Improvements were also registered in the project monitoring grades of states. In 2005, 7 states received a grade of “weak” in project monitoring, 34 were “midlevel,” and 9 registered a “strength.” In 2008, only 2 states were rated weak, 38 states were midlevel, and 10 states were rated strong in project monitoring.

The monitoring criterion assesses the extent to which states regularly collect data on project construction, the means by which data are collected, and the response or correction time once a deficiency or problem has been detected. Specifically, the problem areas for which data and information on reporting and correction time were collected include project efficiency, quality of work, cost overruns, project delays, and safety compliance. For nontransportation projects, a majority of the states with usable responses to the online survey reported being able to implement action to address project inefficiencies (28 out of 39 responding states), control costs (29 states), ensure timeliness of project completion (28) and quality (33), and meet safety standards (33) in 2 to 3 weeks or less. However, for transportation projects, more states reported a longer time period (1-2 months) before being able to respond to problems of cost overruns and project delays (29 out of 40 responding states).

Criterion 3: Maintenance

Maintenance presents the most difficult challenge to most states. In 2005, 54% or 27 states received the lowest grade for poor maintenance of their existing capital stock. Some 36% or 18 states were midlevel, and only 1% or 5 states were considered strong performers. The 2008 assessment finds that most states made some headway in improving their maintenance systems and policies. The number of poor performers decreased from 27 to 23 states, states receiving midlevel scores increased from 18 to 21, and strong performers increased from 5 to 6 states. Still, the fact that some 23 states, or almost half of the states, performed poorly on maintenance is a worrisome trend.

A majority of the states have developed asset management systems designed to assess the condition of and to estimate the intensity and timing of maintenance and repair investments for facilities in transportation, corrections, office buildings, the state capitol, libraries and parks, and recreation. Despite this accomplishment, however, a number of states continue to underfund maintenance activities. Among the 34 states which provided information on maintenance underfunding in the online survey, more than half or 19 states reported that maintenance was underfunded by more than 26% in 2008. Of these states, 7 reported that maintenance was underfunded by 26% to 50%, and 13 states reported more than 50% underfunding.

One reason for the continuing neglect of maintenance activities is that they can be postponed without immediate backlash. Usually, a fixed asset will not deteriorate if it is not maintained properly for a year, but after several years of neglect, the facility deteriorates rapidly.

Criterion 4: Interagency coordination

Interagency coordination is a controversial area because it could be read as favoring a “centralist” orientation, meaning that only states with hierarchical and centralized bureaucracies score well. This, however, is not the case. The issue here is whether, and how, all state infrastructure agencies cooperate in the planning, design, and construction of facilities. The focus is on encouraging the state to look at all its fixed assets holistically as part of a statewide system of investment.

In 2005, 6 states performed poorly in internal coordination, 29 were midlevel, and 15 got the highest rating. In the 2008 round, the number of weak states declined to only 3, states which received a satisfactory rating increased to 37, whereas strong performers decreased to 10.

In terms of capital planning coordination, 17 of the 40 states which provided useable answers to the survey question on interagency coordination reported that their central budget or finance office is primarily responsible for developing capital plans. In another 10 states, individual agencies play a more active role in developing capital plans and formally consult other agencies before submitting their requests to the state budget or finance office. Agencies in 19 states informally consult other relevant agencies in the capital plan development process. In general, there is a certain degree of internal coordination across states but the capital plan is not always prepared in a statewide manner. Most states create a state plan and a transportation plan separately but how those two plans inform each other is often unknown.

Criterion 5: Intergovernmental coordination

This criterion speaks to a form of cooperation and coordination that is similar to the previous criterion but from the intergovernmental perspective. Local governments, as creatures of the states, often receive infrastructure construction and operating support from the state. The mechanisms for intergovernmental cooperation varies from simple communication channels such as newsletters, websites, and other information distribution systems, to more formal arrangements such as state advisory commissions on intergovernmental relationships which are institutions that are designed to promote intergovernmental coordination. Of the 40 states which provided information on interstate coordination in the 2008 survey, half reported that they relied on ad hoc task forces, 15 used regular meetings, and 8 coordinated infrastructure-related issues across state lines through one central state agency or office. Some 5 states reported that they employed no interstate coordination mechanism.

A majority of the states performed well in terms of the extent to which they have created effective intergovernmental and interstate infrastructure management networks. Comparing the grades for intergovernmental coordination in the last two GPP rounds, weak performers declined from 1 in 2005 to none in 2008, strong states increased from 23 to 26, and states that were midlevel declined from 26 to 24.

Literature Review: Drivers of Innovation in the Public Sector

Both the 2005 and 2008 GPP rounds show that there are considerable differences in how the states manage their fixed assets. What factors explain the variation in the quality of infrastructure management systems and processes in the states?

Unfortunately, existing literature that examines how and why state or local governments adopt best practices in infrastructure management is very thin. Hoffmann, Krumholz, O’Brien, and Geyer’s (2000) case study pointed to the positive contributions of educational training of, and shared commitment among planners, to the success of capital planning efforts in Cleveland. Ebdon’s (2004) analysis grouped counties according to their GPP capital management scores to see if they were different in terms of a number of characteristics. The counties were not dissimilar in terms of government form and population change, but Ebdon found that low scorers tended to be found in the West and had smaller population. Without controlling for the effects of other variables, however, it is difficult to rule out the possibility that the findings in both studies are spurious.

A regression analysis by Yusuf, O’Connell, Hackbart, and Wallace (2008) using 2000 GPP infrastructure management scores found that stronger capital management capacity in state transportation departments was associated with having a larger (in terms of membership) transportation commission that played a role in long-term planning. A problem with this finding is that GPP grades also reflected whether transportation departments undertook long-term planning (see Criterion 1)—thus both the dependent and independent variables are, to a certain extent, measuring a similar aspect of infrastructure management.

Exploring the general public administration literature provides more information about the drivers of innovation and management reforms in public organizations. Infrastructure management processes—planning, needs assessment, using technology (specifically web-based tools such as shared database systems) in maintenance assessment and prioritization, and using plans and performance data to drive budgetary decisions—are related to management innovations examined in public administration research including strategic planning, digital government, and performance measurement and evaluation.

In the case of strategic planning, Berry (1994) found that adoption was more likely when state agencies had strong business orientation. In addition, adoption of strategic planning was promoted when there was a new gubernatorial administration, more slack resources, and when other agencies in neighboring states also engaged in planning. Focusing on e-government innovation, Tolbert, Mossberger, and McNeal (2008) found that state governments with stronger institutional capacity—those with highly developed information technology administration and management infrastructure and professional legislatures—were more likely to use digital government.

de Lancer Julnes and Holzer (2001) studied adoption and use of performance measurement in state and local governments and found that adoption was influenced more by rational-technocratic factors such as external requirements to measure performance, available resources, goal-orientation, and technical knowledge. The authors argued that implementation was driven more by politico-cultural factors specifically internal and external interest groups and weaker unionization of workers.

To explain the implementation of reinvention reforms such as quality improvement, benchmarking, and strategic planning in state agencies, Brudney, Herbert, and Wright (1999) examined the effects of agency type and characteristics, state reform effort, environmental influences, and background and attitudes of agency directors. They found that larger agencies and those that had governor-appointed directors, had experienced dramatic changes in goals, and performed staff function were more likely to adopt reinvention-type reforms. Statewide effort at reform and long history of administrative reorganization, as well as interest group influence, were also associated with reinvention. Finally, the extent of agency directors’ policy influence, prioritization of organizational goals, and orientation toward customers correlated positively with reform implementation.

Echoing Brudney, Herbert, and Wright’s (1999) findings on the importance of leadership, Borins (2001) focused on the role of political and administrative leaders and also examined the contributions of staff in promoting innovation in the public sector. Examining recipients of the State and Local Government Innovation Awards from 1990 to 1994, Borins (2001) pointed to the importance of middle managers and frontline staff as innovation initiators. In another study, Borins (2002) focused on the role of political leaders and agency heads in the innovation process and argued that leadership mattered in different types of public management innovation, by, among others, motivating, consulting, and rewarding staff, reaching out to external stakeholders including clients, and publicly promoting bottom-up reforms.

Finally, in a study of English local governments, Boyne, Gould-Williams, Law, and Walker (2005) found that adoption of innovative management practices such as performance plans, action plans, benchmarking, performance indicators, user evaluation, and employee evaluation was promoted by population dispersion, prior experience in reform, and targeting a small number of services.

Explaining Variation in Infrastructure Management Quality: Model and Hypotheses

Clearly, different factors influence the adoption and implementation of management best practices in public organizations. Identifying the sources of variation in infrastructure management quality across state governments becomes all the more complicated given that the analysis does not focus on the adoption of one or two innovations or best practices, but on an entire management system that encompasses planning, project management, maintenance, budgeting, and interagency and intergovernmental cooperation. Consequently, the model we test is exploratory and focuses on macro-level or statewide variables, rather than on micro- or agency-level variables. This limitation should be noted given previous studies showing that management capacity, reforms, or innovation can be influenced by agency leadership, employee attitudes (see Berry, 1994; Borins, 2001, 2002; Brudney, Herbert, & Wright, 1999; de Lancer Julnes & Holzer, 2001), and other factors such as membership composition of agency-level commissions (Yusuf et al., 2008).

We test the following exploratory model to explain state government performance in capital management:

GPP = f {P, FI, FC, E}

where infrastructure management quality (GPP) is hypothesized to be influenced by political variables (P), fiscal institutions (FI), state fiscal condition (FC), and environmental demand factors (E).

Political Factors

Studies have shown that interest groups influence adoption of management innovations (see Berry, 1994; Brudney, Herbert, & Wright, 1999; de Lancer Julnes & Holzer, 2001). Historical evidence suggests that business interest groups are likely to demand more efficient government management to facilitate economic expansion. For example, the local government institutional and management reforms implemented during the early 20th century Progressive Movement—such as nonpartisan ballots, recall elections, council-manager government, and merit system, among others—were largely a result of the demands of local business groups to curtail city corruption and improve government efficiency in support of local economic growth (Judd & Swanstrom, 2002). Adopting the strategy of Gray and Lowery (1988), we measure the size of business interest groups as the percentage of state’s labor force employed in manufacturing and transportation-related industries. It is hypothesized as follows:

Hypothesis 1-Hypothesis 2: States with a higher percentage of jobs in manufacturing and transportation have higher quality infrastructure management systems.

Other mechanisms link politics with management quality. The impact of political competition, for one, has been extensively studied in the literature. Political competition is usually measured by divided government which occurs when the governor and the majorities in both legislative chambers come from different political parties and divided legislature which refers to control of both the upper and lower houses of the legislature by different parties. Poterba (1994) and Alt and Lowery (1994) argue that political competition complicates government decision making. For example, divided government and legislature may lead to pork-barrel type politics as candidates for office distribute benefits to attract more voters. This means that selection of projects in the capital budget are not be based on capital planning priorities but on decisions of legislators to deliver public works projects and contracts to their supporters and constituencies.

Reform gridlock is another possible result of competition within and across the political branches of government. If different parties control majorities in both legislative chambers, and the governorship, it might be more difficult to reach agreement on implementing reforms targeted at improving management of fixed assets, such as implementation of a capital planning process. However, a unified government can reduce decision-making costs increasing possibility of administrative reforms (Horn, 1995). We used dummy variables to measure divided government and legislature. It is expected as follows:

Hypothesis 3-Hypotheses 4: States with divided governments and legislatures have lower quality infrastructure management systems.

Legislative term limits and professionalization are also likely to affect the quality of infrastructure management systems in the states. Legislators play a crucial role in shaping the quality of state infrastructure management systems as they not only select capital projects and control budget authorizations but also can enact policies requiring a comprehensive long-term capital planning process, dictate the regularity of infrastructure condition assessment, and identify regular sources of financing for maintenance of fixed assets.

Legislative term limits facilitate greater legislative focus on improving state management systems to enhance public service delivery. Will (1992) argues that term limits control legislators’ preoccupation with reelection and encourage them to consider broader interests rather than those of their districts alone. Carey, Niemi, and Powell (1998) find that legislators in term-limited states are less likely to spend time seeking out district pork and more likely to focus on state-level needs relative to district interests. Bourdeaux and Chikoto (2008) provide evidence that legislative term limits increase use of performance measures in state agencies. Term limits are measured with an indicator variable.

Legislative professionalization has also been linked to public management reform initiatives in the states. Professionalization of legislatures refer to the degree to which state legislative assemblies have access to resources to undertake informed deliberation of issues and to formulate sound policy initiatives. Tolbert, Mossberger, and McNeal (2008) find that legislative professionalization is a strong predictor of the implementation of digital government in the states. Similarly, Kellough and Selden (2003) find a strong positive relationship between professional legislatures and the implementation of public personnel reform in state governments. A measure of professionalization of state legislatures is that of Squire (2007) which uses the U.S. Congress as a standard against which to assess salaries, staff support, and time in session in state legislatures. It is hypothesized as follows:

Hypothesis 5-Hypothesis 6: States with legislative term limits and more professional legislatures have higher quality infrastructure management systems.

Fiscal Institutions

Institutions play an important role in determining governmental policies. They refer to both the formal and informal rules of the game that constrain or facilitate behavior (Ostrom, 1990). Many state decisions are governed by a plethora of fiscal rules designed to control government profligacy by constraining public revenues, debt, and expenditures. Three fiscal rules are included in the analysis—balanced budget requirements, debt limitations, and tax and expenditure limits (TELs).

Studies show that fiscal institutions can improve fiscal discipline in state governments. Bohn and Inman (1996) find that tight end-of the-year antideficit rules lead to effective control of state general fund deficits. Kiewiet and Szakaly’s (1996) study concludes that states that required referendum approval before debt could be issued and those that actually prohibited guaranteed debt had less debt compared with states which required a supermajority of the legislature to issue debt or those with revenue-based limitations. Finally, Rueben (1995) provides some evidence that TELs do indeed reduce the growth of state governments.

The problem with fiscal institutions, however, is that they can restrict the flexibility of state officials and administrators to deal with various infrastructure management issues. A recent study by Jimenez (2009), which uses state-level data from 1980 to 2005, shows that states with TELs tend to reduce their shares in aggregate state and local expenditures for capital-intensive, development-oriented services such as highways, water and air transportation, sewerage, and sanitation. Jimenez points out that as states simultaneously deal with the problems of rising cost of welfare entitlement programs such as Medicaid and constrained capacity to raise new revenues as a consequence of TELs, policymakers are forced to sacrifice expenditures for other services such as operation and maintenance of fixed assets.

We measure the strictness of antideficit rules using Advisory Commission on Intergovernmental Relations’ (ACIR, 1995) balanced budget stringency index which ranges from 0 to 10, with 10 being the most stringent. With regard to debt limits, we only include the most stringent types—referendum-based debt limits and rules that prohibit guaranteed debt—which are coded as simple dummy variables. Finally, a TEL stringency index is constructed with 0 indicating that state-level TELs are not in effect for state i in year t, 1 if TELS are in effect but are not codified in state constitutions and do not limit increases in revenues or spending to two or more growth factors (such as population, inflation, and income growth), 2 if they are codified in state constitutions or limit increases in revenues or spending to two or more growth factors, and 3 if TELs are both codified in state constitutions and limit increases in revenues or spending to two or more growth factors. It is expected as follows:

Hypothesis 7-Hypothesis 9: States with strict antideficit rules, TELs, and debt limits have higher quality infrastructure management systems.

State Government Fiscal Condition

The literature on public financial management indicates that fiscal stress may hamper the implementation of management reforms precisely because governments do not have the resources to support such reforms. As Schick (1980) points out, periods when governments have sufficient resources to support new programs represent the best time to implement rational management and budgeting techniques such as for instance, capital planning and asset condition assessment systems. States with greater fiscal resources have greater capacity to adopt and experiment with innovative practices (Kellough & Selden, 2003). Berry (1994) and de Lancer Julnes and Holzer’s (2001) studies conclude that government slack resources are associated with the adoption and implementation of strategic planning and performance measurement in state agencies and local governments.

There is no widely accepted indicator of government fiscal condition (Jimenez, 2009). The first measure of state fiscal condition is one which has been frequently used in the literature—general fund balance and budget stabilization fund (BSF) as a percentage of general fund expenditures (Poterba, 1994; Rubin & Willoughby, 2009). General fund balances are the residual equity or the difference between all assets and liabilities in the general fund account. BSFs, often referred to as rainy day funds, represent a specific mechanism for states to save during boom years. Also included in the model is per capita own-source revenue which measures slack resources that can be invested to improve management systems and processes (see Tolbert, Mossberger, & McNeal, 2008). The final measure of state fiscal condition is the growth rate in intergovernmental revenue. Intergovernmental revenues are one of the most important sources of revenues for states. To control for inflation, all monetary data are converted into 2000 dollars using the implicit price deflator published by the Bureau of Economic Analysis. It is expected as follows:

Hypothesis 10-Hypothesis 12: States with higher fund balances, revenue per capita, and intergovernmental revenues have higher quality infrastructure management systems.

Organizational Environment

There is a rich literature in organization theory which argues that organizational structure, reform, and innovation can be explained by attempts of organizations to respond to changes, demands, and threats from the external environment. Gaus’ (1947) organizational ecology model argues that public administration, its development, and its activities are influenced by its setting or ecology. For instance, migration from rural to urban areas required government to finance transportation and sanitation services and to look for new ways of raising taxes and revenues to support such services. Burns and Stalker (1961) aver that management innovation is driven by conditions in the external environment. In the contingency theory, Lawrence and Lorsch (1968) suggest that the complexity or differentiation in internal structures and the rise of different coordinating mechanisms reflect an organization’s attempt to adapt to a complex external environment.

We examine the effects of two environmental factors on infrastructure management quality in the states. Urbanization can be a major impetus for the development of a more developed and integrated infrastructure management system. Rondinelli (1971), for example, avers that urbanization and its negative consequences, such as congestion and physical deterioration, influenced policy makers at the federal, state, and local levels to create new structures to coordinate development issues and implement rational-comprehensive planning methods and techniques. Thus, urbanization serves as a catalyst for the development and integration of infrastructure management systems and processes. The variable percentage of population living in urban areas is included in the analysis to test the urbanization hypothesis.

Age of infrastructure is another environmental factor that can influence infrastructure management quality. Some regions of the country are hit harder by physical infrastructure problems than others, specifically older cities in the Frostbelt states. The aging of so many facilities and the need for careful maintenance and replacement of worn facilities create opportunities to apply the state of the art in asset management in the hopes of obtaining more efficient strategies for maintenance and modernization (ACIR, 1984). It is expected as follows:

Hypothesis 13-Hypothesis 14: States that are highly urbanized and those in the Frostbelt region have higher quality infrastructure management systems.

Data and Estimation Method

We test the hypotheses using the GPP infrastructure management grades. To simplify the analysis, letter grades are transformed into numeric values ranging from F = 1 to A = 12. We use the grades from the 2005 and 2008 GPP rounds. In addition, note that grades for 2008 actually reflect assessment of infrastructure management quality in the states in 2007, and 2005 grades represent 2004 assessments. Consequently, the data for the independent variables are from 2004 and 2007.

As the outcome variable has a lower limit of 1 and upper limit of 12, we estimate the model using pooled Tobit regression.¹ While the natural lower limit is 1, no state received a score lower than 3. However, as at least 2 states received the highest score of 12, we use Tobit with upper limit. We also run the model with ordered logistic regression, with the assumption that the GPP grades are ordinal. The results of the pooled Tobit and ordered logistic regressions are similar, and we discuss only the Tobit estimates.

The interpretation of Tobit coefficients is different from that of ordinary least squares (OLS) coefficients. An OLS coefficient represents the impact of an independent variable on the observed outcome variable, whereas a Tobit coefficient reflects effects on the latent dependent variable (see Note 1). We use the Tobit parameters primarily to confirm if the directions of the observed relationships are consistent with theoretical expectations. To extract more information from the Tobit coefficient, we use the McDonald and Moffitt (1980) decomposition framework and present two types of marginal effects. The first is conditional effects or the impact on the observed outcome variable of a one-unit change in the independent variable conditional on the observations being below the limit. This is simply the change in the GPP grades as a consequence of a change in an independent variable, holding other control variables constant, for states that did not receive the highest grade of A. The second type of marginal effect is the probability of being uncensored which is simply the probability that states will receive GPP grades lower than A as a consequence of a one-unit change in an independent variable, holding other variables fixed. Decomposing the effects gives us more information about the relationship between capital management quality, as measured by the GPP grades, and the independent variables.

Results

Because of its high dependence on severance tax revenues, Alaska tends to have extreme values for its year-end balances compared with other states. We run the model with and without Alaska, but this does not alter the results. The findings are also robust to other changes in the specification of the model, such as the inclusion of other fiscal condition measures, or different measures of urbanization. The final model includes all the original variables described previously and covers all the states except Nebraska which is the only state with a unicameral legislature.

Table 3 presents the maximum likelihood estimates from the pooled Tobit regression. We clustered the observations by states to address the issue of intracluster correlation and use Huber-White sandwich estimators to produce heteroskedasticity-robust standard errors (Wooldridge, 2002).²

Table 3

Results (Dependent Variable: GPP Infrastructure Management Grades)

		Descriptive statistics		Pooled ordered logistic	Pooled Tobit regression
Independent variables	Data source	M	SD	Coefficient (SE)	Coefficient (SE)	Conditional effects^a	Probability uncensored^b
Political factors
% employment in transportation	Bureau of Economic Analysis	3.21	0.73	0.091 (0.371)	0.151 (0.348)	0.144	−.002
% employment in manufacturing	Same	8.27	3.10	0.066 (0.132)	0.078 (0.109)	0.074	−.001
Divided government (1 = yes; 0 = otherwise)	Klarner (2007)	45^c	53^d	−0.192 (0.722)	−0.116 (0.560)	−0.111	.001
Divided legislature (1 = yes; 0 = otherwise)	Same	75^c	23^d	1.770** (0.767)	1.418*** (0.538)	1.311	−.029
Legislative term limits (1 = yes; 0 = otherwise)	Same	70^c	30^d	1.048* (0.643)	1.034** (0.512)	0.971	−.016
Legislative professionalization	Squire (2008)	0.18	0.11	−3.451 (2.333)	−3.451 (2.437)	−3.298	.036
Fiscal institutions
Strict antideficit rules (0-10, w/ 10 most stringent) 0 = otherwise)	ACIR (1995)	8.06	2.60	0.053 (0.139)	0.032 (0.113)	0.030	.000
TEL Stringency Index (0-3, w/ 3 most stringent)	National Conference on State Legislatures	1.06	1.05	−0.544* (0.304)	−0.536** (0.269)	−0.512	.006
Referendum-based debt limit (1 = yes; 0 = otherwise)	Kiewiet and Szakaly (1996)	58^c	42^d	0.305 (0.614)	0.037 (0.527)	0.035	.000
Prohibition of guaranteed debt (1 = yes; 0 = otherwise)	Same	82^c	18^d	−0.485 (0.544)	−0.661 (0.518)	−0.639	.005
Fiscal condition
Fund balance as a % of expenditures	National Association of Budget Officers	11.17	13.32	0.003 (0.024)	0.002 (0.020)	0.002	.000
Real total revenue per capita	Census of Governments	5.59	1.74	−0.110 (0.165)	−0.056 (0.139)	−0.054	.001
Growth rate of real intergovernmental revenue	Same	3.14	8.23	0.025 (0.021)	0.025 (0.018)	0.024	.000
Environmental demand factors
% population in urban areas	Census Bureau	71.69	14.83	0.051* (0.030)	0.054** (0.027)	0.051	−.001
States in Frostbelt (1 = yes; 0 = otherwise)	Same	72^c	28^d	−0.269 (0.593)	−0.301 (0.556)	−0.288	.003
Constant				—		3.669 (2.742)
N				98		98
Wald χ²				59.01		—
Prob > χ²				0.000		—
Censored observations				—		2
LR χ²				—		70.04
Prob > χ²				—		0.000
Log pseudolikelihood				—		–187.837

Note: GPP = GPP = Government Performance Project; TEL = tax and expenditure limits; ACIR = Advisory Commission on Intergovernmental Relations; M = Mean; SD = Standard Deviation. Standard errors (SE) are heteroskedasticity-robust and clustered by states.

Marginal effects of 1-unit change in x on y conditional on the observations being below the limit.

Marginal effects of 1-unit change in x on the probability of being below the upper limit.

Frequency if dummy = 0

Frequency if dummy = 1

Significant at 10%. **Significant at 5%. ***Significant at 1%, two-tailed.

Effects of Political Variables

The analysis shows that the variables percentage employed in manufacturing and transportation, and divided government have no statistically significant effects on the quality of infrastructure management in the states (contrary to Hypotheses 1, 2, and 3). Interestingly, a divided legislature, contrary to Hypothesis 4, is associated with higher GPP grades, and this relationship is highly significant at the .01 level. To more fully understand the relationship between infrastructure management grades and divided legislature, it is necessary to examine the marginal effects.

Marginal effects were calculated by holding continuous covariates at their means, and binary independent variables at zero. The conditional effects in Table 2 show that among states which have grades lower than A, those that have a divided legislature receive higher GPP grades—approximately 1.31 or one-and-a-third letter grade higher—compared with states where a single party controls both legislative chambers, holding other variables constant. However, the probability effects show that a divided legislature reduces the odds that states receive a grade lower than A by at least 2.9%.

The results suggest that legislative professionalization does not systematically affect capital management quality (contrary to Hypothesis 5). However, legislative term limits are significantly correlated with higher GPP grades (confirming Hypothesis 5). This is consistent with literature pointing to the positive policy and management effects of term limits (see Bourdeaux & Chikoto, 2008; Carey, Niemi, & Powell, 1998; Will, 1992). Holding other variables fixed, the conditional effects indicate that among states which did not receive the highest GPP grade, those states with legislative term limits have infrastructure management systems and processes that are 0.97, or almost a full letter grade, better than states without term limits. In addition, states with rules that limit the number of terms legislators can serve in office have a 1.6% lower probability of receiving a letter grade lower than A.

Effects of Fiscal Institutions

Among the fiscal institutions, it is only the coefficient of the TEL stringency index which is significant at the .05 level, and as expected, the negative sign indicates that more stringent TELs are associated with lower infrastructure management grades (supporting Hypothesis 8 but not Hypotheses 7 and 9). A one-unit increase in the TEL stringency index is associated with 0.51 unit or a half-letter grade penalty and a 0.6% increase in the probability of receiving a grade lower than A.

Effects of State Government Fiscal Condition

Not one of the fiscal condition indicators had statistically significant effects on the GPP grades (contrary to Hypotheses 10 through 12). Other possible measures of state fiscal condition, such as short-term end-of-year debt, and changes in tax revenues, or 1- to 2-year lags of these different measures, were introduced in different specifications of the model, but the results were not encouraging. The analysis failed to provide evidence to support the hypothesis that access to sufficient resources is a sine qua non for implementing management reforms. This is contrary to studies which find that slack resources matter for the adoption of innovative management practices in public organizations (Berry, 1994; de Lancer Julnes, & Holzer, 2001) and consistent with Tolbert, Mossberger, and McNeal’s (2008) conclusion that slack resources are not critical for state government officials’ decision to implement management innovations.

Effects of Organizational Environment

Not surprisingly, infrastructure management quality is influenced by external environmental demand factors. Specifically, states that are more urbanized have better infrastructure management systems (in support of Hypothesis 13). This is contrary to Boyne et al.’s (2005) finding that governments in less urbanized localities are more likely to implement management innovations.

It can be argued that public infrastructure issues are more severe and complex in urbanized states. For example, a bigger, more urbanized state such as California faces multifaceted infrastructure issues compared with a small and largely rural state such as Montana. The need to deal with such issues forces states to improve their infrastructure management systems. The effects of urbanization, however, are relatively marginal, with a 1-percentage point increase in state population residing in urban areas associated with a 0.05 improvement in infrastructure management grade and a 0.1% decrease in the probability of having a grade lower than the maximum. Finally, the results suggest that states in the Frostbelt region are also likely to have better infrastructure management systems, but this finding is not statistically significant (contrary to Hypothesis 14).

Discussion

Infrastructure management remains an understudied area in public administration and public financial management. This is unfortunate given the crucial role of public infrastructure in promoting economic growth and development (Munnell, 1992) and in fulfilling basic public health and safety needs (Pagano & Perry, 2008).

Provision of public infrastructure requires very large investments of public resources. Infrastructure represents a sunk cost that cannot easily or quickly be moved to another location or used for another purpose. In addition, efficient functioning of fixed capital requires long-term investment in their maintenance. Given the high value and long life-span of fixed assets, as well as the sunk costs involved, comprehensive and systematic planning, management, and maintenance efforts are clearly very important (Pagano & Perry, 2008; Steiss, 1975). This research explored the factors that explain why some states perform better in managing their fixed assets while others lag behind.

Three groups of findings from the empirical analysis deserve further discussion. First, the results for legislative term limits and divided legislatures point to the importance of legislative bodies in shaping management systems in state governments. For example, an important legislative contribution is the codification of accepted capital planning practices such as future infrastructure needs analysis, assessment of the condition of existing assets, prioritization criteria to determine capital project selection, and formal linkage between the capital plan and the budget. The GPP study showed that the leading states are more likely than midlevel and poor performers to have enacted long-term and comprehensive capital planning requirements as far back as the early 1990s.

Term-limits may create the condition for legislators to focus more on enacting policies that make infrastructure management more comprehensive and rational and focus less on parochial concerns. The nonsignificant effect of legislative professionalization, however, is quite perplexing. It is possible that legislators with more time and staff resources are more likely to micromanage infrastructure-related issues, restricting executive agencies’ autonomy to apply their expertise in addressing such issues (Bourdeaux & Chikoto, 2008). Another possibility is that because full-time legislators depend on their congressional salaries as a primary source of income, there is a powerful incentive for them to focus on policies or projects that will get them reelected, regardless of whether these actions benefit the state. Simply put, full-time legislators can retain their office and thus protect their main source of livelihood by prioritizing the demands of district constituents for particularistic goods (see Rosenthal, 1998).

Divided legislature has been previously associated with symptoms of poor governance including policy gridlock and destructive competition. Others argue that party competition in the legislature does not systematically affect adoption of innovative policies, at least at the federal level (Mayhew, 1991). The results of the empirical analysis in this research suggest that party competition in state legislatures matters, and it matters in a positive way. It is possible that control by different parties of the majorities in both legislative chambers produces a check-and-balance effect limiting success of legislators in seeking out district pork. A more optimistic interpretation is that competition facilitates vibrant and substantive policy debates that in turn can lead to positive changes in the way state governments plan and budget for their infrastructure needs. A different explanation is provided by Bourdeaux and Chikoto (2008), who argue that split control can make it more difficult for state legislatures to micromanage executive agencies. The implication is that giving agencies more autonomy and flexibility can encourage experimentation with management reforms.

Second, a result that should be of particular concern to state government officials is the effect of TELs. The empirical results show that fiscal institutions have produced consequences far beyond the intended purpose of controlling public sector fiscal profligacy. The effects of budget rules spill over in the area of infrastructure management. In particular, precisely because they put a tight leash on government’s ability to adjust revenue and expenditure decisions, an unintended impact is to constrain the capacity of public managers to respond to infrastructure-related issues. As the popularity of fiscal rules such as TELs increases, state governments should be wary of their unintended consequences.

Third, the finding that measures of fiscal condition and slack resources do not systematically affect infrastructure management grades (maintenance is weighted 30% of the final GPP grade) raises questions about the continued underinvestment by state governments in maintenance projects. In 2004, most states suffered from fiscal stress and underinvested in infrastructure maintenance. By 2007, state fiscal condition improved and yet the GPP results showed that a number of state governments continued to neglect their infrastructure maintenance needs. Clearly, encouraging state governments to channel more resources for infrastructure maintenance is difficult, whether in bad or good times. One problem is that, even though maintenance can extend the life of a fixed asset, maintenance-related activities provide less visibility, politically speaking, compared with the construction of new facilities. However, maintenance deferral only pushes back the time when the state eventually will be required to fund an even more costly repair or replacement project.

Conclusion

Using pooled Tobit and ordered logistic regression to analyze multistate, multiyear data, this research identified the determinants of infrastructure management quality across the states. The results of the exploratory empirical model suggest that divided legislatures, legislative term limits, less restrictive TELs, and urbanization are all positively correlated with higher GPP grades for capital management.

While this research has contributed to our knowledge of infrastructure management in the states, it has some limitations that can be addressed in future studies. Foremost among these is the need to examine the effects of administrative leadership, employee attitudes, training and skills, and other agency-level variables on capital management. Equally important is to understand whether similar factors influence state performance in the different areas of asset management. This requires undertaking separate analysis for each component of the GPP infrastructure management grade—capital planning, project monitoring, maintenance, internal cooperation, and intergovernmental coordination.

In addition, there is a need to further explore the insignificant findings from the empirical analysis. Only 4 out of 14 variables reached conventional levels of statistical significance. One explanation is the small sample used in the analysis which weakened the statistical power of the models. Thus, researchers can include additional years of data to the existing data set by using results of future GPP assessments. Future research can also focus on undertaking pairwise comparisons leading to full-blown comparative cases studies that explore, in depth, the key differences in environmental, political, and fiscal factors faced by leading states and those with a poor record in infrastructure management.

Footnotes

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

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Government Performance Project was supported through a grant provided by the Pew Charitable Trust (Grant # G7118). The views expressed in this paper are those of the authors.

Notes

References

Advisory Commission on Intergovernmental Relations. (1984). Financing public physical infrastructure A-96. Washington, DC: ACIR.

Advisory Commission on Intergovernmental Relations. (1995). Significant features of fiscal federalism M-197, 1. Washington, DC: Author.

Alt

J. A.

Lowery

R. C.

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

Baker

J. A.

Lambert

J. H.

(2001). Information system for risks, costs, and benefits of infrastructure improvement projects. Public Works Management & Policy, 5(3), 198-208.

Barrett

Greene

(2005, February). Grading the states 2005: A report card on government performance. Governing, 18(5), 24-95.

Berry

F. S.

(1994). Innovation in public management: The adoption of state strategic planning. Public Administration Review, 54, 322-330.

Bohn

Inman

(1996). Balanced budget rules and public deficits: Evidence from the U.S. states (NBER working paper No. W5533). Cambridge, MA: National Bureau of Economic Research. Retrieved from http://www.nber.org/papers/w5533

Borins

(2001). Public management innovation: Toward a global perspective. American Review of Public Administration,31(1), 5-21.

Borins

(2002). Leadership and innovation in the public sector. Leadership & Organization Development Journal,23, 467-476.

10.

Bourdeaux

Chikoto

(2008). Legislative influences on performance management reform. Public Administration Review, 68, 253-265.

11.

Boyne

G. A.

Gould-Williams

J. S.

Law

Walker

R. M.

(2005). Explaining the adoption of innovation: An empirical analysis of public management reform. Environment and Planning C: Government and Policy,23, 419-435.

12.

Brudney

J. L.

Hebert

F. T.

Wright

D. S.

(1999).Reinventing government in the American states: Measuring and explaining administrative reform. Public Administration Review 59(1): 19-30.

13.

Brudney

J. L.

O’Toole

Jr. Rainey

H. G.

(2000). Advancing public management: New developments in theory, methods, and practice. Washington, DC: Georgetown University Press.

14.

Burns

Stalker

G. M.

(1961). The Management of innovation. Oxford, UK: Oxford University Press.

15.

Carey

J. M.

Niemi

R. G.

Powell

L. W.

(1998). The effects of term limits on state legislatures. Legislative Studies Quarterly, 23, 271-300.

16.

Coggburn

J. D.

Schneider

(2003). The quality of management and government performance: An empirical analysis of the American states. Public Administration Review,63, 206-213.

17.

de Lancer Julnes

Holzer

(2001). Promoting the utilization of performance measures in public organizations: An empirical study of factors affecting adoption and implementation. Public Administration Review,61, 693-708.

18.

Donahue

A. K.

Selden

S. C.

Ingraham

P. W.

(2000). Measuring government management capacity: A comparative analysis of city human resources management. Journal of Public Administration Research and Theory,10, 381-411.

19.

Dowall

D. E.

(2001). Rethinking statewide infrastructure policies: Lessons from California and beyond. Public Works Management & Policy,6(1), 5-17.

20.

Ebdon

(2004). Capital management practices in U.S. counties. Public Works Management & Policy, 8(3), 192-202.

21.

Foster

Franklin

(2000). Expediting infrastructure projects through public agency partnering. Public Works Management & Policy,4, 260-264.

22.

Gaus

J. M.

(1947). Reflections on public administration. Tuscaloosa: University of Alabama Press.

23.

Gray

Lowery

(1988). Interest group politics and economic growth in the U.S. States. American Political Science Review, 82(1), 109-131.

24.

Groden

B. T.

(2000). Meeting great expectations through performance measures: Using performance measures to minimize risk and maximize the effectiveness of highway maintenance and operations contracts. Public Works Management & Policy,5(1), 13-25.

25.

Harder

B. T.

Pedersen

N. J.

Warne

Martin

(2005, July-August). On budget and on time: Using the power of people to get projects done. TR News, 239, 10-15.

26.

Hoffmann

Krumholz

O’Brien

Geyer

(2000). How capital budgeting helped a sick city: Thirty years of capital improvement planning in Cleveland. Public Budgeting and Finance, 20(1), 24-37.

27.

Horn

(1995). The political economy of public administration: Institutional choice in the public sector. Cambridge, UK: Cambridge University Press.

28.

Ingraham

P. W.

(2007). In pursuit of performance: Management systems in state and local government. Baltimore, MD: Johns Hopkins University Press.

29.

Ingraham

Joyce

Donahue

A. K.

(2003). Government performance: Why management matters. Baltimore, MD: Johns Hopkins University Press.

30.

Jimenez

(2009). Fiscal stress and the allocation of expenditure responsibilities between state and local governments: An exploratory study. State and Local Government Review, 41(2), 81-94.

31.

Judd

D. R.

Swanstrom

(2002). City politics: Private power and public policy (3rd ed.). New York, NY: Longman.

32.

Kellough

J. E.

Selden

S. C.

(2003). The reinvention of public personnel administration: An analysis of the diffusion of personnel management reforms in the states. Public Administration Review, 63(2), 165-176.

33.

Kiewiet

D. R.

Szakaty

(1996). Constitutional limitations on borrowing: An analysis of state bonded indebtedness. Journal of Law, Economics, and Organization, 12(1), 62-97.

34.

Klarner

(2007). State partisan balance, 1959 to 2007. State Politics and Policy Quarterly Data Source. Retrieved from www.ipsr.ku.edu/SPPQ/journal_datasets/klarner.shtml

35.

Lawrence

P. R.

Lorsch

J. W.

(1968). Organization and environment: Managing differentiation and integration. Cambridge, MA: Harvard University Press.

36.

Lufkin

P. S.

Desai

Janke

(2005). Estimating the restoration and modernization costs of infrastructure and facilities. Public Works Management & Policy,10(1), 40-52.

37.

Maxwell School of Citizenship and Public Affairs. (2003). Capital management. Paths to performance in state and local government. Syracuse, NY: Syracuse University Press.

38.

Mayhew

D. R.

(1991). Divided we govern: Party control, lawmaking and investigations, 1946-1990. New Haven, CT: Yale University Press.

39.

McDonald

J. F.

Moffitt

R. A.

(1980). The uses of Tobit analysis. Review of Economics and Statistics, 62, 318-321.

40.

McNeil

Tischer

M. L.

DeBlasio

A. J.

(2005). Asset management: What is the fuss? (Transportation Research Record No. 1729). Washington, DC: Transportation Research Board.

41.

Munnell

A. H.

(1992). Policy watch: Infrastructure investment and economic growth. Journal of Economic Perspectives,6(4), 189-198.

42.

Neumann

L. A.

Markow

M. J.

Lambert

L. H.

(2003, November-December). Transportation asset management: New guide advances state of the practice. TR News, 229, 6-11.

43.

Ostrom

(1990). Governing the commons: The evolution of institutions for collective action. Cambridge, UK: Cambridge University Press.

44.

Pagano

M. A.

(2008). Grading the states’ infrastructure management performance: The government performance project’s process and results. In Amekudzi

McNeil

(Eds.), Infrastructure reporting and asset management (pp. 73-80). Reston, VA: American Society of Civil Engineers.

45.

Pagano

M. A.

Perry

(2008). Financing infrastructure in the 21st century city. Public Works Management & Policy,13(1), 22-38.

46.

Petersen

J. E.

(2004). A leaky roof. Governing. Retrieved from http://www.governing.com/columns/public-finance/Leaky-Roof.html

47.

Poterba

(1994). State responses to fiscal crises: The effects of budgetary institutions and politics. Journal of Political Economy, 102, 799-821.

48.

Rondinelli

D. A.

(1971). Adjunctive planning and urban development policy. Urban Affairs Review, 7, 13-39.

49.

Rosenthal

(1998). The decline of representative democracy: Process, participation, and power in state legislatures. Washington, DC: CQ Press.

50.

Rubin

M. M.

Willoughby

(2009). Financial management grades for the states: A prospective use. Public Budgeting and Finance, 29(1), 49-67.

51.

Rueben

(1995). Tax limitations and government growth: The effect of state tax and expenditure limits on state and local government (MIT Working Paper). Cambridge, MA: Department of Economics.

52.

Schick

(1980). Budgetary adaptations to resource scarcity. In Levine

Rubin

(Eds.), Fiscal stress and public policy (pp. 113-134). Beverly Hills, CA: SAGE.

53.

Squire

(2007). Measuring state legislative professionalism: The squire index revisited. State Politics and Policy Quarterly,7, 211-227.

54.

Steiss

A. W.

(1975). Local government finance: Capital facilities planning and debt administration. Lexington, MA: Lexington Books.

55.

Stich

Eagle

(2005). Planning to include the public: Transportation policy implementation with effective citizen involvement. Public Works Management & Policy,9, 319-340.

56.

Tolbert

C. J.

Mossberger

McNeal

(2008). Institutions, policy innovation, and e-government in the American states. Public Administration Review, 68, 549-563.

57.

Transportation Research Board. (2004). Public benefits of highway system preservation and maintenance (NCHRP Synthesis Report No. 330). Washington, DC: Author.

58.

Vogt

A. J.

(2004). Capital budgeting and finance: A guide for local governments. Washington, DC: International City/County Management Association.

59.

Will

G. F.

(1992). Restoration: Congress, term limits, and the recovery of deliberative democracy. New York: Free Press.

60.

Wooldridge

J. M.

(2002). Econometric analysis of cross section and panel data. Cambridge, MA: MIT Press.

61.

Yusuf

J-E.

O’Connell

Hackbart

Wallace

(2008). An empirical examination of the statutory characteristics and effects of highway and transportation commissions on dot capital management capacity. Public Works Management & Policy, 12, 533-543.