What Drives Municipal Capital Investment? A Long-panel Data Analysis of U.S. Central Cities

Abstract

Cities across the United States are facing a severe infrastructure deficit. The challenge of financing infrastructure capital assets has emerged as one of the most urgent issues of the country. Using a panel data consisting of 100 American central cities from 1988 to 2012, this research reveals that city governments are responsive to sociodemographic forces and citizens’ demands, but their abilities to finance capital assets are constrained by city fiscal institutions and revenue capacity. This research also compares the determinants of different components of city capital outlays such as road and water utility capital expenditures.

Keywords

capital investment city United States panel data

In the United States, cities play a key role in funding, operating, and maintaining local roads and streets, bridges, airports, transit facilities, drinking water and sewer systems, and other types of capital assets (Chen and Bartle 2017). However, as widely publicized, cities across the United States are facing a serious infrastructure deficit and struggling to find new ways to finance the needed expansions, upgrades, and repairs. In 2015, more than half of U.S. city mayors highlighted infrastructure issues during their State of the City speeches (National League of Cities 2015). American Society of Civil Engineers (2017) estimated an infrastructure investment gap of $2.1 trillion between 2016 and 2025, and the failure of closing this gap could cost the nation almost $4 trillion in gross domestic product as well as a loss of 2.5 million jobs through 2025.

The urgent infrastructure issue reflects many factors. The fundamental causes perhaps relate to the flaws in current state and local capital budgeting processes. On the expenditure side, those processes tend to underinvest in infrastructure maintenance (Marlowe 2012). American infrastructure is aging and in great need of repair. Yet, building new projects is politically favored even though spending on proper maintenance of existing infrastructure is at lower costs. On the revenue side, as people drive less and cars become more fuel efficient, conventional tax bases for infrastructure are eroded. When elected officials are reluctant to raise taxes and other unfunded liabilities (such as pension and health care) drive governmental revenues away from infrastructure, there is no doubt that America is experiencing infrastructure deficit. Moreover, the investment gap was widening further in business cycles. Building new infrastructure with budget surplus during the economic booms might increase funding demand for maintenance during the subsequent downturns, which results in a “vicious cycle” that further exacerbates the infrastructure crisis (Marlowe 2012).

Although many government and media reports have discussed the aging and decaying American public infrastructure system, a significant research gap is that very few studies empirically examine the determinants of capital investment at the city level. Drawing from public finance theories (the classical median voter model), we construct an econometric model to test how the economic conditions, citizens’ political ideology, fiscal capacity, and institutional elements affect the patterns of urban capital investment. Our sample consists of 100 central American cities from 1988 to 2012. The empirical analysis identifies that city capital investment is significantly explained by varying sociodemographic factors, revenue capacity, and fiscal institutions. It also compares different determinants between city capital investment by main components (road capital outlay and water utility capital outlay).

This study contributes to the field of urban infrastructure finance in several important ways. First, this research focuses on a new institutional context of major American cities and represents one of the first attempts to systematically investigate municipal decisions on capital investment. Second, this study conducts a robust and extensive (-twenty-five years) panel data analysis that addresses limitations in the previous literature. Third, this study improves our understanding of the economic, political, and institutional factors underlying municipal capital investment decisions and informs city policy makers on urban infrastructure finance.

Determinant of State and Local Capital Investment

Capital investment accounts for a significant portion of public budgets. Its decision-making process has been an important topic in state and local government studies. In particular, the question of what factors driving state and local capital investment has received continuous attention. With different emphasis, previous studies suggest that macroeconomic conditions, sociodemographic characteristics, fiscal capacity, political and fiscal institutions, and previous capital investment decisions are likely to affect state and local capital investment decisions. This section reviews relevant literature and discusses the operationalization of the independent variables of this research.

Macroeconomic Environment

The exogenous economic conditions affect governmental capital investment decisions. A weak economic environment reduces the financing resources available for governments to invest in capital projects. Ho (2008) found that states slow their capital spending growth during economic downturns and increase the amounts of capital spending significantly during economic booms. Anecdotal evidence has shown that in response to economic recessions, one of the common cutback strategies adopted by state and local governments is postponing state and local capital project construction and cutting regular maintenance spending (e.g., Bartle 1996; Bumgarner, Martinez-Vazquez, and Sjoquist 1991; Chen 2016b). However, focusing on data from North Carolina municipalities, Rivenbark, Afonso, and Roenigk (2018) suggested that “most municipalities consistently invested in capital assets before, during, and after the Great Recession” at a similar rate (p. 402).

We use the variables of log of real per capita household median income (Ln Median Household Income) and unemployment rate (Unemployment %) to account for the variation in city economic conditions. Prior research identified higher levels of median household income, and lower levels of unemployment rate indicate robust economic environment and therefore increase the financing resources available for more capital spending (Arimah 2005; Pagano 2002; Fisher and Wassmer 2015; Holtz-Eakin 1991; Temple 1994). This positive effect occurs because a higher income level increases citizen demands for infrastructure services.

Fiscal Capacity

Fiscal capacity of a government is crucial in explaining state and local decisions on capital investment. Pagano (2002) conducted a trend analysis using the American city-level data between 1993 and 2000. He indicated that rapid growth in capital spending is mainly associated with the growth in city own-source revenues. Arimah’s (2005) research from a cross-sectional sample of 113 cities in developing countries in 1993 revealed that a robust own-source revenue capacity is positively associated with city infrastructure spending. In addition to own-source revenue capacity, other studies found that intergovernmental aid is a positive predictor of state and local capital investment (e.g., Bates and Santerre 2015; Congleton and Bennett 1995; Fisher and Wassmer 2015; Holtz-Eakin 1991).

To measure city fiscal capacity, we use the following two variables: the log of city real intergovernmental revenues per capita (Ln Intergovernmental Aid) and the log of the share of own-source revenues in city total revenue (Ln Own-Resource Rev). Both fiscal capacity variables are lagged for one year. We expect that cities receiving larger amounts of intergovernmental aid and having stronger own-source revenue capacity are associated with more capital investment.

Socio-demographic Characteristics

According to the median voter model, public expenditures reflect the levels desired by the median voter, which are a function of the median voter’s income, tax price, and taste for public goods and services. Local public goods and services are generally financed through the property tax system. Following Bates and Santerre (2015), the median voter’s tax price is measured as median house value divided by the total market value of all assessed properties in that city. We expect that a higher tax price reduces citizen demands for capital projects.

“Taste” for local public goods and services is typically proxied by a vector of population characteristics thought to be associated with demand for public services such as age, race, educational attainment, school attendance, and homeownership (e.g., Borcherding and Deacon 1972; Bergstrom and Goodman 1973; Rubinfeld and Shapiro 1989). Following previous literature on state and local capital investment, eight variables are employed in this study to control for city variation in sociodemographic conditions (e.g., Bates and Santerre 2015; Wang and Wu 2018; Fisher and Wassmer 2015). They include population density (Ln Pop Density), annual population growth (Pop Growth %), percentage of senior population (Elderly Pop %), percentage of white population (White Pop %), percentage of population with bachelor’s degree or above (Higher Education Pop %), poverty rate (Poverty %), homeownership rate (Homeownership %), and percentage of school-age kids (School-age Kids %).

Population density may drive the need for more infrastructure projects (Arimah 2005; Fisher and Wassmer 2015). White population, college-educated population, and homeowners may have a higher income level and can afford expensive capital projects. It is expected that they may have greater demands for capital projects. Low-income residents, senior adults, and school-age kids may have fewer capital needs (Fisher and Wassmer 2015). Capital expenditures often occur to expand service delivery in response to increased demand from population growth (Arimah 2005; Temple 1994; Fisher and Wassmer 2015). It is expected that faster population growth stimulates more capital expenditures.

Political and Fiscal Institutions

Capital spending decisions are not made in a political and institutional vacuum. Prior research confirmed that politics and fiscal institutions play a role in capital spending (e.g., Chen 2016a; Crain and Oakley 1995; Fisher and Wassmer 2015; Nunn 1996; Poterba 1995; Bruce et al. 2007). Fisher and Wassmer (2015) found that more ideologically, liberal states tend to make more capital investments. We use the variable of the share of Democratic presidential voters (Democratic Voter) to measure citizens’ liberal ideology. We expect that cities with liberal citizens tend to support more capital spending.

Prior studies found that various budgetary and fiscal institutions have differential impacts on state and local government capital finance. Poterba (1995) used a cross-sectional state government data in 1962 and found that states with separate capital budgets and those that do not require pay-as-you-go financing of capital projects have higher amounts of capital spending. Wang, Hou, and Duncombe (2007) confirmed that the imposition of general obligation (GO) bond debt referendum requirement constraints state debt financing capacity for capital projects. Meanwhile, the fiscal rule of tax and expenditure limits (TELs) appears no effect in limiting the use of current revenues for capital projects at the state level. By examining data of 100 large cities across the United States in five selected years over 1992–2012, Wang and Wu (2018) investigated the effects of local TELs and debt limits on municipal capital spending. Their findings are mixed: The coefficient on local TLE index is statistically significant in their ordinary least squares models but not significant in their panel regression models, but the effects of debt limits are only significant in the panel models.

Two fiscal institutions are employed in this study. The first one is the dummy variable of whether state constitutions and city ordinances impose a legal debt limitation on GO bonds (Debt Limit). The second is the state-imposed municipal TELs stringency index. The data of TELs index are drawn from Amiel, Deller, and Stallmann (2009). We expect that cities are fiscally constrained by the imposition of debt limit and TELs; therefore, these fiscal rules will make cities spend less on capital projects.¹ Some cities are merged cities/counties, which also means a different set of functional responsibilities. We created a dummy variable to control for the potential effect of city–county consolidation.

Capital Stock

Current capital outlay depends upon past levels of capital investment decisions or existing capital stock (Chen 2017; Fisher and Wassmer 2015; Holtz-Eakin and Rosen 1989). Two common measures of capital stock are identified in the current literature. They are the cumulative level of past capital investments with depreciation and the condition of current capital stock. Due to the difficulty of obtaining data on city capital asset conditions, we use the monetary value of capital stock.² The sign of capital stock is ambiguous. On the one hand, cities with higher values of capital stock may already have well-developed and maintained capital asset systems. Therefore, there is no need to make more capital investments in the future. On the other hand, larger amounts of capital stock may indicate that city residents and public officials are more willing to invest in capital projects. Hence, they will continue to support future capital investment.

Data and Method

Model Specification

The median voter model is a standard economic model of examining government fiscal choices (Borcherding and Deacon 1972; Holcombe 1989). This model has been widely used by researchers to examine the determinants of state and local expenditures on public services such as public education, public health, and social welfare. Recently, Fisher and Wassmer (2015) applied the median voter model to examine the determinants of state and local capital expenditures. Following their theoretical model of government capital investment, we specify that city capital investment is a function of existing capital stock, macroeconomic condition, fiscal capacity, sociodemographic demand variables, and political and fiscal institutional factors. The econometric model and estimation methods are provided in Supplemental Text 1.

Our dependent variables are city total capital investment, city road capital investment, and city water utility capital investment. These variables are standardized by city population and log-transformed. By definition, “capital outlay” means government direct expenditure for “purchase or construction, by contract or government employee, construction of buildings and other improvements; for the purchase of land, equipment, and existing structures; and for payments on capital leases” (U.S. Bureau of Census 2006, 5).

Supplemental Table 1 ranks cities based on the size of per capita total capital outlay on average from 1988 to 2012. It illustrates the wide variation in average total capital outlay per capita. For total capital spending, Atlanta, Georgia, and Orlando, Florida, are dramatic outliers (over $1,500). Even if one ignores these two cities, the annual city total capital outlay per capita varied over this period from $1,345 in Denver, Colorado, to less than $250 in Providence, Rhode Island, with a coefficient of variation of .38. Supplemental Figure 1 shows the trend of average city capital outlay from 1988 to 2012. On average, city governments reduce total capital spending during the two most recent recessions (1990–1991 and 2007–2009).

To bolster the analysis, we also utilize disaggregate city capital outlay variables by main components. It should be noted that despite the significant amount of variations in the types of assets maintained by cities, in our sample, nearly all central cities are responsible for roads, and most of them are for water utility services.³ Furthermore, road and water utility capital expenditures account for the majority of city capital investment. Therefore, it is valuable to explore the determinants of these two large components.

Data and Sample

We model on a city panel data set from the period of 1988 to 2012. Our main data set comes from three sources. First, data on city-level total expenditures, aggregate and disaggregate capital expenditures, revenues, and intergovernmental aid for years of 1988–2012 are hand collected from the publicly available, time-consistent, Annual Survey of Governments Historical Finance Database (U.S. Bureau of Census n.d.). This data set offers detailed, annual information on city fiscal activities. Second, data on city-level sociodemographic characteristics (employment rate, population, and race) come from the U.S. Bureau of Economic Analysis. Third, the Fiscal Policy Space (FPS) provides data on city fiscal rules and citizen ideology. The FPS data are compiled from a variety of sources for about 100 central cities of the largest metropolitan areas in the United States, including the Census Bureau, Comprehensive Financial Reports, state statutes, city ordinances, Newsbank (Access World News), and others (UIC College of Urban Planning and Public Affairs n.d.).⁴ We merged the FPS and Annual Survey of Governments Historical Finance databases, creating a balanced panel of 2,500 city-year observations from 100 American central cities across 1988–2012. Supplemental Table 2 presents variable definitions, data sources, and descriptive statistics.

Empirical Results

This study employs a two-way panel estimator with city and year dummies to control both city and year-invariant unobserved heterogeneity. Table 1 reports our results. The dependent variable in model 1 is the size of city total capital outlay. The dependent variable in model 2 is the size of city road capital investment. Model 3 is estimated with the dependent variable of the size of city water utility capital investment.

Table 1.

Panel Two-way Fixed Effects Regression Results.

Variables	Model 1	Model 2	Model 3
Variables	Total Capital	Road Capital	Water Capital
Economic and financial variables
Unemployment %	−0.115 (.384)	−.253** (.117)	−1.800* (1.050)
Ln Median Household Income	0.174 (.235)	.588** (.273)	−0.703*** (0.241)
Ln Capital Stock (t − 1)	−0.027 (.100)	−.068 (.108)	0.011 (0.010)
Ln Intergovernmental Aid (t − 1)	0.918*** (.092)	.911*** (.079)	1.290*** (0.261)
Ln Own-Resource Rev Capacity (t − 1)	0.213** (.088)	.245*** (.089)	0.540** (0.220)
Tax Price	−0.031*** (.008)	−.026** (.010)	−0.060*** (0.021)
Sociodemographic variables
Ln Pop Density	−0.039 (.101)	−.029 (.156)	0.500 (0.380)
Pop Growth %	0.055 (.051)	.057 (.046)	−0.060 (0.134)
White Pop %	0.234 (.151)	.340* (.176)	−0.499 (0.300)
Elderly Pop %	−0.705 (.437)	.127 (.524)	−4.41*** (1.402)
Poverty %	−1.251** (.486)	−.841 (.601)	0.409 (1.616)
Homeownership %	0.021 (.159)	−.143 (.193)	1.433 (0.900)
Higher Education Pop %	−0.670** (.312)	−.919** (.432)	2.400*** (0.888)
School-age Kids %	0.233 (.348)	−.076 (.310)	1.601 (1.170)
Political and fiscal institutions variables
Democratic Voter	0.030 (.067)	−.005 (.065)	0.240** (0.120)
Debt Limit	−0.085*** (.015)	−.094*** (.015)	−0.030 (0.030)
State-imposed Municipal TELs Index	−0.110** (.042)	−.154** (.061)	−0.109 (0.170)
Consolidated City–County Government	−0.133 (.080)	−.057 (.095)	−0.440*** (0.150)
City fixed effects	Y	Y	Y
Year fixed effects	Y	Y	Y
N	2,500	2,500	2,000
R ²	.2585	.3585	.1680

Note: The sample size in model 3 is adjusted to exclude the cities without water responsibility. The variables of Capital Stock for models 2 and 3 are adjusted for road capital stock and water capital stock, respectively.

*p < .10.

**p < .05.

***p < .01.

In model 1, the variables of intergovernmental aid and city own-source revenue capacity are estimated as expected. These findings infer that a 10-percent increase in intergovernmental aid per capita lagged results in a 9.2-percent (67 dollars) increase in city total capital outlay per capita. Similarly, a 10-percent increase in city own-source revenue capacity (one-year lag) is associated with a 2.13-percent increase (or 15 dollars) in real per capita city total capital outlay. The variable of tax price is negative and statistically significant. Its negative sign indicates that higher tax prices are associated with lower amounts of total capital investment. Concerning sociodemographic variables, we found that the variables of poverty rate and well-educated residents are negatively associated with the size of city total capital outlay. In terms of variables of political and fiscal institutions, Debt Limit has a negative and statistically significant coefficient. Relative to the absence of a debt limit, the presence of a debt limit reduces per capita total capital spending by 56 dollars. Similarly, state-imposed municipal TELs stringency index exerts a negative impact on the size of real per capita total capital outlay.

Model 2 reports the results of analysis of city road capital investment. Among the economic variables, unemployment rate is statistically significant at the 5-percent level. Its negative signs indicate that deteriorating economic conditions decrease the size of city road capital investment. On the contrary, the positive coefficient of median household income indicates that cities with more affluent residents are associated with a larger size of road capital investment. The positive signs of intergovernmental revenues and own-source revenues indicate that cities receiving larger amounts of intergovernmental aid or having stronger own-source revenue capacity are associated with larger sizes of road capital investment. The variable of tax price is negative and statistically significant, which indicates that higher tax prices are associated with lower amounts of road capital investment. In terms of sociodemographic variables, it is found that the variable of white population increases the size of road capital outlay, while the variable of well-educated population reduces the amount of city road capital investment. Regarding the political and fiscal institution variables, Debt Limit and state-imposed TELs on municipalities have negative and statistically significant coefficients, inferring that stringent fiscal rules constrain the cities’ capacity of investing new roads.

Model 3 reports the analysis of city water utility capital investment.⁵ Among the economic variables, we find that unemployment rate and median household income decrease the size of water utility capital investment. The variables of intergovernmental revenues and own-source revenue capacity are positive and statistically significant, suggesting that better fiscal capacity increases the sizes of water utility capital investment. The variable of tax price is negative and statistically significant, indicating that higher tax prices are associated with lower amounts of water utility capital investment. For the sociodemographic variables, we find elderly population percentage decreases the size of water utility capital investment, while well-educated population percentage increases the investment. Regarding the political and fiscal institution variables, cities with more liberal citizens are associated with larger sizes of water utility capital investment. Different from previous results, the coefficients of debt limit and state-imposed TELs are insignificant, indicating that they might not be effective in constraining municipal capital investment on water utility. The possible reason is that city water utility infrastructure is largely funded by special water utility charges rather than relying on general tax revenues. Thus, the imposition of TELs seems to have little impact on water utility systems. In addition, city governments often issue revenue bonds backed by water utility revenues to finance capital investment. Hence, the levy of GO bonds limits may not hurt city government’s capacity to upgrade capital facilities. The variable of consolidated city–county government has a negative and statistically significant sign. It implies that merged cities are associated with lower amounts of water utility capital investment.

Discussion and Conclusion

The urban infrastructure challenge has emerged as one of the most urgent issues facing the country. In the academic literature, there exists a paucity of empirical studies on what determines capital spending at the U.S. city level. Given the relatively large amount of city capital investment and the vital role of public infrastructure in promoting economic development, understanding urban capital investment decisions is valuable. Using a sample of the 100 U.S. central cities from 1988 to 2012, this research examines the various factors influencing aggregate and disaggregate urban capital investments.

Some of our results have been discussed in previous literature. We reconfirmed the results with better data and methods. The key findings show that (1) social–economic factors such as tax price, poverty rate, and educational attainment affect the aggregate urban capital investments. This infers that city governments are responsive to citizens’ preferences when they make capital investment decisions. The results are consistent with previous findings (e.g., Temple 1994; Arimah 2005; Fisher and Wassmer 2015). (2) The fiscal capacity of city governments (intergovernmental aid and own-source revenue capacity) plays a strong role in capital investment; previous analyses also indicated similar results (e.g., Pagano 2002; Arimah 2005; Bates and Santerre 2015; Congleton and Bennett 1995; Fisher and Wassmer 2015; Holtz-Eakin 1991). (3) Even though previous findings on the effects of fiscal institutions are mixed (e.g., Wang, Hou, and Duncombe 2007; Wang and Wu 2018), our results show that the fiscal institutions of debt limit and TELs constrain city governments’ capacities to invest in capital assets, particularly for city roads.

More intriguingly, we also concluded with some new remarks by delving into different determinants of city capital spending across the aggregate and disaggregated categories. First, some important differences appear on the determinants of city road capital outlays versus water utility capital outlays. Median Household Income is positively associated with city road capital outlay but negatively associated with city water utility capital per capita. This indicates that wealthier cities invest more on roads but less on water utility capital projects. Similarly, Higher Education % shows opposite signs for road and water utility capital expenditures.

Second, we also find some variables are significant determinants for city road outlay but not for water utility capital outlay. Those variables include Debt Limit and State-imposed Municipal TELs Index. City roads are usually funded by general tax revenues and financed by GO s bonds. Hence, debt limits on general bonds issuance and state-imposed fiscal limits on municipal revenues and expenditures restrain city governments’ capacity in investing in road infrastructure. In contrast, funding and financing water utility infrastructure rely more on user-fees charges and revenue bonds. Therefore, the fiscal rules of debt limits and state-imposed municipal TELs do not constrain municipal capital investment in water utility infrastructure. Additionally, being a Consolidated City–County Government reduces water utility capital outlay but not for road capital outlay. Researchers and practitioners should be cautious about the different driving factors when they are considering various types of capital outlays.

Our analysis upgrades existing literature on the inquiry of determinants of state and local capital investment. It utilizes a panel data set of 100 central cities across the United States. The sample cities are contended to be representative of a sizable portion of the municipal sector across the United States (Pagano et al. 2016). It aims to address previous literature’s limitation focusing on municipalities in a single state (e.g., Nunn 1996). Our data set contains extensive twenty-five-year data. With such a panel data structure, we are able to control both city- and year-invariant unobserved heterogeneity using two-way panel fixed effect regression. Thus, it also addresses previous literature’s limitations resulting from relying on cross-sectional data (e.g., Arimah 2005; Bumgarner, Martinez-Vazquez, and Sjoquist 1991). Wang and Wu (2018) conducted an analysis using similar sample cities but only used selected five-year data (1992, 1997, 2002, 2007, and 2012). In contrast, our research is based on a twenty-five-year panel data from 1988 to 2012. The dependent variable in Wang and Wu’s (2018) research is annual capital expenditure from government funds. However, the measurement of their dependent variable is subject to key limitations because it only captures a portion of city total capital outlay and misses out the significant capital outlay in city governments’ business-type funds (e.g., water/wastewater). Instead, we refine their measurement and use the aggregate and disaggregate capital outlay data from the U.S. Census Bureau.

Besides those methodological improvements, our analysis also offers new findings. Previous literature on local capital spending exclusively focused on aggregate capital spending (e.g., Wang and Wu 2018). Our research extended to disaggregate capital investments including road capital outlay and water utility capital outlay. It helps to understand the different driving factors for various types and measurements of city capital expenditures. Furthermore, our analysis presents a more comprehensive model by including an important cost variable in the model (median voters’ tax price) that Wang and Wu (2018) suggested for further study in their conclusion.

This study has broad policy applications. First, our empirical results indicate that an increase in city governments’ intergovernmental aid will increase their capital investments. However, “declining funding, increasing mandates, and misaligned priorities at the federal and state levels” have increasingly made cities greater fiscal responsibility for infrastructure investment (National League of Cities 2016, 2). To address the challenges of funding and financing local infrastructure, cities are encouraged to seek more federal and state infrastructure funding sources. Second, capital projects are often large and expensive, requiring significant amounts of money to be raised and spent on building and operating projects. Maintaining a robust own-source revenue capacity is favorable for cities to make a sustained investment in capital assets. Third, we find that the fiscal institutions of debt limits and TELs have a restrictive effect on city capital investment, particularly for road capital outlay. Without the authorities to raise adequate revenues, cities may not be able to address the vast infrastructure challenges. State and local policy makers should be aware of the consequences of strict fiscal institutions in constraining local infrastructure investment. Besides, as states become increasingly constrained by rising Medicaid and pension costs, a negative spillover is limited intergovernmental aid that could stimulate local capital investment. Finally, we also find different capital outlays (road vs. water utility) may have different determinates. Practitioners should be cautious when they make inferences based on the studies.

Although our findings are intriguing, there are some limitations. First, our major findings may only apply to U.S. central cities. It may raise the issue of the external validity of our research. Future studies may need to examine local capital investment in small and rural communities. Second, due to data limits at the city level, cost factors such as land price and labor cost are not incorporated in our model estimation. This may cause relatively low R ²s for some models. It is suggested to include those possible omitted variables in future analysis. Furthermore, the impact of higher education was unexpected for total and roads capital outlays. This is an area for future research. In addition, cities vary in terms of the types of capital outlays they provide (e.g., hospitals, airports, ports, power plants). One notable limitation in the total capital outlay model is the inclusion of a capital stock measure for aggregating all kinds of capital outlays. It should also be noted that our current measurement of capital stock is limited due to data availability. We call for future research to develop a more robust measurement of capital stock.

Supplemental Material

Supplemental Material, Figure_1_Supplement - What Drives Municipal Capital Investment? A Long-panel Data Analysis of U.S. Central Cities

Supplemental Material, Figure_1_Supplement for What Drives Municipal Capital Investment? A Long-panel Data Analysis of U.S. Central Cities by Can Chen, Yanbing Han and Howard A. Frank in State and Local Government Review

Supplemental Material

Supplemental Material, SLGR_19-0057R3,_Text_1_Supplement_(1) - What Drives Municipal Capital Investment? A Long-panel Data Analysis of U.S. Central Cities

Supplemental Material, SLGR_19-0057R3,_Text_1_Supplement_(1) for What Drives Municipal Capital Investment? A Long-panel Data Analysis of U.S. Central Cities by Can Chen, Yanbing Han and Howard A. Frank in State and Local Government Review

Supplemental Material

Supplemental Material, Table_1_Supplement - What Drives Municipal Capital Investment? A Long-panel Data Analysis of U.S. Central Cities

Supplemental Material, Table_1_Supplement for What Drives Municipal Capital Investment? A Long-panel Data Analysis of U.S. Central Cities by Can Chen, Yanbing Han and Howard A. Frank in State and Local Government Review

Supplemental Material

Supplemental Material, Table_2_Supplement - What Drives Municipal Capital Investment? A Long-panel Data Analysis of U.S. Central Cities

Supplemental Material, Table_2_Supplement for What Drives Municipal Capital Investment? A Long-panel Data Analysis of U.S. Central Cities by Can Chen, Yanbing Han and Howard A. Frank in State and Local Government Review

Footnotes

Declaration of Conflicting Interests

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

Funding

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

ORCID iD

Yanbing Han

Supplemental Material

Supplemental material for this article is available online.

Notes

References

American Society of Civil Engineers. 2017. 2017 infrastructure report card. Washington, DC: American Society of Civil Engineers.

Amiel

Lindsay

Deller

Steven

Stallmann

Judith

. 2009. The construction of a tax and expenditure limitation index for the U.S. Department of Agricultural and Applied Economics Staff Paper Series 536 , University of Wisconsin-Madison, Wisconsin.

Arimah

Ben C.

2005. What drives infrastructure spending in cities of developing countries? Urban Studies 42:1345–68.

Bartle

John

. 1996. Coping with cutbacks: City response to aid cuts in New York State. State and Local Government Review 28:38–48.

Bates

Laurie J.

Santerre

Rexford E.

. 2015. The demand for municipal infrastructure projects. Public Finance Review 43:586–605.

Bergstrom

Theodore C.

Goodman

Robert P.

. 1973. Private demands for public goods. The American Economic Review 63: 280–296.

Borcherding

Thomas E.

Deacon

Robert T.

. 1972. The demand for the services of non-federal governments. The American Economic Review 62:891–901.

Bruce

Donald

Carroll

Deborah A.

Deskins

John A.

Rork

Jonathan C.

. 2007. Road to ruin? A spatial analysis of state highway spending. Public Budgeting & Finance 27:66–85.

Bumgarner

Mary

Martinez-Vazquez

Jorge

Sjoquist

David L.

. 1991. Municipal capital maintenance and fiscal distress. The Review of Economics and Statistics 73:33–39.

10.

Chen

Can

. 2016a. Banking on infrastructure: Exploring the fiscal impacts of State Infrastructure Bank (SIB) loans on leveraging state and local transportation investment. Public Budgeting & Finance 36:94–113.

11.

Chen

Can

. 2016b. Effects of fiscal stress on state highway infrastructure finance: A composite index approach. Municipal Finance Journal 37:1–28.

12.

Chen

Can

. 2017. Does money matter for infrastructure outcomes? The effects of public infrastructure finance on state infrastructure quality. Journal of Public Budgeting, Accounting and Financial Management 29:375–408.

13.

Chen

Can

Bartle

John

. 2017. Infrastructure financing: A guide for local government managers. ICMA and GFOA Policy White Paper. Washington, DC: International City/County Management Association.

14.

Congleton

Roger D.

Bennett

Randall W.

. 1995. On the political economy of state highway expenditures: Some evidence of the relative performance of alternative public choice models. Public Choice 84:1–24.

15.

Crain

W. Mark

Oakley

Lisa K.

. 1995. The politics of infrastructure. The Journal of Law & Economics 38:1–17.

16.

Fisher

Ronald C.

Wassmer

Robert W.

. 2015. An analysis of state-local government capital expenditure during the 2000s. Public Budgeting & Finance 35:3–28.

17.

Alfred Tat-Kei

. 2008. State highway capital expenditure and the economic cycle. International Journal of Public Administration 31:101–16.

18.

Holcombe

Randall G.

1989. The median voter model in public choice theory. Public Choice 61:115–25.

19.

Holtz-Eakin

Douglas

. 1991. Bond market conditions and state-local capital spending. National Tax Journal 44:105–20.

20.

Holtz-Eakin

Douglas

Rosen

Harvey S

. 1989. The “rationality” of municipal capital spending: Evidence from New Jersey. Regional Science and Urban Economics 19:517–36.

21.

Marlowe

Justin

. 2012. Capital budgeting and spending. In The oxford handbook of state and local government finance, ed. Ebel

Robert D.

Petersen

John E.

, 682–726. New York: Oxford University Press.

22.

National League of Cities. 2015. 2015 state of the cities. Washington, DC: National League of Cities.

23.

National League of Cities. 2016. Paying for local infrastructure in a new era of federalism. Washington, DC: National League of Cities.

24.

Nunn

Samuel

. 1996. Urban infrastructure policies and capital spending in city manager and strong mayor cities. The American Review of Public Administration 26:93–112.

25.

Pagano

Michael A.

2002. Municipal capital spending during the “boom.” Public Budgeting & Finance 22:1–20.

26.

Pagano

Michael A.

Hoene

Christopher

Yonghong

Wang

Shu

Shi

Yazici

Nisa

. 2016. The fiscal policy space of cities: A comprehensive framework for city fiscal decisionmaking. Chicago: University of Illinois at Chicago.

27.

Poterba

James

. 1995. Capital budgets, borrowing rules, and state capital spending. Journal of Public Economics 56:165–87.

28.

Rivenbark

William C.

Afonso

Whitney

Roenigk

Dale J.

. 2018. Capital spending in local government: Providing context through the lens of government-wide financial statements. Journal of Public Budgeting, Accounting and Financial Management 30:402–14.

29.

Rubinfeld

Daniel L.

Shapiro

Perry

. 1989. Micro-estimation of the demand for schooling: Evidence from Michigan and Massachusetts. Regional Science and Urban Economics 19: 381–398.

30.

Temple

Judy

. 1994. The debt/tax choice in the financing of state and local capital expenditure. Journal of Regional Science 34:529–47.

31.

UIC College of Urban Planning and Public Affairs. n.d. Fiscal policy space. http://www.srl.uic.edu/fiscalpolicyspace/index.php (accessed January 7, 2018).

32.

U.S. Bureau of Census. n.d. Annual survey of state and local government finances. https://www.census.gov/programs-surveys/gov-finances.html (accessed January 17, 2018).

33.

U.S. Bureau of Census. 2006. Government finance and employment classification manual. Washington, DC: U.S. Bureau of Census.

34.

Wang

Wen

Hou

Yilin

Duncome

William

. 2007. Determinants of pay-as-you-go financing of capital projects: Evidence from the states. Public Budgeting & Finance 27:18–42.

35.

Wang

Wen

Yonghong

. 2018. Why are we lagging behind? An empirical analysis of municipal capital spending in the United States. Public Budgeting & Finance 38:76–91.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.51 MB

0.02 MB