Comprehensive Annual Financial Reports

Comprehensive Annual Financial Reports of the 35 largest popu- lation American cities from 2005 to 2011 to examine how these cities managed the Great Recession, which was a global macroeconomic shock particularly damaging to the housing sector. While broader surveys of local government suggest that the Great Recession has been associated with substantive revenue declines, particularly via the property tax, the Comprehensive Annual Financial Reports data indicate that large cities remained relatively stable in revenue by using higher property taxes to com- pensate for other revenue declines. Furthermore, these cities were able to rely on their net assets to engage in deficit spending. These findings indicate that cities are relying on traditional strengths of local governments, but are also able to engage in the deficit spending that is typically characteristic of national governments. It also seems to be the case that grants for capital projects were largely transferred into highly liquid and spendable assets.

1. INTRODUCTION

This paper explores the fiscal sustainability of America’s 35 largest cities in the years surrounding the Great Recession, using public financial records to assess their ability to cope with fiscal stress. Specifically, these cities were the 35 largest cities according to their population in 2005, with their financial information being derived from Comprehen- sive Annual Financial Reports (CAFRs) for fiscal years 2005 through 2011.1 This highly detailed, audited, data reveal how the city governments experienced changes in different revenue sources during the period as these resources entered into different components

*We appreciate the “Nations and regions/cities after the Great Recession: Austerity or Creative De- struction and Boom?” conference co-organized by the Journal of Regional Science and the International Institute of Housing and Urban Development Studies. We also appreciate helpful comments and sugges- tions from three anonymous referees, Mark Partridge, Skip Krueger, Bill Fischel, and participants at both the 2013 ABFM and NTA conferences.

Received: March 2013; revised: December 2013; accepted: January 2014. 1This list was provided by the 2006 Statistical Abstract of the United States from the U.S. Census

Bureau.

C© 2014 Wiley Periodicals, Inc. DOI: 10.1111/jors.12117

113

 

 

114 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

of their budgets and primary functions. This allows us to observe how local government responded within the budgeting system to navigate through the Great Recession.

The study is motivated by the increasingly important role cities have played, and are expected to continue playing, in the economy. The last several decades have seen a re- markable transition in the spatial configuration of the world’s economy. Within the United States, it has been estimated that cities with more than 150,000 people accounted for 80 percent of the population and 84 percent of gross domestic product in 2010 (Manyika et al., 2012). The globalization of the market place, facilitated in part by declining trans- portation costs and reduced barriers to international trade, seems to have made it possible for many industries to produce anywhere, diminishing the significance of national and state borders. Yet, the consequence of this trend has been to encourage the concentration of economic activity into a relatively small set of cities. A possible interpretation of the modern era is that we are witnessing the demise of the nation-state and the rise of the city.2

If these trends remain a long-run pattern of economic activity, one issue of concern is whether or not the American public finance system is structured to sustainably produce public goods and services. Because taxing powers are inversely related to the mobility of the economic base, the core of the city-state economic divergence strikes directly at the nature of financing a federalist system of government, in which there is a shared sovereignty among several constituent political states over a common economic base.3 A federalist system requires some degree of fiscal autonomy for each sovereign unit, which implies that each level of government needs the independent ability to extract resources from the economic actors they share in common. While only time will ultimately reveal the long-run adaptability of the American federalist system, the Great Recession offers an interesting stress test of these concerns for large cities because it was a global phenomenon that was particularly damaging to wealth in the property sector of the economy. Large cities serve as the residence to many economic actors that compete in a global marketplace, and the municipal governments themselves are heavily reliant on property taxes.

This paper proceeds in two substantial parts. The first part of the paper describes and contrasts the experience of these 35 large cities with the previous findings from broader surveys of local government. We focus our study on the general purpose cen- tral city government as the unit of the analysis, which serves as the urban core of the metropolitan area and is the point of coordination among overlapping units. The general narrative from surveys of practitioners and academics using other data has observed a substantial amount of fiscal stress for American local governments throughout the Great Recession. The CAFR data we have collected suggest that big city governments have fared surprisingly well despite substantive downturns in the local economies, and we unpack where the negative fiscal shocks have occurred within their budgets and evaluate the fiscal sustainability of their responses. In the second part of this paper, we solidify these observations by putting forward and estimate a simple model based on the fiscal options available to city fiscal officers. Interestingly, the results of these analyses suggest that big city governments have the flexibility to raise their total property tax levy, and to draw down their net assets. While raising the property tax levy is a rather traditional local

2For the decline of nation-state power in the context of taxation, see the works of Avi-Yonah (2000) and Hines and Summers (2009). Regarding the rise and importance of city economies, see Brenner (1998), Glaeser and Kohlhase (2004), Glaeser and Resseger (2010), and Glaeser and Saiz (2004).

3In the United States, any given household will be a political member of at least three types of sovereign governments: local, state, and federal. For a fascinating account of the history and institutions of politico-fiscal authority in the Americas, see Kim and Law (2012). See Fisher (1996) for a thorough review of the property tax in the fiscal federalism of the United States.

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 115

public finance response, depleting net assets through one-time sales or debt issuance is a more unconventional practice for a typical local government. Deficit spending is actually much more common among national governments, and not usually associated with local governments.

Since a significant contribution of this paper comes from the exploration and uti- lization of the CAFRs data, the next section describes the importance and uniqueness of CAFR data relative to other, more popularly employed, sources of data. More importantly, we argue that the research question of this paper cannot be adequately addressed with the conventional and popularly accessible data sources. Section 3 describes other recent research on local government fiscal affairs during the Great Recession and contrasts it with important findings from the CAFR data on the condition of our selected large cities. Section 4 models and estimates the regression model of local government fiscal reac- tion during this period, and Section 5 concludes by summarizing the various findings throughout the paper and provides discussion.

2. DATA FROM COMPREHENSIVE ANNUAL FINANCIAL REPORTS

This paper will draw upon data from city CAFRs to examine how large American cities have managed finances through the Great Recession. Conventionally, the primary dataset for academic research on local government revenues comes from the U.S. Census Bureau in the Annual Survey of the Census of Governments (ASCG). As will be discussed, these data are insufficient in detail for drawing definitive conclusions for city government on fiscal sustainability. Better suited for governmental fiscal sustainability is the National League of Cities (NLC), which produces a survey on city fiscal conditions on an annual basis, and then provides a summary report of the data. Although influential and widely known due to the NLC summary reports, it is not widely employed in academic research because of the data’s proprietary nature. In order to understand the contribution associated with the CAFR data, it is helpful to compare the data from ASCG and NLC in terms of their nature and limitations.

The ASCG provides full coverage of all local jurisdictions every five years, with local governments selected using probability proportion to size sampling in the intervening years. The collected data are used to infer state-level aggregates of local government activity that are downloadable from the Census Bureau’s Governments Division, as well as selected city government statistics that appear in the Statistical Abstract of the United States. For a local government officer that receives the survey (U.S. Census Bureau Form F-28), they are given a set of general instructions with five enumerated points, three of which are of direct relevance for contrasting with CAFR data (#3 and #5 are omitted here):

1. Please report amounts covering all funds and accounts of your government except for any employee retirement funds administered by your government. Include bond redemption and interest funds, and construction or development funds, as well as current funds. Exclude refunds and transfers between funds or accounts of your government.

2. You may report on either a cash or accrual basis.

4. Do not delay reporting to await finally audited figures, if substantially accurate figures can be supplied on a preliminary basis.

In the enumerated instructions above, the underlined portions of text is our own added emphasis. While the ASCG data are reasonably good at determining the type of instrument generating revenue, a point of substantial interest for the taxpayers and

C© 2014 Wiley Periodicals, Inc.

 

 

116 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

residents, it is less indicative of the fiscal changes as they are experienced by the govern- mental units. The classification system is different between revenues and expenditures: the revenue information of Census data is gathered by instrument and the expenditure information is grouped by functions with no clear information on where the resource inflows and outflows are located within the fund structure. Different from private organi- zations, the financial resources of the public sector are collected, allocated, and managed based on different funds. This is an important piece of information for the public in order to understand the structure and function of a government. CAFR data track revenues as they flow into the governmental funds with an emphasis on how they can be spent. All inflows of resources must be accounted for, and resources for specified purposes must be designated for its corresponding fund. The fund with the most flexibility in terms of the purposes for which the government may direct the resources is the general fund, and it is often the largest of the funds.

Second, there is no distinction in ASCG as to whether or not the amounts are re- ported on a cash or accrual basis.4 Unfortunately, these two accounting concepts are very different, and comparing across governments operating under different accounting basis can be very misleading.5 The cash basis of accounting only records revenues or expen- ditures at the time the cash enters or leaves the government balance. The accrual basis of accounting recognizes all flows of resources when they are earned or spent during the year regardless of when the actual cash flow will take place. To compare to a household, a purchase made on a credit card would not be recognized in a cash basis of accounting system until the money was withdrawn from their bank accounts. An accrual basis of accounting recognizes that credit card purchases represent a reduction in the availability of resources at the moment even if the cash has not yet been withdrawn to pay for it. This simple motivation has made accrual basis of accounting a popular reform in the movement to encourage fiscal transparency to the public. A government which receives a service with an agreement to pay at some future point in time can improve its current cash position, but not when it is operating under an accrual basis. Therefore, accrual basis represents a broader and more accurate picture of the financial condition of the government.

Within CAFRs, we draw upon two types of data, governmental fund and government- wide financial statement data. The governmental fund data report on modified-accrual accounting basis and reflects the current financial resources and liabilities of the govern- ment. Governmental funds include the general fund and other funds associated with general governmental activities, such as special revenue funds. Government-wide fi- nancial statements report on the long-term resources and liabilities of the government. Government-wide financial statements cover the entire government, both governmental funds and proprietary funds. The different funds are combined, or reconciled, and re- ported in the government-wide financial statements on a full-accrual basis. Therefore, the government-wide financial statements reflect the broadest and longest-term measure of the finances of the government.

Finally, the information contained in individual CAFRs is audited by an independent certified public accountant (CPA) on an annual basis. The auditing firm certifies that the financial statements are presently fairly in conformity with generally accepted accounting

4Most local governments have historically uses a cash basis of accounting because of its simplicity. The larger a government becomes, the more likely it is to employ accountants that track resources on an accrual basis.

5Reconciliation is the process of adjusting numbers between financial statements with different basis of accounting. In our example, through the reconciliation process governmental fund numbers are converted from the modified accrual basis to an accrual basis and reported on the government-wide financial statements.

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 117

principles (GAAP) and free from material misstatement (in a clean or unqualified audit opinion), and if not, where and why not (in a qualified audit opinion). Auditing provides a professional assessment beyond self-reported data that the government’s financial infor- mation is accounted for and reported based on GAAP, and therefore, that the information reported by the government is accurate, consistent, and comprehensive.

Another survey to contrast with the CAFR data of this paper is the one produced by the NLC. The NLC survey is popularly known as the best national representation of the current fiscal conditions of local governmental units, and is better suited to that purpose than the ASCG. The NLC survey is conducted annually, and though the underlying micro- data it is proprietary, annual reports summarizing the data are publically available.6 The approach to determining which cities are included in the underlying data has changed between 2004 and 2012, the survey can be described as being sent to a universal sample of the largest cities, and then a random sample of city governments with populations greater than 10,000.7 Each year, a little more than 1,000 cities would receive the survey, with responses being received by 234–385 cities, and generally the response rate would increase with the city population. The surveys contain self-reported information on the change in general fund revenue by major tax type, as well as some the city financial offi- cers’ responses to questions intended to gauge their own local perceptions of fiscal stress. The limitations of the NLC survey include the concern that sample selection bias may affect which cities respond in a particular year, as well as a limited set of financial data that is actually being gathered since revenue flowing into the general fund is all that is being collected.

A few qualifications about the CAFR data are in order. The CAFR data employed for this paper are audited, and we are able to follow the same set of cities over time. The trade-off to this accuracy, level of detail, consistency, and comprehensiveness offered by CAFR data is limited availability. Even if all CAFRs produced in the nation were obtained, they would still represent a biased sample in which larger governments would be dispro- portionately represented. Local governments are not mandated in any state to create CAFRs, and many smaller governments do not because of the expense of hiring qualified independent CPAs, along with the expense for implementing accrual accounting. There- fore, it would not be correct to consider CAFR data as representing a random selection of local governments, but a sample that consists mostly of large population governments where they are more likely to be able to cover the fixed costs of CAFR reporting.

Additionally, although most aspects of CAFRs are standardized, there remain several points of discretion among the local officials in describing revenue sources that require sig- nificant attention for researchers collecting data to ensure it is comparable longitudinally and cross-sectionally. For instance, every CAFR will include a Statement of Revenues in the financial section of the report that includes detail on the revenue sources by fund, and this statement will list at least the general fund as a total fund aggregate. However, the Statement of Revenues will vary over time and city in terms of which funds are list as “major funds,” with an accompanying section devoted to providing the details of the nonmajor funds. Furthermore, a city might list revenue sources at a high level of detail (e.g., property taxes, income taxes, sales taxes, etc.) or a low level such as “total taxes” that will require deeper investigation into other supplementary components of the re- ports. A similar level of detail is lost to aggregation with respect to intergovernmental funds, as it is often not possible to discern from which government(s) the locality has

6NLC website: www.nlc.org. 7The largest cities were defined as those with populations greater than 50,000 until 2011, before

redefining to the largest 100 cities by population.

C© 2014 Wiley Periodicals, Inc.

 

 

118 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

received transfers. This is particularly true with respect to American Recovery and Rein- vestment Act (ARRA) funds directed to state and local “fiscal relief.”8 Operationally, the federal government directed funds to states which then “passed through” these funds to local units, at which point local governments’ CAFRs report them as “transfers from the state” as they would appear any other general support they might receive from the state. Likewise, state CAFR data report transfers to local units without specific declarations of which units received the funds, although many states produce reports separate from the audited CAFR data which detail how these funds were disbursed.9

In addition to these limitations in data aggregation, there are other important local government phenomena that the CAFR data cannot reliably ascertain. GASB 34 requires net pension liabilities to be disclosed in the Statement of Net Assets. Prior to the im- plementation of GASB 34 the impact of unfunded pension liabilities were hidden in the fiduciary fund and not disclosed in an overall financial statement, nor were unfunded pension liabilities ever reported in the general fund since it has never accounted for unfunded, long-term accrued liabilities. While pension fund accounting specifically, and unfunded liability reporting generally, has improved since the implementation of GASB 34, the measurement of net pension liabilities is still nonuniform, subject to substantial discretion, and reporting is not transparent to researchers or the general public.10

Even with these qualifications and limitations, the CAFR data represent the best available and most comprehensive opportunity to investigate fiscal sustainability, espe- cially when the subject of interest is large city governments.

3. LOCAL PUBLIC FINANCE DURING THE GREAT RECESSION

Overview

The housing crash associated with the Great Recession has drawn academics and think-tanks to predominantly focus on the fiscal well-being of local governments and cities via the property tax channel, a natural choice given the heavy reliance of local governments on that particular revenue instrument. In a policy report from The Rock- efeller Institute, it is stated that “significant declines in housing prices caused by the Great Recession had a noticeable impact on local property tax revenues” (Dadayan, 2012, p. 6). Using Census data, they go on to document nearly continuous quarterly declines in aggregate state and local property tax revenues since end of 2007 and leading into 2012. A similar narrative is drawn from research briefs on the NLC surveys, which document declines in local government property tax revenues entering the general fund in each fiscal year from 2010 to 2012 (Pagano et al., 2012). Furthermore, the NLC data indicate

8Section 3 of the American Recovery and Reinvestment Act. 9For example of such a report, see New York City’s online “Stimulus Tracker” (http://www

.nyc.gov/html/ops/nycstim) or Milwaukee’s special ARRA webpage (http://city.milwaukee.gov/arra). 10On June 25, 2012, GASB issued Statements 67 and 68 on June 25, 2012 requiring state and lo-

cal governments to implement new public pension accounting and reporting standards effective for fiscal years beginning after June 15, 2013. The new accounting and reporting standards require governments to report their net pension liability (NPL) in a Statement of Fiduciary Net Position, and prescribes other disclosure and technical requirements to improve the accuracy of pension asset and liability estimates, such as guidelines for determining and reporting the appropriate discount rate. Therefore, it will be signif- icantly harder for governments to hide substantial underfunding of their accrued pension liabilities after Statements 67 and 68 are fully implemented (GASB. June 2012. “Statement No. 67 of the Governmental Accounting Standards Board. Financial Reporting for Pension Plans (an amendment of GASB Statement No. 25)” Governmental Accounting Standards Board, Norwalk, CT).

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 119

that total receipts into the general fund first declined in 2007 and continued to decline in each year through 2012, with 2010 being the hardest year at –8.4 percent.

In the academic literature, the connection between local fiscal performance and the severity of the Great Recession and the housing crisis in particular has been more difficult to identify. Lutz et al. (2011) found a small-to-modest connection between housing prices and local property tax revenue from 2006 to 2009, and concluded that the general economic recession was a more significant source of stress than the housing market itself. Doerner and Ihlanfeldt (2011) found a weak correlation between housing prices and property taxes in Florida cities between 1994 and 2008, with this work drawing upon data from the Florida Department of Revenue. Particularly relevant to this study is that they found that the largest cities, those above the 75th percentile in the sample’s city population distribution, found no house price relationship to overall per capita city revenue even after testing for an asymmetric response that would allow for the revenue elasticity to vary differently between positive and negative housing price shocks.

Alm and Sjoquist (2011) examine a panel dataset of 396 property taxing districts (counties, cities, townships, and independent school districts) observed every quarter in the Census survey from 1998 through 2009, and find that a substantial number had avoided the significant declines in property tax revenues. This would be consistent with the NLC survey data that found property tax revenue to continue to increase through 2009, before tumbling in 2010 and 2011. Alm and Sjoquist (2011) note that, of the 396 observations, 377 were component units of Metropolitan Statistical Areas to suggest that the sample representation was weighted toward population. More importantly for the purposes of this paper, the authors also note that a close examination of the data “sug- gested that there is reason to suspect reporting or recording errors” in the Census reports, which they go on to catalogue the magnitudes of the different errors they discovered as part of their caveats to the reported findings.

Chernick et al. (2011) employ a different approach in that they use Census data to generate “constructed cities,” which aggregates all local public sectors with the idea that it reflects the local government condition as it is experienced by the citizens living within that geographic unit, and is arguably the type of concept the ACSG data are better suited for addressing.11 They use a panel of these “constructed cities” imputed over the largest (by population) city areas from 1997 to 2008 to estimate the relationship between the economic base and constructed city expenditures, which then allows them to produce an out-of-sample revenue-expenditure forecast for these areas to estimate the Great Recession’s impact during the period 2009 to 2013. The forecast results indicate that the Great Recession would reduce constructed city spending per capita experienced by the average citizen by about 7 percent.

In summary, the indicators all point to the view that the average local government has fared rather poorly during the Great Recession. Our attention now turns to the largest cities only, and uses the audited CAFR data that allow us to follow them over time for many relevant variables.

Big Cities in the Great Recession

The cities chosen for the study were the 35 largest on the basis of the 2005 popu- lation residing within the city boundaries, as identified by the U.S. Census Book of the

11The data created by Chernick et al. (2011) have been updated and made available through the Lincoln Institute of Land Policy as a database of “Fiscally Standardized Cities.”

C© 2014 Wiley Periodicals, Inc.

 

 

120 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

TABLE 1: Annual Percent Housing Price Growth Rates

Year Mean Std. Dev. Min Max

2005 7.88 7.13 1.05 23.65 2006 5.46 6.13 −2.74 19.86 2007 1.24 3.82 −6.95 9.04 2008 −3.44 5.45 −24.91 4.2 2009 −3.44 3.01 −24.74 2.1 2010 −2.66 3.03 −11.77 0.65 2011 −2.31 3.06 −13.3 1.17

Source: Authors’ calculation from FHFA Housing Price Index for the MSA’s of the 35 largest American cities.

States.12 The data bear out the significant economic impacts of the Great Recession on these cities. In 2005 to 2006 the Federal Housing Finance Agency’s Housing Price Index indicated average annual growth rates of 7.9 and 5.5 percent, respectively.13 As demon- strated in Table 1, however, in the 2008 to 2011 period the average annual growth rates were negative in every year and some areas experienced double digit rates of decline. A similar, albeit less drastic in percentage terms, story emerges when examining annual employment figures. The county hosts of these cities all experienced downturns in total annual employment around 2008 to 2009, with only three of the 35 areas returning to their 2008 levels by 2011. On average, in 2011 the employment level was still just 92.2 percent of the level in 2008. While the most recent data on gross domestic product by metropolitan area only date to 2009, it is clear from the preliminary estimates that these cities took a substantive economic hit.

Turning attention to some direct public finance figures, Figures 1 and 2 display, by city, the annual trends for alternative measures of public finance revenue by indexing them to their respective 2008 levels. Figure 1 represents total governmental revenue, whereas Figure 2 is restricted to general fund revenue only.14 Most of the main functions of government are financed out of the general fund, whereas the other funds might be set up for a special purpose or capital project, and as a result it is the general fund that receives most of the attention. When a public service requires a distribution of revenues, it must be drawn from its appropriate fund, but since money is fungible it is possible for money to be passed between funds indirectly in any accounting framework.

The largest recession since the Great Depression has produced visible downward effects in economic activity in city areas, so perhaps the most surprising observation that arises from Figures 1 and 2 is the level of general stability in revenues, especially given what has been inferred from more nationally representative surveys of governments. To provide as transparent display of the data as possible, these figures present the data in its nominal terms and are not scaled by other variables such as population or personal

12The cities are: Albuquerque (NM), Atlanta (GA), Austin (TX), Baltimore (MD), Boston (MA), Char- lotte (NC), Chicago (IL), Columbus (OH), District of Columbia, Dallas (TX), Denver (CO), Detroit (MI), El Paso (TX), Ft. Worth (TX), Honolulu (HI), Houston (TX), Indianapolis (IN), Jacksonville (FL), Las Vegas (NV), Los Angeles (CA), Louisville (KY), Memphis (TN), Milwaukee (WI), Nashville (TN), New York (NY), Oklahoma City (OK), Philadelphia (PA), Phoenix (AZ), Portland (OR), San Antonio (TX), San Diego (CA), San Francisco (SF), San Jose (CA), Seattle (WA), and Tucson (AZ).

13Most economic data are not reported annually at the city level, and as a result the closest approxi- mate unit is used in this paper. CAFR data are all the city governmental unit level data.

14Figure 2 continues to look very stable across city governments even when Phoenix is eliminated to allow a more sensitive scaling of the y-axes.

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 121

.4 .6

.8 1

1. 2

.4 .6

.8 1

1. 2

.4 .6

.8 1

1. 2

.4 .6

.8 1

1. 2

.4 .6

.8 1

1. 2

.4 .6

.8 1

1. 2

2004 2006 2008 2010 2012

2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012

Albuqureque Atlanta Austin Baltimore Boston Charlotte

Chicago Columbus DC Dallas Denver Detroit

El Paso Ft Worth Honolulu Houston Indianapolis Jacksonville

Las Vegas Los Angeles Louisville Memphis Milwaukee Nashville

New York OklahomaCity Philadelphia Phoenix Portland San Antonio

San Diego San Francisco San José Seattle Tucson

T ot

al R

ev en

ue (

In de

x: 2

00 8=

1. 0)

Year

Source: Authors’ compilation from city CAFR reports.

FIGURE 1: Total Governmental Funds by City, 2005–2011.

income.15 The nominal values are particularly illustrative because it is the more volatile form of the data as a result of inflation during the pre-recession period and deflation in the immediate aftermath.16 Despite the severity of the Great Recession on the economic data, large city budgets generally lack similarly dramatic fluctuations and appear to be on the road to recovery in revenue terms even if they have not returned to previous peaks. If one had fallen into a coma in early 2007, and upon awakening in 2012 been immediately presented with Figures 1 and 2, they would probably not infer that a recession of such considerable magnitude had just taken place. Figures 1 and 2 also nicely complement each other in terms of gaining a complete snapshot of a city. For instance, the General Fund for Phoenix in 2011 on Figure 2 takes a substantial jump, but as one can quickly infer from looking at Figure 1, this is simply the result of the city subsuming a separate fund into the general fund.17

CAFR data are not particularly well-suited to understanding the composition of spending by government functions, as it instead emphasizes spending categories for cur- rent functions, capital functions, and debt service. Current spending is to pay for day-to- day basic government functions and a variety of public services. Debt service spending is

15Estimates of population or personal income introduce a statistical fluctuation, but more importantly since the focus of this paper is on fiscal sustainability of governments that produce a considerable amount of nonrivalrous public goods, changes in levels are the most appropriate unit of analysis on theoretical grounds for the research topic. A paper focused on citizen demand for government services would more appropriately focus on changes in per capita levels.

16The consumer price index for urban areas experienced a –0.4 percent decline in 2009. 17Indeed, Phoenix eliminated a special fund for excise tax revenue collection and began reporting the

tax as a collection into their general fund.

C© 2014 Wiley Periodicals, Inc.

 

 

122 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

0 1

2 3

0 1

2 3

0 1

2 3

0 1

2 3

0 1

2 3

0 1

2 3

2004 2006 2008 2010 2012

2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012

Albuqureque Atlanta Austin Baltimore Boston Charlotte

Chicago Columbus DC Dallas Denver Detroit

El Paso Ft Worth Honolulu Houston Indianapolis Jacksonville

Las Vegas Los Angeles Louisville Memphis Milwaukee Nashville

New York OklahomaCity Philadelphia Phoenix Portland San Antonio

San Diego San Francisco San José Seattle TucsonT ot

al G

en er

al F

un d

R ev

en ue

( In

de x:

2 00

8= 1.

0)

Year

Source: Author’s compilation from city CAFR data. The spike in Phoenix 2011 is caused by the combining of the city’s excise tax fund with the general fund.

FIGURE 2: Total General Fund Revenue by City, 2005–2011.

TABLE 2: Annual Percent Net Change in City General Fund Balance

Year Mean Std. Dev. Min Max

2005 2.36 3.6 −6.44 12.43 2006 3.47 4.92 −2.90 24.17 2007 1.40 5.33 −17.99 20.01 2008 −1.94 4.43 −13.7 5.27 2009 −0.95 4.51 −9.54 13.81 2010 −0.34 4.65 −7.75 15.01 2011 2.9 6.12 −4.66 33.33

Source: Authors’ calculation from city CAFR data, expressed as a percent of total general fund revenue. Atlanta in 2006 is excluded due to a fiscal calendar change that shortened the fiscal year to only six months.

responsible for retiring the outstanding debt accumulated from the past as well its interest cost. Capital outlay mainly serve two functions: (1) paying for the operation, maintenance, and repairs of the infrastructures, which is an essential part of the operating budget, and (2) payment from the current revenue for all the newly initiated capital projects, and these resources usually go to capital project funds and ultimately become part of the capital budget. As a share of total governmental expenditures, on average these cities slightly shifted from spending on current operations (78.1 percent in 2005 to 75.1 percent in 2011) toward capital spending (12.4 percent in 2005 to 14.0 percent in 2011), but it is difficult to confirm if there were more substantive shifts within these broad categories.

Despite the stability of overall governmental revenues, it is still the case that the recession likely adds to some additional public service costs. Table 2 displays summary

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 123

TABLE 3: Mean Annual Percent Change in Selected City Cash Flows, 2006–2011

Cash Flow 2006 2007 2008 2009 2010 2011

1. Total Governmental Revenue

7.55 7.87 3.8 0.02 1.29 2.81

2. Property Tax Levy 6.6 8.36 7.35 4.58 2.58 −0.21 3. Total General Fund

Revenue 6.05 8.7 3.13 −1.32 0.48 5.76

4. General Fund Revenue from Property Taxes

6.28 10.43 4.31 6.31 4.89 0.34

5. Intergovernmental Grants into General Fund

14.24 −39.55 60.11 −13.34 −7.56 −30.79

6. Other General Fund Revenue

6.11 9.59 0.51 −3.63 −1.53 7.76

7. Total Net Assets 3.43 2.88 −1.33 10.2 −76.8 −103.66 8. Unrestricted Net

Assets 185.4 68.66 −248.13 −281.31 −9.1 122.97

9. Capital Grants and Contributions

15.43 3.06 −2.53 7.02 8.82 −3.67

All variables are for total primary government.

statistics by year for the annual net change in city general fund balance, scaled as a per- centage of general fund revenue. This is perhaps the most commonly employed indicator for whether or not a city is experiencing fiscal stress in terms of its ability to finance the expenditure needs made from its general fund. Recessionary periods to bring greater demands for particular city services, particularly programs aimed at poverty or employ- ment assistance. Although 2008 saw total general fund and governmental revenue grow at 3.13 and 3.80 percent, respectively, on average there was a slight deficit in the general fund balance of –1.94 percent. Of course, the standard deviations, minimums, and maxi- mums reveal that there was considerable variation among the cities. But a different city served as the minimum in each of the years, so it was generally not the case that cities were annually and continuously running these substantive deficits in their general fund. Rather, it seemed that the slow-downs in revenue growth and accelerated expenditure demands tended to hit the cities at different points in time. By 2011, only six cities ran any deficit at all, and only half of those cities ran a deficit greater than 1 percent, which is pretty similar to the pre-Great Recession years of 2005 to 2007. So while it is clear that the Great Recession provided a serious fiscal shock, it seems that large city governments are demonstrating some impressive resiliency.

We now turn to examine some of the underlying components of the cash flows being received by these governments, both to address how they were meeting these expenditure demands and how they maintained revenue stability. Table 3 provides the mean annual percentage change in selected city cash flows from 2006 to 2011. Within Table 3, lines 1 and 3 correspond to the means of the observed city-by-city revenue figures displayed in Figures 1 and 2. Table 3 assists with identifying where the Great Recession’s impact can be most observable. Sources of data and a more detailed description of the variables can be found in Table 4. Importantly, it would seem that general fund revenue from intergovernmental grants (line 5) started choking off in 2007 with a 39.6 percent drop, with some relief in 2008 as it rebounded by 60 percent before declining each year since. The significance of these effects can also be observed in the reconciliation of budget

C© 2014 Wiley Periodicals, Inc.

 

 

124 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

0 5

10 15

20 25

% (

B ud

ge te

d −

A ct

ua l)/

A ct

ua l

2004 2006 2008 2010 2012 Year

GF Total Revenue Budget Error GF Intergovernmental Budget Error

Source: Author’s compilation from city CAFR data. Note: A positive percentage error indicates that city budgeted for more revenues from state and federal government than was actually received during the fiscal year.

FIGURE 3: Average City Error in Expected General Fund (GF) Revenues by Year, 2005–2011.

variance statements found within CAFRs. CAFRs require that cities report what amount was budgeted to be received into the general fund and compare it to what was actually received.18 Figure 3 plots the mean budget percentage error for both total general fund revenue and intergovernmental revenue. In 2005 and 2006, these cities followed the fairly conservative practice of budgeting with the expectation of less revenue than it would actually receive, as indicated by the negative errors. In 2008 and through 2011, however, governments were actually receiving less than expected. Figure 3 demonstrates that intergovernmental transfers that were expected but not arriving seem to have played a substantial role in this, as even in 2008 when cities received an increase of 60 percent, it remained less than what was expected.19 As late as 2009, the cities had budgeted for revenues that were about 25 percent higher than what they would actually receive that year. This may be partially due to states and localities expecting to receive ARRA funds before they were disbursed by the federal government. The initial federal outlay of ARRA funds was $55.9 billion in fiscal 2009. In 2010 ARRA federal outlays peaked at $111.9 billion and 2011 outlays were $68.8 billion.20

ARRA created a State Fiscal Stabilization Fund (SFSF) to help state and local gov- ernments stabilize their budgets by minimizing the need to cuts budgets in essential services. By July of 2009, the federal government had allocated $31.09 billion in SFSF

18Unfortunately for our purposes, cities are not required to perform these budget variance statements for total governmental funds, only the general fund.

19Although we cannot systematically discern it from the CAFRs, part of the 2007 to 2008 fluctuations in the intergovernmental transfers seem to be driven by a slow-down in the timing of transfer payments from the state to the local governments. In short, the onset of the recession reduced the state’s cash flow, and caused them to delay some payments to local governments until the next fiscal year. This is likely what causes such a large drop in 2007 followed by such a large increase in 2008.

20Source: Government Accounting Office (hhtp://www.gao.gov/recovery, downloaded October 7, 2013).

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 125

TABLE 4: Variable Descriptions, Sources, and Statistics

Variable Description Mean (Std. Dev.)

Property Tax Levy Total governmental property tax levy issued. 0.049 (0.067)†

General Fund Property Tax Revenue

Revenue collected from property taxes for the general fund. Collected from Revenue Statements within CAFRs.

0.054 (0.128)†

Total Net Assets Assets minus liabilities for total primary government. Collected from Statement of Fund Net Assets of CAFRs.

−0.240 (3.858)†

Unrestricted Net Assets Unrestricted assets minus liabilities for total primary government. Collected from Statement of Fund Net Assets of CAFRs.

−0.290 (8.251)†

Total Employment Total nonfarm employment in the city’s county. Collected from Bureau of Labor Statistics.

−0.001 (0.027)†

Other Own Source General Fund Revenue

Total general fund revenue less property tax revenue and intergovernmental fund transfers. This is mostly consisting of other tax revenues, revenue from licenses, fees, and permits, as well as income earned on investments. Computed from Statement of Revenues within city CAFRs.

0.031 (0.154)†

Aggregate Property Tax Dummy variables used to indicate 0.286 (0.453) Limit whether or not there are limitations

Individual Property Tax Limit

placed on measures of the aggregate property tax levies/growth in the government and/or what could be levied from an individual property owner. These limits are coded based on data source provided in Seljan (2013), with the exception that Indiana adopted an Individual property tax limitation in 2010 that was not reflected in that table.

0.498 (0.501)

Capital Grants Capital grants and contributions for total primary government. Collected from Statement of Activities within city CAFRs.

0.046 (0.546)†

Operating Grants Operating grants and contributions for total primary government. Collected from Statement of Activities within city CAFRs.

0.040 (0.317)†

Intergovernmental General Fund Transfers

State and federal transfer revenue into the general fund. Collected from Statement of Revenues within city CAFRs.

−0.03 (2.650)†

Notes: †Indicates the mean and standard deviation of the variable is calculated from the year-over-year change in the log transformation (d.ln()) for comparability with regression pointe estimates. Multiply by 100 to convert to percentage change. For CAFR revenue data, due to a change in the closing date of the fiscal year Atlanta’s 2006 values reflected only six months, so this observation was smoothed by attributing half of its previous year revenues and expenditures.

funds to states. The vast majority of SFSF funds, 81 percent, had to be used to “alleviate shortfalls in state support for education to school districts and public institutions of higher education,” the remaining 18 percent, had to be used “for the government services fund for public safety and other government services, which may include education.”21 By the

21GAO 09-829 July 2009 RECOVERY ACT: States’ and Localities’ Current and Planned Uses of Funds While Facing Fiscal Stresses.

C© 2014 Wiley Periodicals, Inc.

 

 

126 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

end of August 2013, the Department of the Treasury had paid out over $269 billion in stimulus funds for use in states and localities (GAO, 2013).

Table 3 also reveals that general fund revenue from sources other than property taxes and intergovernmental funds (line 6) also took a fairly substantive hit in terms of slowing growth and negative growth.22 The table information suggests that the stressors seem to have been offset by contributions from the property tax and drawdowns on net assets. The total governmental property tax levy (line 2) on average increased each year until 2011, and the amount of the property taxes being funneled into the general fund never declined on average. Even to the extent these percentage increases appear small, the large amount of property taxes being levied means that small percentage gains can offset larger percentage losses from other sources. Negative changes in net assets (i.e., assets minus liabilities), measured either by the total or the subset of unrestricted assets, indicate that cash was being injected either with the liquidation of assets or by the issuing of debt.

The role of net assets and property taxes in this process of determining the level of public funds is important. In principle, local policy makers determine the total expendi- tures to allocate across the different governmental functions. To meet these expenditures, local fiscal officers must draw from some combination of revenues, accumulated assets, and issuing liabilities. Restated, expenditures must equal revenues plus the change in net assets. The total amount of revenue is composed of property taxes and all other sources of revenue, which includes other own source revenues and intergovernmental transfers. The role of the property tax is a significant one, however, because the property tax levy is used to account for any difference between the desired own source revenue and total revenues. City fiscal structure is remarkably heterogeneous in terms of revenue instruments, but the property tax differs from all other taxes. All other tax instruments used by cities re- quire the setting of a rate, with revenues then being based upon the volume of subsequent exchanges in the tax base. By contrast, since property is a stock of infrequently traded as- sets, rather than a flow of transactions such as in the case with labor income or household consumption, the government is responsible for both the determination of taxable value and setting the rate. Within important statutory guidelines that limit the use of property tax levies, the collection of property tax revenue is really a political choice rather than the outcome of base fluctuations. As such, the only formal role the assessment process plays is to determine the distribution of the property tax burden across parcel owners, and the market value of property is no more a determinant of property tax revenues than is household income (Netzer, 1964; Fisher, 2007; Mikesell, 2011).

Since the property tax levy and net assets are choice variables set to meet the planned expenditures, the choice over whether to raise or decrease total spending is really a choice reflecting what the government can alter in resources drawn from property taxes and net assets. Increases in revenues from other nonproperty sources result in increases in expenditures only if policy makers deliberately choose to not completely offset the revenue increase with reduced property tax levies or increases in net assets. With this understand- ing of the property tax levy and net assets as tools for compensating revenue shocks, we return to the data. Figure 4 displays the total property tax levy for the cities across all governmental funds. These figures reveal that the property tax levy has generally been far less stable relative to the total revenue in Figures 1 and 2. While the property tax occupies a substantial proportion of city revenues, it does not generally appear that large fluctuations in property tax levies have consequently produced large fluctuations in the

22General fund revenues from sources other than property and intergovernmental transfers typically include other excise taxes, fees, permits, investment income, and a host of other miscellaneous items.

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 127

.6 .8

1 1.

21 .4

.6 .8

1 1.

21 .4

.6 .8

1 1.

21 .4

.6 .8

1 1.

21 .4

.6 .8

1 1.

21 .4

.6 .8

1 1.

21 .4

2004 2006 2008 2010 2012

2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012

Albuquerque Atlanta Austin Baltimore Boston Charlotte

Chicago Columbus DC Dallas Denver Detroit

El Paso Ft Worth Honolulu Houston Indianapolis Jacksonville

Las Vegas Los Angeles Louisville Memphis Milwaukee Nashville

New York Oklahoma City Philadelphia Phoenix Portland San Antonio

San Diego San Francisco San José Seattle Tucson

P ro

pe rt

y T

ax L

ev y

(2 00

8= 1.

0)

Year

Source: City CAFR data. Note: This accounting of the property tax levy is missing for all years of Tucson and 2011 for Seattle.

FIGURE 4: Annual Total Property Tax Levy, Total Primary Government (2008 = 1.0).

overall revenues. This is consistent with city officials being able to recalibrate the property tax levies to negate the impact of fluctuating revenue sources. A similar interpretation can be drawn when examining only property tax contributions to the general fund for the main operations of government in Figure 5. The fact that the movements in property tax levies are not identical to the observed changes in property tax revenues in the general fund reflects a conscious decision to recalibrate spending decisions through the property tax mechanism. Likewise, 2006 and 2007 saw a building up of net assets (Table 3, line 7), particularly unrestricted net assets (Table 3, line 8), which suggests that these monies were being added to contingency or rainy day funds.

Figure 6 depicts the beginning of year balance and direction of change in unrestricted net assets from 2006 to 2011. The intention of this figure is to provide a general sense of the sustainability of these cities’ fiscal decisions. A city with a positive balance which is increased by year’s end (white solid-fill) is likely to be on a fiscally sustainable path, whereas a city with a negative balance that is becoming more negative over the course of the year (black solid-fill) is an unsustainable path. Somewhere in-between these two extremes is a city that is withdrawing from its positive balance (white with black dots) and another that is rebuilding from its negative balance (white with black diagonal lines). The proportion of cities with a negative balance grew from 2006 to 2009, as did the proportion that was negative and decreasing in their unrestricted net assets. By 2011, there was just one more city in 2011 than in 2006 that was running both a negative balance and drawing down from those net assets. Importantly, these were not the same cities in 2006 as were in 2011 with the exception of New York City, which was negative and decreasing in every year. Chicago, Detroit, Houston, Indianapolis, Jacksonville, Philadelphia, and Portland

C© 2014 Wiley Periodicals, Inc.

 

 

128 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

0 2

0 2

0 2

0 2

0 2

0 2

2004 2006 2008 2010 2012

2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012 2004 2006 2008 2010 2012

Albuquerque Atlanta Austin Baltimore Boston Charlotte

Chicago Columbus DC Dallas Denver Detroit

El Paso Ft Worth Honolulu Houston Indianapolis Jacksonville

Las Vegas Los Angeles Louisville Memphis Milwaukee Nashville

New York Oklahoma City Philadelphia Phoenix Portland San Antonio

San Diego San Francisco San José Seattle Tucson

G en

er al

F un

d R

ev en

ue s

fr om

P ro

pe rt

y T

ax es

( In

de x:

2 00

8= 1.

0)

Year

Source: City CAFR data. Note: This accounting of the property tax levy is missing for Chicago, Oklahoma City, Indianapolis, and Tucson

(2005 only).

FIGURE 5: Annual City Total General Fund Revenues from Property Tax Collections.

Source: City CAFR data. Note: 2006 has only 33 values due to missing data for Phoenix and Las Vegas.

FIGURE 6: Balance and Growth of Unrestricted Net Assets in 35 Largest Cities by Year, 2006–2011.

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 129

carried negative balances for the entire time period but had at least some periods in which they increased their annual holdings of unrestricted net assets. On the other side of the spectrum, San Diego and Honolulu carried positive and increasing balances the entire time period, and another 16 cities carried positive balances the entire span but had periodic draw downs on those balances.

Although the nature of the data differs for reasons previously described, the ASCG and NLC surveys indicate that the typical local government is reducing property taxes right along with their total receipts. If the data were directly comparable to the CAFR data of our large cities, the analysis of this paper would suggest that large cities are faring the Great Recession with the property tax in a manner that is much more successful than other, smaller local governments. Why the larger cities would be better able to manipulate their property tax levies than smaller ones is an open and interesting question.23 Given the data differences that observation can only be offered with qualifications, however, it can at least be said that big cities are passing through the Great Recession with a substantive amount of revenue stability that is driven by the manipulation of the property tax. Perhaps more surprising is the extensive use of net assets, as the acquisition of debt is considered to be more of a characteristic of a national government, as opposed to a local government. The academic literature on local governments has generally treated spending and revenues as a near identity, but this look at large city government suggests that view to be incomplete.

The final item presented on Table 3 is capital grants and contributions to total pri- mary government, which represents all capital grants received by the city within the fiscal year. This measure captures intergovernmental grants into general fund (line 5), and other revenues flowing into other funds as well. It is worthy of attention because there was at least a popular narrative around stimulus payments being made to local governments as part of the American Recovery and Reinvestment Act of 2009 (“stimu- lus” funding, henceforth). While we cannot determine how much or if any capital grants were stimulus money, this is a significant measure for those funds to appear because it includes all governmental and business-type capital grants.24 Interestingly, even as to- tal intergovernmental grants to the general fund were declining, this measure of capital grants was increasing for total primary government in fiscal years 2009 and 2010, which would fit with the timing of the stimulus delivery, although the amount of growth was not particularly exceptional.25

It is with the findings of this section that we now turn to an empirical model to separate out how these different revenue sources are passing through to net assets and property taxes.

23The local public’s tolerance for taxes or debt, for instance, may differ in large cities from the rest of country. Another possibility is that many state laws differ in their application to local governments in a way that is directly related to population size. This is usually a way to specifically impose or exempt a particular city from a statute without directly naming it.

24Business-type activities include functions like public utilities, libraries, golf courses, etc. 25By design, a significant amount of stimulus funds that were directly sent to cities were not intended

for capital spending. In New York City, for example, stimulus funds were used across the city’s budget: Infrastructure ($951 million); Energy Efficiency ($123 million); Economic and Workforce Development ($67 million); Health and Support ($2.4 billion); Education ($2.6 billion); Public Safety ($80 million); Neighborhood Stabilization ($227 million); Medicaid Relief ($2.6 billion). New York City reports receiving $9.1 billion in stimulus funding, and that $241.17 million (roughly 3 percent) of that funding “displaced” funding from New York City, mostly on infrastructure projects planned by the city prior to the ARRA.

C© 2014 Wiley Periodicals, Inc.

 

 

130 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

4. CITY GOVERNMENT FISCAL MANAGEMENT IN THE GREAT RECESSION

Regression-Based Empirical Investigation

The previous sections established the significance of the property tax in helping America’s largest cities weather the Great Recession, despite the fact that the recession has been particularly damaging for property markets. Furthermore, it became clear that intergovernmental transfers and capital grants were important fluctuations that affected local governments. Finally, unlike most local governments, understanding net asset with- drawal is required for monitoring the fiscal performance of large city governments.

The implied theoretical model adopted for this paper is one in which a budget setter is updating the city budget from the previous year, making accommodations to reflect changes in economic conditions and financial resources. In response to a negative fiscal shock, for example, the local official can maintain current spending by increasing the property tax levy, selling assets for cash, or issuing bonds. This can be stated as the government official responding to the current circumstances by making a simultaneous choice over how to revise the property tax levy or reallocate their net assets (assets minus liabilities), which we state as a system of seemingly unrelated regressions as

. Lit = Xit�1 + εit,(1) .

NAit = Xit�2 + �it,(2) where

. Lit and

. NAit represent property tax levy and net asset growth, respectively, in city

i in year t. Since the decision to be made on total spending requires the utilization of these two factors, it is likely that the error terms, εit and �it, are contemporaneously correlated. Any factor that implicates one of these dependent variables should, in principle, have the potential to affect the other. For the moment, all other potential variables are represented in the 1 × k vector in X, although the corresponding coefficients may differ. The use of the CAFR data allows for the introduction of equation (2) to allow for a simultaneous modeling of the choice between net asset repositioning and property tax levy growth that allows for this paper to depart from much of the previous literature. Most econometric work on the change in government spending employs total spending as the dependent variable (e.g., Bergstrom and Goodman, 1973; McMillan et al., 1981; Edwards, 1986; Holcombe and Sobel, 1995), or only the property tax levy equation (e.g., Bloom and Ladd, 1982; Ladd, 1991; Ross and Yan, 2013). Since the last dollar of spending is the result of a choice between these two general instruments, the marginal effects of a covariate in a regression can be interpreted as a reduced form of the utilization of these two equations. Using only equation (1) assumes that there is no contemporaneous cross-equation error, and so there are potential efficiency gains by estimating simultaneously.

In estimating the equations (1) and (2), we have different and informative measures of the dependent variables that can be operationalized. For equation (1), we first employ the governmental property tax levy, which includes the property tax levies for all services and funds of the primary government. Since most of the popular attention falls on the general fund revenue sources, such as in the NLC, general fund revenue from property taxes will also be implemented and estimated.

For equation (2), the dependent variable will be unrestricted net assets (UNA). Un- restricted net assets represent the city’s net holdings of accumulated assets that are free of liabilities, and can be used for any purpose at the city’s discretion. While all unre- stricted net assets may not be available and read cash for expenditures, they can be used to manipulate budget balances, often by underfunding accrued long-term liabilities like pensions, for example. Also, these funds may be liquefied to cover cash flow shortages in the governmental budget (Johnson et al., 2012). Unrestricted net assets are in contrast

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 131

to restricted and capital net assets, which represent varying degrees of stringency over how the funds can be employed. The second operationalization of net assets then will be total net assets. In both instances, if the city issues a liability that has no accompanying asset to off-set it, then net assets decline. Both measures of net assets also decline if the city reduces its holdings of assets, perhaps by selling them, without also reducing a corresponding liability. Therefore, reductions in net assets may represent injections of ad- ditional resources, with the total and unrestricted measures representing different levels of flexibility in how the funds can be employed.

Turning attention now to other factors be introduced to the system regression, we include a basic control to capture the area specific economic conditions (E). For the pre- sented results, the economic conditions variable is the year-over-year change in total nonfarm employment in the county and lagged by one year so as to better contempora- neously align the budgeting period with the economic conditions.26 Also included are tax revenues that are not property taxes (R), and alternative measures of intergovernmental transfers (G). From the CAFRs, we adopt total primary government capital grants, total primary government operating grants, and intergovernmental transfers to the general fund. Intergovernmental transfers to the general fund are the more popularly identified measure of transfers from higher levels of government, and is usually identified in bud- get reconciliation statements. Operating grants will oftentimes require some form of local matching from the city government, whereas capital grants are less likely so. The purpose and target of these transfers can be different, however, as a state or federal transfer to a fund outside of the general fund may allow for the movement of other unrestricted funds to other components of the budget. Intergovernmental transfers into the general fund are likely to be less restrictive cash flows, but the CAFRs typically do not make it clear whether these are intended for operating or capital purposes, and they are likely included in the other measures of operating and capital grants. As a result, we employ all three as alternative measures of intergovernmental transfers.

Some state governments, however, impose limitations on property taxes that limit the ability to make these revisions to the levy. The details of these property tax limits vary considerably, but following Seljan (2013) they are categorized here as either limits on the aggregate property tax levy (APTL), or on the property tax burden that may be levied against an individual (IPTL). Indiana, for example, limits the property tax burden an individual parcel owner may face as a percentage of the fair market value. The percentage varies with the classification of property, and any bill levied in excess of the percentage cap against an owner simply represents a revenue loss for the local government. Massachusetts has a limit of 2.5 percent in the potential aggregate property tax levy growth from one year to the next. The introduction of these aggregate limits is intended serve as constraints on the ability to raise the levies, and they are potentially more binding on the ability of a local government to raise revenue. As long as some individuals have not reached the limit in a state with an IPTL, increases in the levy could still produce revenue. Although the effect of property tax limits on property tax levies and spending has been studied extensively with mixed findings, this paper would seem to be the first to consider the effect of such limits on net assets, especially in a system of equations with the property tax levy. To the extent a property tax limit is binding, it may likewise cause city governments to respond to fiscal shocks by selling assets or bonds to

26The county measure of employment is adopted for two reasons. First, it is more reliably estimated by the Bureau of Labor Statistics than the city level is, both in terms of frequency and methodology. Second, it is usually the case that the home county of the city remains home to a significant share of the actors who spend and/or earn taxable income in the city’s local economy.

C© 2014 Wiley Periodicals, Inc.

 

 

132 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

TABLE 5: System Estimation of Seemingly Unrelated Regressions on Total Governmental Property Tax Levy Growth and Change in Total Net Assets (TNA)

d.ln(PTL) d.ln(TNA) d.ln(PTL) d.ln(TNA) d.ln(PTL) d.ln(TNA)

d.ln(Lag Total 0.023 −14.454 0.032 −14.086 0.002 −14.495 Employment) (0.237) (17.350) (0.238) (17.370) (0.239) (17.335)

d.ln(Other Own Source −0.056* 0.781 −0.053* 0.781 −0.052* 0.790 GF Revenue) (0.030) (2.198) (0.030) (2.193) (0.030) (2.195)

Aggregate Level −0.019* −1.579** −0.019* −1.569** −0.020* −1.582** Property Tax Limit (0.011) (0.783) (0.011) (0.782) (0.011) (0.781)

Individual Level 0.009 0.364 0.008 0.363 0.008 0.362 Property Tax Limit (0.009) (0.640) (0.009) (0.639) (0.009) (0.640)

d.ln(Capital Grants) −0.015* −0.027 (0.008) (0.561)

d.ln(Operating Grants) 0.024* 0.331 (0.014) (0.987)

d.ln(General Fund 0.001 0.002 Intergovernmental Grants)

(0.002) (0.114)

Intercept 0.013 −1.201 0.011 −1.221 0.013 −1.201 (0.011) (0.798) (0.011) (0.800) (0.011) (0.799)

R2 0.241 0.037 0.238 0.038 0.226 0.037 Notes: All specifications include year-fixed effects. The prefix d.ln() indicates the year-over-year change in

the natural log of the variable in parentheses. Sample is unbalanced panel of 188 observations across 34 cities during 2005–2011. Standard errors appear in parentheses with statistical significance indicated at the 1 percent (***), 5 percent (**), and 10 percent (*) confidence level.

generate cash, causing it to be negatively influence net assets in addition to the property tax levy.

At this point, the system of equations can be specified as . Lit = �1

. Eit + �1

. R+ �1

. Git + �1APTL + �1IPTL + �1t + ε̇it,(3)

. NAit = �2

. Eit + �2

. R+ �2

. Git + �2APTL + �2IPTL + �2t + �it,(4)

where � represents a vector of year fixed effects.27 The point estimate for employment must also be taken with some caution due to potential endogeneity concerns, for which no valid instrument was discovered.28 Certainly, willingness to support levies and deficits may be impacted causally by employment, but so too might employment be affected by the spending programs supported or complemented by these finances. Likewise, intergov- ernmental transfers may have been targeted toward cities with more limited ability to adjust levies or engage in deficit spending, which would bias both point estimates closer to zero than their true marginal effect.

Regression Results

Table 5 presents the system estimates of equations (3) and (4) using the total govern- mental property tax levy and the change in total net assets as the respective dependent variables. As previously discussed, the three specifications differ by the alternative mea- sures of intergovernmental funds, but all specifications include the use of year fixed effects.

27The use of city fixed effects would eliminate the variation in the property tax limit variables. 28Both the employment variable lagged additional periods and employment in the nearest same-state

MSA were attempted as instruments, but they failed the first stage F-test for instrument validity.

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 133

Given the relatively small sample size and the historically volatile period, unsurprisingly the specifications have low levels of statistical significance, but generally the magnitude is similar for the common variables across the specifications. The basic control for the local business cycle, total employment growth, is positively correlated with the property tax levy and is consistent with the expectation that cities thriving more economically will be more capable of supporting a property tax levy. The effect of employment on net assets was ambiguous in expected sign because additional growth could make it more acceptable to issue debt for maintaining services or it might encourage the savings of “rainy day” type assets. The results here suggest that it is the former effect that dominates, and a city with a mean level of employment growth (–0.10 percent) would see a positive increase in net assets of about 1.4 percent, although the effect is not statistically significant. For a sense of scale, a 1 percent increase in the sample’s mean property tax levy corresponds to about $8.6 million.

Other general fund revenue is inversely related to the property tax levy by a statis- tically significant margin in Table 5, a finding consistent with the expectations that the property tax serves as a compensating mechanism for overall revenue shortfalls. A stan- dard deviation increase in the year-over-year change in other general fund tax revenue, about 15 percent, is correlated with about a 0.84 percent decrease in the property tax levy.29 Other general fund revenues had no statistically significant impact on total net assets in any specification.

Although individual property tax limits had no statistically significant effect on prop- erty tax levies, aggregate property tax limits served to levy growth by a statistically sig- nificant margin of about 2 percent. The findings of these limits on total net assets were similar in direction and statistical significance. Restrictions on the ability to revise prop- erty tax levies seem to have a consequential impact on either the selling off of assets or the issuance of new liabilities. The point estimate suggests that the presence of the aggregate property tax limits reduced net assets growth by about 158 percent, which is a little less than one-half a standard deviation. If a goal of property tax limits were to limit both spending and government deficits, this evidence would suggest that these limits imply a trade-off rather than a complementary effect, as the inability to access property taxes induces some additional deficit spending.

Turning attention to the alternative definitions of grants used across specifications of Table 5, the annual change in capital grants was statistically significant at the 10 percent confidence level. The point estimate on capital grant growth indicates that a standard deviation increase would result in a 0.81 percent decline in the property tax levy, which is relatively small compared to the mean levy growth rate of 4.9 percent.30 Operating grants, by contrast, was associated with positive increases in the property tax levies that were also statistically significant at the 10 percent level, with a standard deviation increase implying a 0.768 percent increase.31 Since total net assets include capital and restricted assets, it is particularly surprising that increasing capital grants do not seem to translate into increases in total net assets. The insignificance of operating grants and intergovernmental grants might simply result in the termination of the public services they were being drawn for, so it is less surprising that they have no effect on total net assets. It should be expected that capital grants tend to result at least in some additional net assets. With this finding in mind, we turn to Table 6.

In Table 6, unrestricted net assets are employed as the dependent variable instead of the total net assets measure. The point estimates for the property tax levy specifications

29Calculation: 15×(–0.056) = –0.84 30Calculation: 54×(–0.015) = –0.81 31Calculation: 32×(0.024) = 0.768

C© 2014 Wiley Periodicals, Inc.

 

 

134 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

TABLE 6: System Estimation of Seemingly Unrelated Regressions on Property Tax Levy (PTL) and Unrestricted Net Asset (UNA) Growth

d.ln(PTL) d.ln(UNA) d.ln(PTL) d.ln(UNA) d.ln(PTL) d.ln(UNA)

d.ln(Lag Total 0.023 26.415 0.032 28.819 0.002 30.150 Employment) (0.237) (30.570) (0.238) (31.069) (0.239) (30.903)

d.ln(Other Own Source −0.056* 1.688 −0.053* 1.087 −0.052* 0.927 GF Revenue) (0.030) (3.873) (0.030) (3.922) (0.030) (3.912)

Aggregate Level −0.019* −1.201 −0.019* −0.984 −0.020* −0.930 Property Tax Limit (0.011) (1.380) (0.011) (1.398) (0.011) (1.393)

Individual Level 0.009 0.489 0.008 0.549 0.008 0.595 Property Tax Limit (0.009) (1.127) (0.009) (1.144) (0.009) (1.141)

d.ln(Capital Grants) −0.015* 2.373** (0.008) (0.988)

d.ln(Operating Grants) 0.024* −0.672 (0.014) (1.766)

d.ln(General Fund 0.001 −0.231 Intergovernmental Grants)

(0.002) (0.203)

Intercept 0.013 1.213 0.011 1.321 0.013 1.221 (0.011) (1.407) (0.011) (1.431) (0.011) (1.424)

R2 0.241 0.096 0.238 0.069 0.226 0.074 Notes: All specifications include year-fixed effects. The prefix d.ln() indicates the year-over-year change in

the natural log of the variable in parentheses. Sample is unbalanced panel of 188 observations across 34 cities during 2005–2011. Standard errors appear in parentheses with statistical significance indicated at the 1 percent (***), 5 percent (**), and 10 percent (*) confidence level.

will be unchanged from Table 5, but the standard errors differ slightly due to the error structure being simultaneously estimated with a different second equation, although observable differences that can appear in the table are limited due to rounding. The results of Table 6 indicate that a 10 percent increase in capital grants increased unrestricted net assets by 23 percent, a finding statistically significant at the 5 percent level. Operating grants and a general measure of intergovernmental transfers continue to have largely no effect. Capital grants also had a negative, statistically significant effect on property tax levies, just as in Table 5. These results suggest that capital grants from state and federal governments were being used to partly reduce property taxes and restore cash reserves. The absence of an effect on total net assets suggests that they were not being used solely to finance designated capital projects, although it could be that these unrestricted assets are ultimately being used for that purpose at some point in the future. Although it cannot be discerned from CAFRs, it would be an interesting forensic accounting exercise to trace these grants to their origins to determine whether or not these capital grants were related to the American Recovery and Reinvestment Act of 2009. Elsewhere in Table 6 the results are generally the same as in the previous tables.Tables 7 and 8 repeat these specifications, but instead we use property tax revenues accrued to the general fund, a measure that has been preferred by some researchers and the NLC. Unlike the total governmental funds property tax levy, this general funds specific measure demonstrates no statistically significant elasticity with respect to capital grants. This suggests that capital grants are being used to lower the aggregate property tax burden and add to the city’s unrestricted net assets, but they are not being directed into the cities most visible fund. The sign on own source general fund revenue indicates that it carries a direct positive correlation property tax revenue in the general fund. That the increase in other revenue is only met with a substitution in the total governmental property tax levy is indicative that this general fund money is being used to lower the property tax burden on other special projects rather

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 135

TABLE 7: System Estimation Results of Seemingly Unrelated Regressions on General Fund Revenue from Property Taxes (GPT) and Total Net Assets (TNA)

d.ln(GPT) d.ln(TNA) d.ln(GPT) d.ln(TNA) d.ln(GPT) d.ln(TNA)

d.ln(Lag Total −0.275 −18.261 −0.210 −17.951 −0.282 −18.398 Employment) (0.499) (15.635) (0.497) (15.668) (0.497) (15.623)

d.ln(Other Own Source 0.182*** 0.564 0.178*** 0.556 0.182*** 0.584 GF Revenue) (0.060) (1.891) (0.060) (1.890) (0.060) (1.889)

Aggregate Level −0.011 −1.732** −0.009 −1.726** −0.011 −1.742*** Property Tax Limit (0.022) (0.676) (0.021) (0.675) (0.021) (0.674)

Individual Level 0.002 0.129 0.003 0.127 0.002 0.126 Property Tax Limit (0.018) (0.574) (0.018) (0.574) (0.018) (0.574)

d.ln(Capital Grants) 0.001 −0.105 (0.016) (0.504)

d.ln(Operating Grants) 0.044 0.313 (0.028) (0.881)

d.ln(General Fund 0.004 −0.003 Intergovernmental Grants)

(0.003) (0.104)

Intercept 0.014 −1.071 0.012 −1.091 0.015 −1.073 (0.023) (0.720) (0.023) (0.722) (0.023) (0.721)

R2 0.094 0.064 0.107 0.065 0.101 0.064 Notes: All specifications include year-fixed effects. The prefix d.ln() indicates the year-over-year change in

the natural log of the variable in parentheses. Sample is unbalanced panel of 184 observations across 33 cities during 2005–2011. Standard errors appear in parentheses with statistical significance indicated at the 1 percent (***), 5 percent (**), and 10 percent (*) confidence level.

TABLE 8: System Estimation Results of Seemingly Unrelated Regressions on General Fund Revenue from Property Taxes (GPT) and Unrestricted Net Assets (UNA)

d.ln(GPT) d.ln(UNA) d.ln(GPT) d.ln(UNA) d.ln(GPT) d.ln(UNA)

d.ln(Lag Total −0.275 12.658 −0.210 14.773 −0.282 16.314 Employment) (0.499) (32.649) (0.497) (33.146) (0.497) (32.956)

d.ln(Other Own Source 0.182*** 1.940 0.178*** 1.551 0.182*** 1.477 GF Revenue) (0.060) (3.949) (0.060) (3.998) (0.060) (3.984)

Aggregate Level −0.011 −1.885 −0.009 −1.695 −0.011 −1.635 Property Tax Limit (0.022) (1.411) (0.021) (1.429) (0.021) (1.422)

Individual Level 0.002 0.265 0.003 0.338 0.002 0.363 Property Tax Limit (0.018) (1.200) (0.018) (1.214) (0.018) (1.212)

d.ln(Capital Grants) 0.001 2.330**

(0.016) (1.051) d.ln(Operating Grants) 0.044 −0.732

(0.028) (1.864) d.ln(General Fund 0.004 −0.229

Intergovernmental Grants)

(0.003) (0.220)

Intercept 0.014 1.218 0.012 1.293 0.015 1.193 (0.023) (1.504) (0.023) (1.527) (0.023) (1.520)

R2 0.094 0.08 0.107 0.057 0.101 0.061 Notes: All specifications include year-fixed effects. The prefix d.ln() indicates the year-over-year change in

the natural log of the variable in parentheses. Sample is unbalanced panel of 184 observations across 33 cities during 2005–2011. Standard errors appear in parentheses with statistical significance indicated at the 1 percent (***), 5 percent (**), and 10 percent (*) confidence level.

C© 2014 Wiley Periodicals, Inc.

 

 

136 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

than general activity. Otherwise, the findings in Tables 7 and 8 are largely similar to their counterpart specifications in Tables 5 and 6.

5. CONCLUDING REMARKS ON FINDINGS

The implementation of CAFRs under GASB standards has been embraced by practi- tioners who wish to increase financial transparency of governments. Whether or not the CAFRs are useful for the citizenry’s understanding of city government fiscal behavior, they do provide the opportunity for a much more in-depth investigation into the behav- iors of local governments. For this study, we target America’s largest 35 cities according to their 2005 population, and investigate their ability to cope with the Great Recession. This is a particularly relevant analysis because large cities are expected by many to play an increasingly significant role in the economy with a public finance system that is remark- ably unchanged since the country’s origins. The Great Recession was a global phenomenon that targeted the tax base of cities’ most important revenue source: property. This might be indicative of the ability of cities to manage future fiscal problems over the much longer run. Our investigation of city CAFRs reveals several interesting findings:

� Even though large cities received significant economic shocks (as measured by hous- ing price index, total nonfarm employment), there was little evidence of comparable volatility in total governmental or general fund revenue among any of the cities in the aftermath of the Great Recession. The recession has been a fiscal shock, but cities are demonstrating adaptability and fiscal resiliency for the time being.

� The property tax has played a very significant role in offsetting the declines from other revenue sources, particularly in the management of resources across funds.

� Declines in intergovernmental grants and transfers were commonplace for these large cities between 2007 and 2011, with fiscal year 2008 being the only year where there was a positive change from the previous year. In all periods, the cities in our sample had budgeted with the expectation of receiving even more intergovernmental transfers than they actually received.

� Even after levy increases, city governments still experienced deficits that they financed by reducing their net assets, a behavior more commonplace with federal governments.

� The balance and growth of unrestricted net assets suggests that cities are generally returning to the same pattern of fiscal sustainability by 2011 that they had in 2006.

� Capital grants from state and federal governments were largely translated into prop- erty tax levy reductions and new unrestricted net assets (assets that can be utilized for any purpose) instead of restricted capital assets.

� Not only did aggregate property tax limitations restrict access to property taxes, but it also resulted in the drawdown of net assets. This suggests that property tax limits may be successful in lowering the ability of governments to spend from their property tax base, but also induces the spending to come from a position of deficits. This suggests property tax limitations offer something of a trade-off between lower spending and greater deficits.

The dominant narrative that has emerged from surveys of American local govern- ments with broader coverage is that they have fared rather poorly through the Great Recession. Our review of the literature and other surveys reveals that the commonly accepted view is that total revenue declines have occurred and that negative growth in property tax revenue is the main culprit. As we demonstrate, the fiscal experience of large city governments does not appear to be particularly bleak given the severity of the economic shock. Although we think the CAFR data are much better suited for gaining a comprehensive view of local government finance, it could very well be the case that

C© 2014 Wiley Periodicals, Inc.

 

 

ROSS, YAN, AND JOHNSON: THE PUBLIC FINANCING OF AMERICA’S LARGEST CITIES 137

these larger governments are able to respond to negative fiscal shocks differently than their smaller counterparts. For example, states generally carve out various exemptions from other policies, or pass targeted legislation, for their largest cities. New York City, for example, has a different set of applied laws than all other local governments in the state of New York. Illinois is a Dillon Rule state, i.e., local governments only have powers specif- ically granted by the state, but the city of Chicago has been granted Home Rule powers that grant them much more autonomy. Indiana frequently passes laws whose application differs based on a population range that only has a consequence for Indianapolis. It is likely that the policy space for large cities is very different from other local governments, even within the same state, and that this results in very different fiscal responses during volatile times.

Another important qualification of this study is that it is an analysis of fiscal sustain- ability of city government, which differs from a study of whether or not citizens living in a city are receiving the services they demand. There are many overlapping governmental units in most cities, such as school districts, counties, and a variety of special districts. Their functions, fiscal rules, and access to revenue instruments are different in many ways that likely affect their coping abilities in volatile periods of fiscal stress (Pagano and Hoene, 2010). To understand fiscal sustainability in these units requires further analysis similar to the approach of this research. Another approach is to study cities from the citizen perspective, which requires some methodology for measuring what an “average citizen” of a given city experiences in their local government given the many different possible combinations of governments that exist within cities.32

If the Great Recession is a useful stress test for the long run viability of city gov- ernment in the future economy where an economic divide exists between the city and national government, then the results of this study are encouraging for cities. These large cities were able to tolerate a large macroeconomic shock using the traditional property tax mechanism and with deficit spending.

REFERENCES Alm, James, Robert D. Buschman, and David L. Sjoquist. 2011. “Rethinking Local Government Reliance on the

Property Tax.” Regional Science and Urban Economics, 41, 320–331. Avi-Yonah, Reuven. 2000. “Globalization, Tax Competition and the Fiscal Crisis of the Welfare State.” Harvard

Law Review, 113(7), 1573–1676. Bergstrom, Theodore C. and Robert P. Goodman. 1973. “Private Demand for Public Goods.” American Economic

Review, 63(3), 280–296. Bloom, Howard S. and Helen F. Ladd. 1982. “Property Tax Revaluation and Tax Levy Growth.” Journal of Urban

Economics, 11(1), 73–84. Brenner, Neil. 1998. “Global Cities, Glocal States: Global City Formation and State Territorial Restructuring in

Contemporary Europe.” Review of International Political Economy, 5(1), 1–37. Chernick, Howard, Adam Langley, and Andrew Reschovsky. 2011. “The Impact of the Great Recession and the

Housing Crisis on the Financing of America’s Largest Cities.” Regional Science and Urban Economics, 41, 372–381.

Dadayan, Lucy. 2012. “The Impact of the Great Recession on Local Property Taxes,” Rockefeller Institute Brief (July).

Doerner, William M. and Keith R. Ihlanfeldt. 2011. “House Prices and City Revenues.” Regional Science and Urban Economics, 41, 332–343.

Edwards, John H.Y. 1986. “A Note on the Publicness of Local Goods: Evidence from New York State Municipali- ties.” Canadian Journal of Economics, 19(3), 568–573.

Fisher, Glenn W. 1996. The Worst Tax? A History of the Property Tax in America. Lawrence, KS: University of Kansas Press.

32A new intriguing project along these lines is the Lincoln Institute of Land Policy’s “Fiscally Stan- dardized Cities.”

C© 2014 Wiley Periodicals, Inc.

 

 

138 JOURNAL OF REGIONAL SCIENCE, VOL. 55, NO. 1, 2015

Fisher, Ronald C. 2007. State & Local Public Finance, 3rd Edition. Mason, OH: Thomson South- Western.

Glaeser, Edward L. and Janet E. Kohlhase. 2004. “Cities, Regions and the Decline of Transport Costs.” Papers in Regional Science, 83(1), 197–228.

Glaeser, Edward L. and Matthew G. Resseger. 2010. “The Complementary between Cities and Skills.” Journal of Regional Science, 50(1), 221–244.

Glaeser, Edward L. and Albert Saiz. 2004. “The Rise of the Skilled City.” Brookings-Wharton Papers on Urban Affairs, 5, 47–94.

Government Accountability Office, 2013. “States Implementing Teacher and Principal Evaluation Systems De- spite Challenges.” GAO-713-777 (September 18).

Hines, James H. and Lawrence H. Summers. 2009. “How Globalization Affects Tax Design.” NBER Working Paper No. 14664.

Holcombe, Randall G. and Russ S. Sobel. 1995. “Empirical Evidence on the Publicness of State Legislative Activities.” Public Choice, 83(1–2), 47–58.

Johnson, Craig L., Sharon N. Kioko, and W. Bartley Hildreth. 2012. “Government-wide Financial Statements and Credit Risk.” Public Budgeting and Finance, 32(1), 80–104.

Kim, Sukkoo and Marc T. Law. 2012. “History, Institutions, and Cities: A View from the Americas.” Journal of Regional Science, 52(1), 10–39.

Ladd, Helen. 1991. “Property Tax Revaluation and Tax Levy Growth Revisited.” Journal of Urban Economics, 30(1), 83–99.

Lutz, Byron, Raven Malloy, and Hui Shan. 2011. “The Housing Crisis and State and Local Government Tax Revenue: Five Channels.” Regional Science and Urban Economics, 41, 306–319.

Manyika, James, Jaana Remes, Richard Dobbs, Javier Orellana, and Fabian Schaer. 2012. “Urban America: US cities in the global economy.” McKinsey Global Institute Technical Report.

McMillan, Melville L., W. Robert Wilson, and Louise M. Arthur. 1981. “The Publicness of Local Public Goods: Evidence from Ontario Municipalities.” Canadian Journal of Economics, 14(4), 596–608.

Mikesell, John L. 2011. Fiscal Administration: Analysis and Applications for the Public Sector, 8th Edition. Boston, MA: Wadsworth Cengage Learning.

Netzer, Dick. 1964. “Income Elasticity of the Property Tax: A Post-Mortem Note.” National Tax Journal, 17(2), 205–207.

Pagano, Michael A. and Christopher W. Hoene. 2010. “States and the Fiscal Policy Space of Cities” in Michael Bell, David Brunori, and Joan Youngman (eds.), The Property Tax and Local Autonomy. Cambridge, MA: Lincoln Institute of Land Policy.

Pagano, Michael A., Christopher W. Hoene, and Christiana McFarland. 2012. “Research Brief on America’s Cities: City Fiscal Conditions in 2012,” National League of Cities (September).

Ross, Justin M. and Wenli Yan. 2013. “Fiscal Illusion from Property Reassessment? An Empirical Test of the Residual View.” National Tax Journal, 66(1), 7–32.

Seljan,Ellen C. 2013. “The Limits of Tax and Expenditure Limits: TEL Implementation as a Principal-Agent Problem.” Public Choice (January), DOI 10.1007/s11127-013-0063-0.

C© 2014 Wiley Periodicals, Inc.

 

 

Copyright of Journal of Regional Science is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.

Get professional assignment help cheaply

Are you busy and do not have time to handle your assignment? Are you scared that your paper will not make the grade? Do you have responsibilities that may hinder you from turning in your assignment on time? Are you tired and can barely handle your assignment? Are your grades inconsistent?

Whichever your reason may is, it is valid! You can get professional academic help from our service at affordable rates. We have a team of professional academic writers who can handle all your assignments.

Our essay writers are graduates with diplomas, bachelor, masters, Ph.D., and doctorate degrees in various subjects. The minimum requirement to be an essay writer with our essay writing service is to have a college diploma. When assigning your order, we match the paper subject with the area of specialization of the writer.

Why choose our academic writing service?

  • Plagiarism free papers
  • Timely delivery
  • Any deadline
  • Skilled, Experienced Native English Writers
  • Subject-relevant academic writer
  • Adherence to paper instructions
  • Ability to tackle bulk assignments
  • Reasonable prices
  • 24/7 Customer Support
  • Get superb grades consistently

 


Get Professional Assignment Help Cheaply

fast coursework help

Are you busy and do not have time to handle your assignment? Are you scared that your paper will not make the grade? Do you have responsibilities that may hinder you from turning in your assignment on time? Are you tired and can barely handle your assignment? Are your grades inconsistent?

Whichever your reason may is, it is valid! You can get professional academic help from our service at affordable rates. We have a team of professional academic writers who can handle all your assignments.

Our essay writers are graduates with diplomas, bachelor's, masters, Ph.D., and doctorate degrees in various subjects. The minimum requirement to be an essay writer with our essay writing service is to have a college diploma. When assigning your order, we match the paper subject with the area of specialization of the writer.

Why Choose Our Academic Writing Service?

  • Plagiarism free papers
  • Timely delivery
  • Any deadline
  • Skilled, Experienced Native English Writers
  • Subject-relevant academic writer
  • Adherence to paper instructions
  • Ability to tackle bulk assignments
  • Reasonable prices
  • 24/7 Customer Support
  • Get superb grades consistently

How It Works

1.      Place an order

You fill all the paper instructions in the order form. Make sure you include all the helpful materials so that our academic writers can deliver the perfect paper. It will also help to eliminate unnecessary revisions.

2.      Pay for the order

Proceed to pay for the paper so that it can be assigned to one of our expert academic writers. The paper subject is matched with the writer’s area of specialization.

3.      Track the progress

You communicate with the writer and know about the progress of the paper. The client can ask the writer for drafts of the paper. The client can upload extra material and include additional instructions from the lecturer. Receive a paper.

4.      Download the paper

The paper is sent to your email and uploaded to your personal account. You also get a plagiarism report attached to your paper.

 

smile and order essaysmile and order essayPLACE THIS ORDER OR A SIMILAR ORDER WITH STUDENT ASSIGNMENT AID TODAY AND GET AN AMAZING DISCOUNT

order custom essay paper