Determinants of debt burdens: evidence from California counties.

Author:Wang, Janey Qian

    Public debt plays a critical role at the sub national economy level because government's ability to borrow provides an essential source of financing for longer term investments in transportation systems, building and rebuilding schools, hospitals, prisons, natural resources, energy and economic development. While state and local government outstanding debt is a relatively smaller component of total US debt (according to the Federal Reserve statistics- it was 6.7% of the total $35.5 trillion debt in the U.S. in 2010) it has increased significantly since 2000 from $1.1 trillion to $2.4 trillion in 2010.

    This research study will focus primarily on California counties' long-term debt. While debt financing can be and has been used to cover short-term budget shortfalls and to smooth financial management cycles, this has not been a really significant "bad" fiscal habit of California counties. Although debt is not solely issued for capital purposes, it is most often used for this purpose; therefore, it helps smooth investment on infrastructure over time and spreads the repayment burden to future generations who will benefit from the investment.

    Future infrastructure development will undoubtedly rely on both debt financing and tax revenues. However, to what extent can governments shift the costs of its infrastructure from the General Fund to debt financing is a hard question. Solely relying on tax revenue, fees and intergovernmental transfers may result in under-development of infrastructure, especially in the current economic recession and what has already been shown to be a slow and painful recovery. Understanding the determinants and limitations of local governments' debt financing has already emerged as a critical issue for the future of California.

    This research study aims to explain what factors have affected changes in debt levels, but goes beyond existing research by developing and implementing a measure of local governments' relative debt reliance. Our research contributes to the field of public finance in two different ways. First, our study develops an improved methodology for measuring debt burden. Second, our results provide insight into how variables such as budget institutions, demographic reserve funds, and levels of intergovernmental transfers affect debt burden, and add to recent research on debt reliance.

    The paper proceeds as follows. Section 2 discusses previous research. Section 3 describes the data. Section 4 and 5 discuss the model selection and empirical results. In section 6, we compute the debt reliance indices and discuss policy implications, and in section 7, we give concluding remarks.


    The debt burden of each jurisdiction is typically determined by the economic situation, government management performance, political environment, and market borrowing cost. The literature on debt burdens has generally followed four broad categories of explanatory variables- institutional, political, economic, and structural (Garand & Kapeluck, 2004).

    A first group focuses on the economic situation. Theoretically, high-income residents prefer debt financing, because they can benefit from lower current tax burden (Cunningham, 1989). In addition, a wealthier state can more easily finance capital expenditures through their savings and current tax revenue rather than through borrowing (Clingermayer and Wood, 1995). Cunningham (1989) additionally finds that the unemployment rate has a positive impact on the debt burden.

    The second group examines demographic variables. Ellis and Schansberg (1999) find that the proportion of younger voters has a positive impact on debt, while old voters have a negative impact on debt. This could be explained with the fact that state governments issue debt disproportionately to spend on programs that benefit younger voters, such as education, welfare and transportation.

    A third group explores politics. The party affiliation of localities is also related to their debt level. For example, it is generally believed that Democrats have a propensity to spend more and borrow more (Ellis & Schansberg, 1999). The degree of competition between political parties also has a significant impact on states' indebtedness (Baber and Sen 1986; Poterba 1994, Clingermayer and Wood, 1995, Kreuger and Walker 2010).

    The fourth group turns attention to fiscal institutions. Line item veto power for governors, various balanced budget requirements, and debt limitations have helped determine states' debt levels, although the research has not found consistent results regarding fiscal institutions' impact on public debt levels (Rowley, Shughart and Tollison, 1987; Von Hagen 1991; Kiewiet and Szakaly, 1992).

    Finally, the demand function of debt level includes factors such as the level of government capital outlay, the price of debt, the tax price, and the current capital investment level. Eberts and Fox (1992) demonstrate that the share of total city funds obtained through borrowing has a positive relationship with the level of capital investment, which means the more capital outlay currently underway, the more likely it is that localities are going to issue debt, since the increased capital outlay is usually financed by long term debt. Tax prices will affect the debt level as well. For example, the higher the tax price, the more likely the demand for debt will go up. The price of the debt is the borrowing cost, which is supposed to have a negative impact on debt level.

    There are some important similarities to note about the previous research. First, most of the existing research uses per capita debt measures as the dependent variable. This measure may be problematic as it ignores the wealth of a community. Communities that are relatively wealthier may be able to better afford more debt per capita. While some of the studies have controlled some what for this by entering per capita income into their models, a much more straightforward way to proceed would be to measure debt as a percentage of a variable that captures wealth such as assessed valuation. Second, most of the prior research has focused on the determinants of the level of debt in a jurisdiction. These papers have completed their analysis at the point where they are able to test whether some variable affects debt burden. From a practical standpoint, a much more interesting question is whether a given jurisdiction has a debt burden that is greater than or less than predicted. The answer to this question can help jurisdictions (and those monitoring their finances) assess their debt burden in the context of known information about the debt burdens of other jurisdictions.

  3. DATA

    Although there is an extensive literature on the determinants of state debt financing, there has been little research on the determinants of local debt financing, especially at the county level. This study therefore uses panel data on counties from the state of California to examine local debt affordability issues.

    California has 58 counties and 480 cities and towns. San Francisco, due to its special characteristics-it's both city and county, is...

To continue reading