An evaluation of the connection between economic volatility and governmental savings levels for Illinois, North Carolina, and Mississippi counties.

Author:Stewart, LaShonda M.

    Scholars and practitioners alike studying government savings have long grappled with understanding what constitutes an appropriate level of savings to ensure fiscal stability. Identifying an appropriate level of governmental savings, however, is a difficult task. Each government will differ in its cyclical vulnerabilities, but this depends on the composition of its revenue sources (Wolkoff, 1987; Stewart, 2009). Governments also respond differently to savings. Gold (1984) suggests, for example, that the level of savings varies from state to state and over time, but the optimal size depends on the government's "desire for stability in tax rates and expenditures" (p. 426). But this difficulty does not mean that we cannot gain some leverage toward understanding what motivates governments to save when previous researchers like Massey and Tyer (1990) note a large discrepancy in the amount of savings local governments hold. They find that governments in South Carolina maintain anywhere from a negative fund balance to over 700% of general fund revenues. Stewart (2009) identifies a similar pattern in some counties in Mississippi, which maintained savings ranging from below zero to over 200% of expenditures. While 700% and 200%, respectively, may appear to be excessive levels of savings, economic conditions of a particular government may suggest otherwise and warrant savings of this magnitude. Hence, what is considered an appropriate level for one jurisdiction may not be appropriate for another, thus making it challenging to identify a standard benchmark that is fitting across all jurisdictions (Stewart, 2009). The issue of identifying an appropriate level of savings is further compounded by the fact that there is a lack of recommendations to help guide local governments in determining appropriate savings levels for their environment.

    The key to understanding the appropriate level of government savings is to identify potential risks of economic volatility. Some governments experience higher levels of volatility and need to save more, while other governments have very stable economic environments and can function with lower levels of savings (Joyce, 2001; GFOA, 2009). If governments are saving based on the needs of their environment, there should be a direct link between the amount of money governments save and their level of volatility.

    Wolkoff (1987) speculates that local governments facing a more risky or uncertain environment are more likely to maintain a higher level of savings. However, he finds, after examining 27 of the most populous U.S. cities, no relationship between savings and volatility in the environment. He concludes that the governments that are least likely to need a rainy day fund (RDF) are the ones that are most likely to have one. Over a decade later, Joyce (2001) finds similar results and argues that there is no significant relationship between volatility and the level of states' rainy day funds. Building on these works as well as the work of Stewart (2009), who examines local governments in Mississippi, Hendrick (2006), who focuses on local governments in Illinois, and Wang and Hou (2012), who focus on local governments in North Carolina, the authors of this study assess if local governments in Mississippi, North Carolina, and Illinois are saving according to their economic volatility for FY 2005-2010. If local government savings levels are in line with their economic volatility, it can be assumed that they understand their unique environments and save accordingly.



    Although local government savings are nothing new (Aronson, 1968), maintaining a recommended level of an unreserved fund balance (1) (left over money that remained unspent by the end of the fiscal year) has recently received attention in the literature. This has gained the attention of researchers because some local governments have been found to maintain large amounts of savings to deal with both short- and long-term community needs (Stewart, 2009). Massey and Tyer (1990), for example, find that "fund balances as a percentage of general fund expenditures ranged from zero to over 700 percent" (p. 44) for cities in South Carolina with populations under 1,000 residents. Hembree and Shelton (1999, p. 17) suggest, however, that, "One of the most difficult questions that policymakers and finance professionals confront ... is the adequacy of the level that should be maintained by governments."

    Despite the endorsement of professional organizations like the National Association of State Budget Officers (NASBO), National Conference of State Legislators (NCSL), and the Government Finance Officers Association (GFOA), the widely cited recommended benchmark of 5-15% or two months of general fund expenditures for local governments has been disputed by researchers of both local and state governments. These researchers argue that the proposed level is not sufficient to sustain expenditures during economic downturns (Hou, 2003; Douglas and Gaddie, 2002; Sobel and Holcombe, 1996; Navin and Navin, 1997; Wagner, 2003; Stewart, 2009, 2011; Marlowe, 2005). Thus, the savings recommendations that currently exist do not appropriately guide governments in determining adequate funding levels, especially those governments that have higher levels of economic volatility.

    The literature provides a wide range of recommendations and insights on what is considered to be an appropriate level of savings. Kriz (2003) examines cities in Minnesota and attempts to establish an optimal level of savings for local governments. He concludes that a reasonable estimate of local government savings is at least 30% of total revenues. Shelton et al. (2000) note that local governments in North Carolina and South Carolina have average savings of between 20% and 50%. They note, however, that a common rule of thumb for local savings is no more than three months of general fund revenues. Marlowe (2011) further implies that savings above 50% may be too high and exceed optimal slack resources. (2) Still, researchers are not in agreement on the upper limits of how much governments should save. Joyce (2001) and Wolkoff (1987), however, suggest that the level of savings should match the volatility of the government's economic environment.


    Joyce (2001) builds upon previous research by assembling a national sample and assesses if states save enough to match their unique volatility. He examines various measures (unemployment, corporate income taxes, intergovernmental revenues, gaming revenues, and Medicaid spending) that potentially induce volatility into state governments' economic environments. These variables were used to create an instrument that measures the overall level of volatility for each state. After determining the overall level of volatility for each state, Joyce compares each state's potential for volatility to its savings, under the assumption that state savings should, at the very least, match its volatility. He argues, for example, that unemployment is a proxy for volatile economic environments. In other words, economic environments that are comprised of volatile industries will have higher unemployment rates over the years and a higher unemployment rate may indicate a need for higher savings levels in good economic conditions to help stabilize expenses during poor economic times. Higher unemployment rates during economic downturns lead to decreased revenue and greater demands for government resources (Wang and Hou, 2012); thus, we should expect higher rates of unemployment to decrease unreserved savings. GFOA also acknowledges that governments should consider the predictability of their revenues as well as the volatility of expenditures when planning savings levels (GFOA, 2009).

    Joyce (2001) argues that during tough economic times, state governments will look for ways to cut expenses. Money transferred to other governments is likely to be considered for budgetary cuts. Thus, as local governments increasingly rely on intergovernmental transfers as a significant portion of their income, the potential for volatility also increases. Both Hendrick (2006) and Stewart (2009) further argue that a heavy reliance on unstable sources like intergovernmental revenues will positively impact the level of saving if governments are cognizant of the risk involved in relying on this source of income. This argument is applicable to other revenue streams (e.g., user fees, sales tax, etc.) that are sensitive to the economy. Property tax revenues are more stable than other revenue sources because changes in personal income do not have an immediate impact on property value (Stewart, 2009). Being consistent with the arguments of Wolkoff (1987) and Joyce (2001), we expect those counties facing a more risky or uncertain environment to maintain a higher level of savings. So, as property tax revenue volatility, other revenue volatility, intergovernmental revenue volatility, and expenditure volatility increase, savings should also increase.

    Hendrick (2006) finds a negative relationship between debt per capita and governmental savings. Based on this finding, the authors argue that long-term debt holds the potential to induce volatility during times of fiscal stress. Governments with high levels of debt may find it difficult to meet financial obligations when revenues are below expectations. Moreover, high debt increases borrowing cost, especially during poor economic conditions (Grizzle, 2010; Wagner, 2003; Marlowe, 2011). Savings can relieve the pressure of large debt obligations by giving governments a cushion to fall back on during lean years. It is equally valid to assume that a county engaging in debt creation may not have any savings to rely on. Thus, we should expect a connection between higher levels of debt and lower levels of savings.



To continue reading