The effects of tax and expenditure limits on state fiscal reserves

Abstract

To help policy makers manage expenditures during periods of economic downturns, most states have formal budget stabilization funds and unreserved fund balances. Using indices of tax and expenditure limitations laws restrictiveness, we examine the relationship between tax and expenditure limitations and state reserves for years 1992–2010 to help determine the extent to which tax and expenditure limitations constrain or in other ways affect how states manage fiscal reserves. This time period is particularly relevant because it includes two recessions and most states had budget stabilization funds and tax and expenditure limitations. Findings suggest that state-constructed tax and expenditure limitations have little effect on state capacity to react to fiscal shocks.

Keywords

State government economic recession fiscal condition budget stabilization funds unreserved fund balances tax and expenditure limitations

Introduction

The Great Recession of 2008–2009 highlighted a number of troublesome financial management issues at all levels of government in the United States (US). One of these issues at the state level is the adequacy of reserves to weather revenue shocks caused by instability in the economy. While historical management best practices have called for state reserves, or rainy day funds, to be equal to at least five percent of general expenditures, the fiscal crisis created by the Great Recession has called into question the appropriate level of reserves. A recent report by the Pew Charitable Trusts (2014) finds that state rainy day funds were largely inadequate and thus unable to smooth out the fiscal shocks caused by the Great Recession. According to this, “[s]tates should analyze the amount of savings needed to effectively mitigate volatility, and they should routinely evaluate whether those levels are appropriate” (p.6). This approach is consistent with the best practice recommendations of both professional fiscal officers (Government Finance Officers Association, 2014) and scholars of public administration and finance (Joyce, 2001; Marlowe, 2011, 2013).

As budgets slowly recover, should the increased revenues be used to restore previous spending levels, returned to the taxpayer, used to pay down debt that might have been incurred during the fiscal crisis, or set aside in a rainy day fund as an “insurance policy” against future downturns? Indeed, the most recent state fiscal data suggest that the recovery from the Great Recession has been less than robust with strong evidence of continued instability (Dadayan and Boyd, 2014). In the 10-year period prior to the Great Recession, the average quarterly increase in state and local government revenues was about 1.4 percent but the average quarterly increase since the end of the Great Recession was only 0.9 percent. According to Dadayan and Boyd (2014), increased instability in revenues post the Great Recession can be attributed to uncertainty as to the strength of the overall economic recovery and specifically federal fiscal policies.

In addition to rainy day funds, also called budget stabilization funds (BSFs), states have ending fund balances. Hou's (2003) analysis of the extent to which these two funds act to stabilize state spending revealed that BSFs helped stabilize expenditures but unreserved undesignated fund balances did not. Regarding the latter, Hou (2003) infers that following state creation of BSFs, the traditional counter-cyclical role of unreserved undesignated fund balances has changed and thus leaves open the current/future function of these fund balances.

A more theoretically rich analysis by Hou et al. (2003) links BSFs and the rules governing their deposits, withdraws, and balances to organizational capacity and management. These BSF rules are treated as constraints on public officials that can positively impact financial management outcomes as measured by BSFs relative to expenditures. According to the authors

[r]estricting governmental discretion by requiring adequate contribution to the fund and preventing inappropriate use of the fund will lead to higher [BSF] balances and a greater likelihood that the fund will be solely used for the purpose it was designed. (p.301)

Hou et al. (2003) find that, in general, financial management constraints (stricter BSF rules) are positively associated with financial performance.

Within this theoretical context, tax and expenditure limitations (TELs) by their very definition can be conceptualized in a similar manner: TELs are fiscal constraints that should affect government financial management capacity. In theory, the rules of the BSF and/or TELs impose fiscal discipline on political decision makers. The question is whether the constraints created by TELs have a positive or negative effect on financial performance, narrowly defined for this study as the size of reserves.

Hou's (2003) work coupled with that of Wagner and Sobel (2006) find that BSF adoption and structure (rules of adoption and withdrawal) are linked to TELs and call into question the extent to which TELs influence end of year unrestricted balances and BSFs. A testable question for this study is whether states with more constraints on the flexibility of fiscal policy are more likely to set revenues aside in a rainy day fund or are they more likely to use every dollar available to maintain funding levels and/or reduce taxes? We would argue that formal rainy day funds would be more difficult to tap into from a political perspective than unrestricted fund balances and in order to maintain some flexibility states with more restrictive TELs are more likely to utilize unrestricted fund balances.

To test the question of how the restrictiveness of TELs influences the levels of state stabilization funds and end of year balances, this study uses a panel of U.S. state data from 1992 to 2010 and a modification of an index of TEL restrictiveness developed by Amiel et al. (2009). Beyond these introductory comments, the study is composed of six sections: (1) a review the relevant literature on budget stability; (2) an overview of research on TELs; (3) a descriptive analysis of BSF, fund reserves, and TELs; (4) the empirical modeling results; and (5) a review of the key findings of the study.

Literature review

Both academic and professional-focused public finance literatures appear to accept a priori that state governments need to take an active role in crafting policies that help mitigate the effects of economic downturns and instability on service provision (Hou and Moynihan, 2008; Kwak, 2014). This is, historically, the principal justification for the retention of tax dollars in funds without a designated purpose. Now, given the nearly universal adoption of formal BSFs by states and a literature questioning the effectiveness of the complementary unreserved undesignated fund balances in mitigating expenditure variability, we are left questioning the current/future function of these reserves (Hou, 2003). A further question is if TELs affect the ability of state reserves to fulfill their central function.

Literature on state BSFs can generally be divided into four categories: (1) state adoption of BSFs and associated deposit/withdraw rules; (2) the nature of BSFs (e.g., are they counter-cyclical); (3) the adequacy of BSFs and; (4) determinants of BSF and total reserves balances. It should be pointed out that the literature tends to be less theoretical and focused more on the observed outcomes affected by variation in fund structures. In other words, the literature tends to be inductive in nature drawing on analysis of the available data rather than deductive from stylized theoretical constructs.

Forty-seven of the 50 states currently have some form of BSF, only Illinois, Kansas, and Montana have no such fund. The first BSF was established by New York in 1945 (Wagner and Sobel, 2006), but the majority were adopted in the 1980s following the recession of the early part of that decade. The argument has generally been that state adoption of BSFs was based on the realization that policy makers were inadequately prepared for the fiscal shocks caused by economic downturns and instability (Douglas and Gaddie, 2002; Gold, 1983). BSFs are similar to a self-funded insurance policy against lagging revenues and social support program expenditure increases during poor economic times.

An alternative argument, offered by Wagner and Sobel (2006), is that BSFs were created less to cope with fiscal shocks flowing from fluctuations in the economy than as a mechanism to circumvent TELs. They describe the deposit/withdrawal rules associated with BSFs, noting that many did not have strict rules governing deposits into or withdrawals from these funds. They argue that state lawmakers would be able to take advantage of loosely structured deposit and withdrawal rules to circumvent TEL-based restrictions. “By adopting and depositing surplus funds into a stabilization fund, the state decision makers may effectively nullify the effect of a tax or expenditure limit law since money in stabilization funds are not subject to TEL oversight” (Wagner and Sobel 2006, p.182). Their findings offer compelling evidence that the design of BSFs, both in terms of whether the rules are governed by statute or state constitution and the degree of stringency of deposit and withdrawal rules, are associated with the TEL requirements pertaining to the management of budget surpluses.

Regardless of the circumstances surrounding the adoption of BSFs, their effectiveness at smoothing fiscal shocks has been the focus of several scholars, most notably Hou (2003, 2005, 2006; with Moynihan, 2008; with Duncombe, 2008; with Smith, 2006). Hou (2003) finds that BSFs were effective in counteracting the effects of economic downturns on state expenditures, while unreserved undesignated fund balances did not have the same counter-cyclical effect. The findings are consistent with Sobel and Holcombe (1996) and Douglas and Gaddie (2002) that BSFs helped alleviate some of the fiscal pressures on states caused by the 1990–1991 recession. Wagner and Elder (2005) found that, “state expenditures are approximately 20 percent less volatile following the adoption of a rule-bounded budget stabilization fund” (p.439). A smaller-scale study of Midwestern states in the 1980s to early 1990s finds that policy makers in Michigan and Indiana consistently funded their BSF and, in turn, weathered the 1990–1991 recession better (stabilized expenditures) than the other states in the region. (Navin and Navin, 1994). Navin and Navin argue that BSFs in the other states (Ohio, Wisconsin, Iowa, Minnesota and Missouri) are typically underfunded and drawn down for purposes other than counter-acting down swings in the economy. In a more recent study of state fiscal behavior, Kwak (2014) finds that the existence of BSFs had no measurable effect on state expenditure gaps, measured as deviations in per capita real expenditures from the trend. The differences in findings across studies may be the result of different time frames, different research questions, and differences in measurement; Kwak measures BSFs as a dichotomous variable, whereas Hou and others measure the BSF funding levels.

The third focus of the BSFs literature has been to assess the adequacy of their size. One of the more common caps imposed by states on their BSFs is five or 10 percent of either appropriations or revenue (Joyce, 2001; Pew, 2014). Hou et al. (2003) seminal piece examined the size of BSFs as a proportion of general fund expenditures from the perspectives of funding sources, constraints on balances, deposit, and withdrawal rules and party politics. They find that the stricter rules generally constrained policymakers' ability to draw down the reserves and led to stronger balances. More recently, the Pew Charitable Trusts (2014) studied the size of BSFs, deposit rules, and state revenue volatility. The takeaway from the Pew study is that state BSFs were inadequate to overcome the fiscal effects of the Great Recession – and preceding recessions for that matter. The Pew authors argue that part of the issue is that state policy makers often cap BSFs too low. In addition to problems with deposit rules, the Pew study identifies concerns with withdrawal rules that can be either too lenient or too restrictive. The key finding by the Pew (2014) authors is, “[a]lthough most states recognize the importance of having a fund to smooth the booms and busts of the revenue cycle, few base the size of the rainy day fund on their own typical revenue fluctuations” (p.4). From our perspective, not only are rules governing BSFs important but other institutional rules such as TELs are worthy of examination within the context of state reserves.

Lastly, previous work has examined the factors affecting state fiscal reserves and BSF balances. As discussed above, rules governing BSF practices have been identified as one important factor: the structural or design attributes of deposit/withdraw rules (e.g., strict vs. weak) are directly associated with BSF balances (Gold, 1984; Hou, 2004; Knight and Levinson, 1999; Vasche and Williams, 1987). State economic conditions also have been found to shape state saving behavior. For example, studies have shown that state personal income, a proxy of state economic capacity (Knight and Levinson, 1999; Navin and Navin, 1994), and unemployment rates, a measure of public service demand (Hou and Brewer, 2010; Knight and Levinson, 1999), are positively and negatively associated with state fiscal reserves, respectively. State budgetary features have also been examined in the sense that higher state spending has been negatively associated with state reserves (Hou and Brewer, 2010; Vasche and Williams, 1987). In contrast, higher levels of own source revenues (Navin and Navin, 1994) and tax structure stabilization (Rodriguez-Tejedo, 2012; Zahradnik and Johnson, 2002) have been reported as factors playing a role in state fiscal reserves. The impact of other political and institutional factors has also been explored (Gold, 1995; Hou, 2004; Hou and Brewer, 2010; Lauth, 2003; Rose, 2008). Although studies have shown different results depending on their approaches, there are three common themes: (1) fiscal institutions which constrain a state's fiscal choices (e.g., balance budget requirements) contribute to increases in fiscal reserves; (2) policy makers' political motivations tend to affect the increasing use of reserves, and; (3) political conflict among policy makers within a state is likely to prevent the discretional withdraw of fiscal reserves.

As previously discussed, Wagner and Sobel (2006) call into question the role of TELs on state reserves but on a broader scale, the effects of TELs are increasingly found to affect fiscal outcomes. These include poorer infrastructure (Deller et al., 2013a) varying effects on debt levels (Deller et al., 2013b), effects on bond ratings (Stallmann et al., 2012; Johnson and Kriz, 2005), and economic performance (Deller et al., 2012; McGuire and Rueben, 2006). Furthermore, while Poterba (1994) does not directly test the relationship between TELs and fund balances, he does find that TELs affect tax policy and that fund balances affect spending and tax policy. This leaves open the question of the direct effects of TELs on fund balances. It is this dynamic between institutional constraints and fiscal condition, as measured by fund balances (Poterba, 1994), that is understudied.

As discussed above, much of the scholarship regarding state reserves has been inductive in nature where scholars are trying to extract insights from the data. From a more theoretical perspective, we can think of state TELs as institutional constraints imposed by political actors to control the taxing and spending behavior of governments in accordance with their preferences (Hur, 2007; Mullins and Wallin, 2004; Poterba and Rueben, 1999b). These restrictions on government fiscal or budgetary decision-making are not only means to limit policy alternatives (von Hagen, 2002), but also instruments designed to establish rules to be followed in allocating resources (Poterba and von Hagen, 2008; von Hagen and Harden, 1996). In addition, fiscal reserves can be viewed as budgetary choices which are relatively counter-cyclical, long-term, cumulative, and discretionary (Hou, 2003; Jacob and Hendrick, 2012; Rose and Smith, 2012). In this sense, the political incentive structure is key in understanding the relationship between TELs and state reserves.

Depending on one's theoretical standpoint, the direction of the relationship between TELs and state reserves varies. For instance, the stringency of TELs may be negatively associated with the level of fiscal reserves in order for policy makers to be perceived as more fiscally efficient (Brennan and Buchanan, 1979; Mullins and Wallin, 2004). That is, strict TELs are likely to be implemented with the intention of reducing government size and decreasing tax revenues without cutting services (Ladd and Wilson, 1982; Lowery and Sigelman, 1981). From a principal–agent perspective, institutional constraints are expected to make up for incomplete contractual arrangements between principals and agents (Seljan, 2014; von Hagen, 2002). Conversely, the stringency of TELs may be positively associated with the level of fiscal reserves as policy makers may be inclined to fulfill their self-preferences and maximize their own utility (Eisenhardt, 1989; Miller, 2005). Another possibility is that policy makers define themselves as professional managers/trustees who are primarily concerned about exposure to future risk and collective benefits (Balk and Calista, 2001). In either case, increasing fiscal reserves is a practical means of securing fiscal flexibility. The stringency of TELs is therefore likely to motivate decision makers to have a higher level of fiscal reserves.

The relationship between TELs and state reserves may not be as static as described above; it may vary depending on the institutional attributes of TELs and state fiscal reserves. For example, more stringent TELs on revenues (e.g., caps on all revenues) may encourage decision makers either to reduce the size of fiscal reserves or to set more current revenues aside in a reserve fund by means of “arbitrage” strategies (e.g., overestimating expenditures; Metcalf, 1989) for future spending which is not constrained by revenue TELs (Danziger and Ring, 1982). Similarly, strict expenditure TELs can also result in the reduction of fiscal reserves as a demonstration of political will to reduce the size of government.

Conversely, TELs with expenditure limits tend to leave room for the excessive collection of revenues (Amiel et al., 2014) and for increases in fiscal reserves. Under the most restrictive TELs – both revenue and expenditure limits – decision makers also encounter similar competing motivations but the means of having higher level of reserves may be more limited than in other cases due to the TELs stringency. It should also be noted that these hypothetical relationships can be affected by the type of fiscal reserves. That is, reserve funds such as BSFs may be less usable for decision makers than unrestricted fund balances because they generally governed by formal rules (Rose and Smith, 2012). We are, therefore, left with theoretical and empirical justification for a relationship between TELs and state reserves; however, the direction of the relationship remains elusive.

TELs research

Our primary interest is modeling the determinants of state reserves as measured by ending balances and BSFs. More specifically, we focus on the effects of state-imposed TELs on BSFs, ending fund balances, and total reserves (BSFs plus fund balances). The effects of TELs have been the focus of numerous studies which have operationalized TELs in several different ways. The earlier literature uses a cross section of state or municipal observations and employs simple descriptive analysis such as Howard (1989). The second is a case-study approach using a variation on the with-and-without quasi-experimental design to examine fiscal policy in a state pre- and post-imposition of the particular TEL (Dye et al., 2005, 2006; Fisher and Gade, 1991; Maher et al., 2011; Skidmore et al., 2010; Springer et al., 2009). Within the case-study approach literature, the two TELs that have been examined the most are Massachusetts's Proposition 2½ (e.g., Bradbury et al., 2001; Cutler et al., 1999; Lang and Jain, 2004) and California's Proposition 13 (e.g., Downes, 1996; Wasi and White, 2005).

The third approach uses panel data at either the state or local level within a quasi-experimental with-and-without framework. A metric to capture the presence of a particular type of TEL, traditionally a simple dummy variable (taking on a value of one if a particular TEL is present, the treatment, zero otherwise, or untreated), is regressed on a metric of government revenues and/or expenditures (e.g., Deller and Stallmann, 2007; Mullins, 2004; Preston and Ichniowski, 1991). Shadbegain (1999, 2003) uses a 1972 to 1992 panel of local government budgets, aggregated to the state level, to see if state-imposed TELs influenced the size of local governments. In both studies, Shadbegain concludes that TELs have had minor impacts on local government. Blankenau and Skidmore (2004) use a panel of state data from 1971 to 1993 to examine the interplay between state-imposed TELs and court-mandated decreases in public education spending inequality. They find that court-order reforms generally dampen the impacts of TELs. Kousser et al. (2008) use panel data from 1969 to 2000 to look at the impact of TELs on state spending and find that, other than in a small number of states, such as Colorado and Missouri, TELs have not curtailed state spending.

The complexity or heterogeneity of TELs across states is a major reason why researchers often take a case-study approach, focusing only on a particular state. To address this heterogeneity, some researchers focus on specific types TELs, such as McCubbins and Moule (2010) who focus their analysis on limitations that target the property tax through the use of a dummy variable. Others, such as Kousser et al. (2008) and Poterba and Rueben (1999a and 1999b), focus on TELs that constrain the whole of taxing and spending, not single segments, such as property tax limits. But again, a dummy variable is used to capture the presence of the TEL. Primo (2006) uses a dummy variable if a spending limit TEL is present. Others that focus on fiscal responses to TELs and use dummy variables include Abrams and Dougan (1986), Bowler and Donovan (2004), Dye, et al. (2005), New (2001), Preston and Ichniowski (1991), Shadbegian (1998 and 1999) and Skidmore (1999).

The problem with the dummy variable approach is that it masks important differences in restrictiveness across states and even within states over time. No two states are exactly alike and states seldom put a TEL in place and do not alter it over time. Amiel et al. (2014) appeal to information theory to argue that compressing the complexity of how individual TELs are structured into a simply dummy variable loses important information and masks the impact of TELs on policy and fiscal outcomes. There are several studies that have taken a different approach, constructing indices to reflect the strength or restrictiveness of a state's tax and expenditure limit. Bae and Gais (2007), for example, build an index that ranges from zero for no TEL to three for the strongest TEL. They find that higher values of the index are associated with modestly lower levels of state government per capita spending.

Poulson (2005) uses five dimensions to build an ordinal index of state TEL restrictiveness for the year 2005, which ranges from zero for states that have no effect TEL in place to a maximum possible score of 25. Poulson's (2005) index reflects the heterogeneity of TELs, but is limited to one year. A one year index does not capture when the TEL was implemented or how it has changed over time, and it cannot be used for time-series or longitudinal analysis. Building on Poulson (2005), Amiel, et al. (2009) construct an annual index for each state, 1973–2005. Their index includes six characteristics of TELs: (1) type of TEL, (2) constitutional or statutory, (3) growth restriction, (4) method of approval, (5) override provisions, (6) exemptions. This index has been used by Nicholson-Crotty and Theobald (2011), Bae and Jung (2011) at the state-level and Marlowe (2011), Maher and Deller (2013) at the local-level.

The Amiel et al. (2009) was deconstructed into a set of indices by type of state TEL: (1) only revenue limits; (2) only expenditure limits; and (3) both revenue and expenditure limits. Additionally, each exemption to the TEL is assigned a value of negative one, because exemptions make a given TEL less restrictive. The highest score for revenue TELs is 20, for the expenditure TELs is 22, and for TELs that restricts both revenues and expenditures is 18. In 2009, 32 of the 49 states studied had some form of TEL (Table 1). The most common type is an expenditure limit (18 states), followed by tax and expenditure (eight states) and revenue limits (six states). It is also interesting to note that during the study period, TELs in 18 states were either modified, changed, eliminated, or suspended. Note that Colorado's Taxpayer Bill of Rights (TABOR) was suspended from 2005–2010.

Table 1.

State TEL patterns during study period

States with revenue TELs in 2009: 6
AR, FL, KY, MA, MI, SD
States with expenditure TELs in 2009: 18
AK, CT, HI, IA, ID, IN, MD, ME, MS, NC, NJ, OH, RI, SC, TV, TX, UT, WA
States with revenue and expenditure TELs in 2009: 8
AZ, CA, DE, LA, MO, NV, OK, OR
States without TELs in 2009: 17
AL, CO, GA, IL, KS, MN, MT, HH, NM, NY, ND, PA, VT, VA, WI, WY, WV
Note: CO's TABOR was suspended 2005–2010
Number of states that changed TEL between study period: 18
Two states ended TELs: CO, MT
Five changed to both revenue/expenditure TELs
CO (1992), LA (1993), NV (1996), OK (1992), OR (1996)
WA switched to expenditure TEL (2000)
Remaining states 10 modified existing TELs

As noted by Kioko (2011), while the movement to indices is a marked improvement over simple dummy variables, the limitation of an ordinal ranking is that individual coefficients lack a direct interpretation; at best, we can infer direction of the relationship and level of statistical significance. Despite this limitation, we believe that this restrictiveness index is an improvement over dummy variables for several reasons.¹ Dummy variables cannot capture the heterogeneity, or in a pure statistical sense, the variation that exists across TELs. In addition, legislatures and public initiatives amend existing TELs, making them more or less strict overtime. A dummy variable for the presence of a TEL cannot capture these policy changes. Finally, dummy variables can only test for linear shifts in the relationships between TELs and fiscal measures.

An index accounting for the restrictiveness of the TEL focuses attention on what state and local officials are most concerned with: how much flexibility they have in setting budgets and revenue policies. Second, the index adds greater variance to explore in the statistical modeling and also allows for changes in the TEL over time. Third, use of an index expands the estimation methods that can be employed. Fixed effects estimation is often the preferred method for panel data studies because it allows researchers to control for all of the time invariant characteristics of the states, even if they are not explicitly included in the data. Fixed effects estimators, however, cannot estimate coefficients for variables that do not vary over time, such as dummy variable for a state that has a TEL for the entire study period, because they are collinear with the intercept term (Greene, 2008). Therefore, researchers must employ another method, such as a random effects or pooled regression.

TELs and state reserves

During the period of this study, the U.S. endured two recessions, a relatively mild one in 2001 and a severe one in 2008–2009. How these recessions affected state revenues relative to expenditures is described in Figure 1. Leading up to the 2001 recession, revenues exceeded expenditures by one or two percentage points annually. Note how similar the effects of the recessions were on the revenue/expenditure proportions. At the troughs in 2002 and 2009, expenditures were greater than revenues by five percentage points. Also similar after both recessions in that it took three years for revenues to “catch up” with expenditures. The trends reflected in the graph also reveals that, as Joyce (2001) and others have been contending, the five-percent norm when it comes to reserves was inadequate in both recessions for many states (the graph shows a national average) and that a reserve in the range of eight to nine percent should be the standard.

Figure 1.

State revenues as a percentage of expenditures.

If we just look at national averages, states were responding to economic cycles in the prescribed manner, during times of growth (revenues exceeding expenditures) reserves grew and were drawn down during recession (Figure 2). Between 1992 and 2000, ending balances grew from one to seven percent of expenditures, and BSFs grew to six percent of expenditures during the same period. The dramatic effects of the 2001 and 2008 recessions on state reserves are also plainly visible. By 2002, nearly all state reserves were eliminated and did not begin recovering until 2004. The trend line also reveals that between the 2001 and 2008 recessions, ending balances recovered more rapidly than BSFs, being nearly twice as large at the peak in 2006 (nine percent versus five percent).

Figure 2.

BSFs and ending fund balance as percentage of expenditures.

Not surprisingly, this simple descriptive analysis masks the variability across states. State reserves, along with the corresponding value of the TEL index, for both 2001 and 2008 are provided in Table 2. Reserves ranged from levels exceeding total expenditures in Alaska (due to mineral/oil income) in both 2001 and 2008, to Arkansas carrying neither a budget stabilization fund nor ending balances in 2001 and 2008. Wisconsin and North Carolina also carried minimal reserves (less than two percent of expenditures) in 2001. In 2008, Wisconsin was still carrying a modest reserve (one percent), whereas North Carolina grew its BSF and ending balance so that total reserves were nearly seven percent in 2008. Interestingly, Arkansas has a revenue limit TEL, North Carolina has an expenditure limit TEL, and Wisconsin does not have a TEL, calling into question, at least anecdotally, our hypothesis that state TELs influence reserves.

Table 2.

State reserves and TEL measures, 2001 and 2008

	Year 2001						Year 2008
	Reserves as Pct of expenditures			Type of TEL			Reserves as Pct of Expenditures			Type of TEL
	Ending balance (%)	Stabilization funds (%)	Total reserves (%)	Rev.	Expend.	Both	Ending balance (%)	Stabilization funds (%)	Total reserves (%)	Rev.	Expend.	Both
AL	1.0	0.0	1.0	0	0	0	3.0	3.0	6.0	0	0	0
AR	0.0	0.0	0.0	10	0	0	0.0	0.0	0.0	10	0	0
AZ	0.0	6.0	6.0	0	0	16	0.0	2.0	2.0	0	0	16
CA	4.0	2.0	6.0	0	0	11	2.0	0.0	2.0	0	0	11
CO	7.0	0.0	7.0	0	0	17	4.0	4.0	8.0	0	0	0
CT	0.0	5.0	5.0	0	10	0	1.0	8.0	9.0	0	10	0
DE	21.0	5.0	26.0	0	0	10	15.0	5.0	20.0	0	0	10
FL	2.0	4.0	6.0	20	0	0	1.0	5.0	6.0	20	0	0
GA	17.0	5.0	22.0	0	0	0	11.0	5.0	16.0	0	0	0
HI	10.0	1.0	11.0	0	15	0	6.0	1.0	7.0	0	15	0
IA	0.0	8.0	8.0	0	13	0	1.0	10.0	11.0	0	13	0
ID	10.0	3.0	13.0	0	12	0	9.0	5.0	14.0	0	12	0
IL	5.0	1.0	6.0	0	0	0	1.0	1.0	2.0	0	0	0
IN	0.0	6.0	6.0	0	0	0	8.0	3.0	11.0	0	11	0
KS	8.0	0.0	8.0	0	0	0	9.0	0.0	9.0	0	0	0
KY	0.0	3.0	3.0	11	0	0	1.0	2.0	3.0	11	0	0
LA	1.0	3.0	4.0	0	0	18	9.0	8.0	17.0	0	0	18
MA	14.0	10.0	24.0	11	0	0	7.0	6.0	13.0	10	0	0
MD	5.0	9.0	14.0	0	12	0	3.0	5.0	8.0	0	12	0
ME	1.0	5.0	6.0	0	0	0	0.0	4.0	4.0	0	19	0
MI	0.0	10.0	10.0	15	0	0	5.0	0.0	5.0	15	0	0
MN	12.0	12.0	24.0	0	0	0	11.0	7.0	18.0	0	0	0
MO	1.0	2.0	3.0	0	0	13	10.0	3.0	13.0	0	0	13
MS	0.0	5.0	5.0	0	15	0	1.0	7.0	8.0	0	15	0
MT	14.0	0.0	14.0	0	11	0	21.0	0.0	21.0	0	0	0
NC	0.0	1.0	1.0	0	14	0	3.0	4.0	7.0	0	14	0
ND	8.0	0.0	8.0	0	0	0	38.0	17.0	55.0	0	0	0
NH	0.0	5.0	5.0	0	0	0	1.0	6.0	7.0	0	0	0
NJ	6.0	3.0	9.0	0	14	0	4.0	2.0	6.0	0	14	0
NM	12.0	0.0	12.0	0	0	0	12.0	12.0	24.0	0	0	0
NV	7.0	7.0	14.0	0	0	17	9.0	2.0	11.0	0	0	17
NY	3.0	2.0	5.0	0	0	0	5.0	2.0	7.0	0	0	0
OH	1.0	5.0	6.0	0	0	0	6.0	4.0	10.0	0	12	0
OK	6.0	7.0	13.0	0	0	17	4.0	9.0	13.0	0	0	17
OR	7.0	0.0	7.0	0	0	15	0.0	9.0	9.0	0	0	16
PA	2.0	6.0	8.0	0	0	0	2.0	3.0	5.0	0	0	0
RI	5.0	3.0	8.0	0	17	0	−1.0	3.0	2.0	0	17	0
SC	2.0	1.0	3.0	0	12	0	5.0	1.0	6.0	0	12	0
SD	0.0	14.0	14.0	12	0	0	0.0	9.0	9.0	12	0	0
TN	0.0	3.0	3.0	0	16	0	3.0	7.0	10.0	0	16	0
TX	14.0	1.0	15.0	0	12	0	18.0	11.0	29.0	0	12	0
UT	0.0	3.0	3.0	0	18	0	0.0	7.0	7.0	0	18	0
VA	0.0	6.0	6.0	0	0	0	2.0	6.0	8.0	0	0	0
VT	0.0	5.0	5.0	0	0	0	0.0	5.0	5.0	0	0	0
WA	6.0	4.0	10.0	0	14	0	5.0	2.0	7.0	0	14	0
WI	2.0	0.0	2.0	0	0	0	1.0	0.0	1.0	0	0	0
WV	6.0	3.0	9.0	0	0	0	15.0	15.0	30.0	0	0	0
WY	1.0	9.0	10.0	0	0	0	1.0	16.0	17.0	0	0	0

On average, in 2001, states with no TEL carried the smallest reserves (8.2 percent of expenditures), followed by states with expenditures TELs (8.3 percent, Alaska excluded; 16.0 percent including Alaska) and states with both revenue and expenditure TELs (9.7 percent) and revenue TELs (9.8 percent). The pattern changes in 2008, when states with revenue TELs had the smallest reserves (6 percent of expenditures). Reserves were 9.2 percent of expenditures in states with expenditure TELs (14.7 percent if Alaska is included); 11.3 percent in states with both tax and expenditure TELs, and 14 percent in states with no TELs.

In summary, TELs research has evolved from case-studies to more sophisticated panel studies and from the use of dichotomous measures to ordinal measures. The importance of the adequacy of state reserves has been the focus of several studies, which find that reserve funding levels are affected by institutional rules that include BSF deposit, limits and withdraw rules, as well as TELs. The latter, however, has not been empirically examined over the past two recessions or with the use of ordinal TEL measures that vary over time.

Research design

Based on the review of literature, our models seek to test three hypotheses:

State revenue limits are negatively associated with state reserves;

State expenditure limits are positively associated with state reserves, and;

State expenditure and revenue limits are negatively associated with state reserves

The variables of interest, the dependent variables, are total state reserves, end of year balances, and BSFs for years 1992 to 2010.² The dependent variables are measured as a percentage of expenditures, which is the generally accepted approach within the literature (Brown, 1989; Douglas and Gaddie, 2002; Hou, 2003; Hou et al., 2003; Maher and Nollenberger, 2009; Poterba, 1994; Sobel and Holcombe, 1996).³ These data were collected from the National Association of State Budget Officers (NASBO) annual reports.⁴ The Breusch and Pegan test rejected pooled regression and given that we are examining the population of states, fixed effects models were run.

Our key independent variables consist of three measures of state TELs:

Revenue TEL (lagged one year)

Expenditure TEL (lagged one year)

Revenue and Expenditure TEL (lagged one year)

The model includes a set of controls to capture economic, fiscal, demographic, and political effects. The economic, fiscal, and demographic measures are:

Per capita income

Percent of population under age 18

Percent of population age 65 or older

Revenue stability

Intergovernmental aids as a percentage of revenues

These economic and demographic data were collected from the US Bureau of Census.

The political measures, collected from the National Council of State Legislatures, are:

Divided government

Government control by Democrats

To capture recessionary effects, year dummies were included for 2001, 2002, 2007–2010.

Per capita income serves as a measure of state fiscal capacity, the expectation being that the higher the state's per capita income, the larger its reserves. Revenue stability Rodriguez-Tejedo (2012) and federal aid as a percentage of operating revenues (Joyce, 2001) have been identified as important variables in earlier work on state reserves. Revenue stability is measured as state sales and personal income tax collections as a percentage of operating revenues. The expectation is that in states with higher revenue volatility and states more reliant on federal aids, fund balances will be larger. Intergovernmental aid has historically been perceived as a measure of fiscal condition; the more dependent an entity on intergovernmental aid, the worse its fiscal condition and by extension, the greater the need for reserves to compensate for potential aid cuts (Maher and Nollenberger, 2009).

We include two political variables in this analysis: divided government measured as one if the legislature and the executive branches of state government are controlled by different parties and zero otherwise. McCubbins (1991) found that at the federal level, divided government resulted in increased deficits. At the state level, Alt and Lowry (1994) found that states with divided government (the legislature and governor representing different parties) were more likely to run deficits than in states with single-party control. Poterba (1994) found that states with single-party control were more inclined to have stricter anti-tax policies and preferred to cut spending in response to fiscal shocks (Poterba, 1994). We can thus infer that in states where there is single-party control, states should have stronger reserves than in states with divided government because in the latter states, elected officials are less willing to make politically risky decisions that could affect fiscal policies, including reserves (Poterba, 1994). The second political variable is coded one if the legislative and executive branches are controlled by the Democratic Party (zero otherwise). Generally, Republicans are expected to be more fiscally conservative than Democrats resulting in the expectation that reserves will be lower when Democrats are in power.

Results

A set of fixed effects models where the states are fixed effects is presented in Table 3. The models offer limited evidence of a relationship between TELs and state reserves. In two models, there a marginally significant relationship between TELs and the dependent variables. Contrary to expectations, expenditure TELs are negatively associated with total state reserves and consistent expectations, revenue and expenditure TELs are negatively associated with total reserves and ending balances. We find negative, but statistically insignificant relationships between BSFs and all three TELs measures. There is also no statistically significant relationship between revenue TELs and state reserves.

Table 3.

Fixed effects models of TELs and state reserves 1992–2010

	Total reserves		BSFs		Ending balance
	As Pct expends		As Pct expends		As Pct expends
	Coeff.	S.E	Coeff.	S.E.	Coeff.	S.E.
Tax and expenditure limitations
Revenue and expenditure	−0.184	0.989*	−0.573	0.059	−0.123	0.075*
Expenditure	−0.158	0.090*	−0.088	0.054	−0.066	0.068
Revenue	0.028	0.142	−0.380	0.085	0.595	0.107
Political make-up
Mixed control	−1.664	0.678**	−1.199	0.404***	−0.421	0.511
Democratic control	−3.256	0.746***	−1.777	0.444***	−1.429	0.562**
Revenue structure
Revenue volatility	0.276	0.044***	0.140	0.026***	0.138	0.033***
Pct. Intergovernmental	−0.424	0.059***	−0.168	0.035***	−0.254	0.044***
Demographic/economic
Pct Pop < 18yrs	0.007	0.020	0.004	0.012	0.001	0.015
Pct Pop>64 yrs	0.877	0.415**	0.788	0.247***	0.119	0.313
Per capital income	0.0004	0.000***	0.0002	0.000***	0.0001	0.000***
Recession year dummies
2001	−1.063	0.903	0.120	0.538	−1.265	0.680*
2002	−3.887	0.936***	−1.166	0.557**	−2.760	0.705***
2007	0.742	1.004	−0.205	0.598	0.988	0.756
2008	−2.533	1.071**	−1.108	0.637*	−1.493	0.806*
2009	−2.865	1.350**	−0.754	0.804	−2.158	1.016**
2010	−2.960	1.099***	−0.904	0.655***	−2.065	0.828**
Constant	−14.393	5.798**	−15.118	3.451***	0.247	4.365
***p<.01 Number of obs	912		912		912
**p<.05 R-sq within	0.1254		0.1891		0.1328
*p<.10 R-sq between	0.0001		0.0033		0.0027
R-sq: overall	0.0353		0.0257		0.0343
F-value	14.55***		12.36***		8.12***

The political makeup of a state, measured by Democratic Party control of the legislative and executive branches, is significantly and negatively associated with total reserves, ending balances and BSFs, as predicted. This suggests that Republican control may increase reserves. Divided government is also negatively associated with both total reserves and BSFs. This suggests that the divided governments may have difficulty agreeing on budgets, specifically on allocating revenues to reserves.

The fiscal and demographic variables are also consistent in each of the six models. Per capita income, the proxy for state fiscal capacity, is positively associated with total reserves, ending balances and BSFs. The state fiscal variables show consistent findings across the models. Unlike Rodriguez-Tejedo (2012), we find a significant positive relationship between revenue volatility and each of our three measures of state reserves. That is states with more volatile revenues carry larger ending balances, BSFs, and total reserves. The proportion of state intergovernmental aid is negatively associated with ending balances, BSFs, and total reserves. The expectation was that states more reliant on federal aids would have larger reserves to prepare withdrawal of such aid. The proportion of population less than 18 years is not associated with state reserves; however, the proportion of the state's population is 64 or older is positively associated BSFs and total reserves. The year dummies show consistent decreases in each of the three measures of reserves in years 2002, 2008, and 2010. Total reserves dropped in 2002 and in years 2008–2010 reflecting the magnitude of the Great Recession on state budgets.

Conclusions

The size of state reserves and the ability of policy makers to rely on those reserves to weather economic shocks has been the focus of several studies (Hou, 2003, 2005, 2006; Joyce, 2001; Pew, 2014). The arguments have been that despite the growth of BSFs following the early 1980s, state reserves remain inadequate to sustain state spending during economic downturns (Pew, 2014). The focus of scholars' attention has been on the role of BSF deposit, withdraw, and cap rules as explanations as to why state reserves are, generally, poorly funded (Pew, 2014).

Less covered in the literature are the effects of other fiscal rules that have been hypothesized to affect state reserves. The current literature suggests that TELs, as institutional constraints, should have a positive effect on state reserves (Hou et al., 2003). Similarly, Wagner and Sobel (2006) have argued that TELs were created as instruments for states to circumvent BSF rules. The evidence presented here suggests that the effects of TELs on state reserves are weak, at best. As hypothesized, in the cases of total reserves and ending balances, we found marginally significant associations with combined revenue and expenditure TELs. Contrary to our hypothesis, expenditure TELs were marginally associated with total reserves. In these cases, the direction is counter to Hou et al.'s (2003) assertion that institutional constraints positively affect state reserves. The practical implications of these findings are that TELs seem to matter much less than political will when it comes to the size of state reserves. This is an important point when you consider how poorly states were prepared for the 2008–2009 recession.

Our findings also provide evidence that fiscal structures and political composition are associated with state reserves. Revenue volatility is positively associated with ending balances, BSFs, and total reserves. This is sound budget practice as variation in revenue generation should affect reserves. An important takeaway is that during the period of this study, were building reserves during “good years” and drawing down reserves during fiscal/economic downturns, that is, states were using reserves in the prescribed manner. In addition, states with more volatility in their revenue structure had larger ending balances, which, again is a sign of sound fiscal policy. Similarly, Democratic-led state governments tend to have lower funded reserves (all types) and that divided governments tend to have lower ending balances.

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Abrams

Dougan

(1986) The effects of constitutional restraints on government spending. Public Choice 49(2): 101–116.

Alt

Lowry

(1994) Divided government and budget deficits: Evidence from the states. American Political Science Review 88(4): 811–828.

Amiel L, Deller SC and Stallmann JI (2009) The Construction of a Tax and Expenditure Limitation Index for the US. Department of Agriculture and Applied Economics Staff Paper No. 536.

Amiel

Deller

Stallmann

(2014) Does the restrictiveness of state tax and expenditure limitations affect state revenues and expenditures? International Journal of Public Administration 37(4): 237–248.

Bae

Gais

(2007) The effects of state-level tax and expenditure limitations on revenues and expenditures. Rockefeller Institute Policy Brief (May 21), New York: State University of New York at Albany.

Bae

Jung

(2011) The effects of tax and expenditure limitation (TEL) stringency on the level of state expenditure and revenue. International Review of Administrative Sciences 77(4): 647–669.

Balk

Calista

(2001) A framework for professional action in government: A facts and values perspective. Public Performance and Management Review 25(1): 88–104.

Blankenau

Skidmore

(2004) School finance litigation, tax and expenditure limitations, and education spending. Contemporary Economic Policy 22(1): 127–143.

Bowler

Donovan

(2004) Evolution in state governance structures: Unintended consequences of state tax and expenditure limitations. Political Research Quarterly 57(2): 189–196.

10.

Bradbury

Mayer

Case

(2001) Property tax limits, local fiscal behavior, and property values: Evidence from Massachusetts under proposition 2½. Journal of Public Economics 80(2): 287–311.

11.

Brennan

Buchanan

(1979) The logic of tax limits: alternative constitutional constraints on the power to tax. National Tax Journal 32(2): 11–22.

12.

Brown

(1989) Ten-point test of financial condition: Toward an easy-to-use assessment tool for small cities. Government Finance Review 9(6): 21–26.

13.

Cutler

Elmendorf

Zeckhauser

(1999) Restraining the Leviathan: Property tax limitation in Massachusetts. Journal of Public Economics 71(3): 313–334.

14.

Dadayan L and Boyd DJ (2014) After four years of uninterrupted growth, tax revenues decline in the first quarter. The Nelson A. Rockefeller Institute of Government. Available at: www.rockinst.org/pdf/government_finance/state_revenue _report/2014-08-14-SRR_96.pdf. (accessed 13 July 2016).

15.

Danziger

Ring

(1982) Fiscal limitations: A selective review of recent research. Public Administration Review 42(1): 47–55.

16.

Deller SC and Stallman JI (2007) Tax and expenditure limitations and economic growths. Marquette Law Review 90(3): 497–554.

17.

Deller SC, Stallmann JI and Amiel L (2012) The impact of state and local tax and expenditure limitations on state economic growth. Growth and Change 43(1): 56–84.

18.

Deller SC, Amiel L, Stallmann JI, et al (2013a) Do tax and expenditure limits hinder the condition of public infrastructure? The case of the Nation's system of bridges. Public Works Management and Policy 18(4): 379–397.

19.

Deller SC, Maher CS, Amiel L, et al (2013b) The restrictiveness of state tax and expenditure limitations and state debt. Public Finance and Management 13(1): 1–25.

20.

Douglas

Gaddie

(2002) State rainy day funds and fiscal crises: Rainy day funds and the 1990–91 recession revisited. Public Budgeting & Finance 22(1): 19–30.

21.

Downes

(1996) An examination of the structure of governance in California school districts before and after proposition 13. Public Choice 86(3–4): 279–307.

22.

Dye

McGuire

McMillen

(2005) Are property tax limitations more binding over time? National Tax Journal 58(2): 215–225.

23.

Dye

McMillen

Merriman

(2006) Illinois' response to rising residential property values: an assessment growth cap in Cook County. National Tax Journal 59(3): 707–716.

24.

Eisenhardt

(1989) Agency theory: An assessment and review. Academy of Management Review 14(1): 57–74.

25.

Fisher

Gade

(1991) Local property tax and expenditure limits. In: McGuire

Therese J

Naimark

Dana Wolfe

(eds) State and Local Finance for the 1990s: A Case Study of Arizona, Tempe, AZ: Arizona State University.

26.

Gold SD (1983) Preparing for the next recession: rainy day funds and other tools for the states. Legislative Fiscal Paper No. 41. National Conference of State Legislatures.

27.

Gold

(1984) Contingency measures and fiscal limitations: The real world significance of some recent state budget innovations. National Tax Journal 37(3): 421–432.

28.

Gold

(1995) The Fiscal Crisis of the States: Lessons for the Future, Washington, D.C: Georgetown University Press.

29.

Government Finance Officers Association (2014) Financial policy examples – General fund reserves. Available at: www.gfoa.org/financial-policy-examples-general-fund-reserves. (accessed 28 August 2014).

30.

Greene

(2008) Econometric Analysis, 6th ed. Upper Saddle River, NJ: Prentice Hall.

31.

Hou

(2003) What stabilizes state general fund expenditures in downturn years – Budget stabilization fund or general fund unreserved undesignated balance? Public Budgeting & Finance 23(3): 64–91.

32.

Hou

(2004) Budget stabilization fund: structural features of the enabling legislation and balance levels. Public Budgeting & Finance 24(3): 38–64.

33.

Hou

(2005) Fiscal reserves and state own-source expenditure in downturn years. Public Finance Review 33(1): 117–144.

34.

Hou

(2006) Budgeting for fiscal stability over the business cycle. Public Administration Review 66(5): 730–741.

35.

Hou

Brewer

(2010) Substitution and supplementation between co-functional policy instruments: Evidence from state budget stabilization practices. Public Administration Review 70(6): 914–924.

36.

Hou

Duncombe

(2008) State saving behavior: Effects of two fiscal and budgetary institutions. Public Budgeting & Finance 28(3): 48–67.

37.

Hou

Moynihan

(2008) The case for countercyclical fiscal capacity. Journal of Public Administration Research & Theory 18(1): 139–159.

38.

Hou

Moynihan

Ingraham

(2003) Capacity, management and performance: exploring the links. The American Review of Public Administration 33(3): 295–315.

39.

Hou

Smith

(2006) A framework for understanding state balanced budget requirement systems: Reexamining distinctive features and an optimal definition. Public Budgeting & Finance 26(3): 22–45.

40.

Howard

(1989) State tax and expenditure limitations: There is no story. Public Budgeting & Finance 9(2): 83–90.

41.

Hur

(2007) Fiscal limits and state fiscal structure: An analysis of state revenue structure and indebtedness. Municipal Finance Journal 28(3): 19–35.

42.

Jacob

Hendrick

(2012) Fiscal health and sustainability concepts and measures. In: Levine

Justice

Scorsone

(eds) Handbook of Local Government Fiscal Health, Burlington, MA: Jones and Bartlett Learning.

43.

Johnson

Kriz

(2005) Fiscal institutions, credit ratings, and borrowing costs. Public Budgeting & Finance 25(1): 84–103.

44.

Joyce

(2001) What's so magical about five percent? A nationwide look at factors that influence the optimal size of state rainy day funds. Public Budgeting & Finance 21(2): 62–87.

45.

Kioko

(2011) Structure of state-level tax and expenditure limits. Public Budgeting & Finance 31(2): 43–78.

46.

Knight

Levinson

(1999) Rainy day funds and state government savings. National Tax Journal 52(3): 459–472.

47.

Kousser

McCubbins

Moule

(2008) For whom the TEL tolls: Can state tax and expenditure limits effectively reduce spending? State Politics & Policy Quarterly 8(4): 331–361.

48.

Kwak

(2014) The dynamics of state fiscal behavior over the business cycle: Are state fiscal polices procyclical? The American Review of Public Administration 44(5): 550–574.

49.

Ladd

Wilson

(1982) Why voters support tax limitations: Evidence from Massachusetts' proposition 2-1/2. National Tax Journal 35(2): 121–148.

50.

Lang

Jian

(2004) Property taxes and property values: Evidence from proposition 2-1/2. Journal of Urban Economics 55(3): 439–457.

51.

Lauth

(2003) Budgeting during a recession phase of the business cycle: The Georgia experience. Public Budgeting & Finance 23(2): 26–38.

52.

Lowery

Sigelman

(1981) Understanding the tax revolt: Eight explanations. The American Political Science Review 75(4): 963–974.

53.

Maher

Deller

Amiel

(2011) Property tax limits and fiscal burdens: The role of organizational structure. Public Administration Quarterly 35(2): 205–240.

54.

Maher

Deller

(2013) Measuring the impacts of TELs on municipal financial condition. In: Levine

Justice

Scorsone

(eds) Handbook of Local Government Fiscal Health, Burlington, MA: Jones and Bartlett.

55.

Maher

Nollenberger

(2009) Revisiting Kenneth Brown's 10-point test. Government Finance Review 25(5): 61–66.

56.

Marlowe

(2011) Beyond 5 percent: Optimal municipal slack resources and credit ratings. Public Budgeting & Finance 31(4): 93–108.

57.

Marlowe

(2013) Fiscal slack, reserves, and rainy day funds. In: Levine

Justice

Scorsone

(eds) Handbook of Local Government Fiscal Health, Burlington, MA: Jones & Bartlett Learning.

58.

McCubbins

(1991) Party governance and U.S. budget deficits: Divided government and fiscal stalemate. In: Alberto

Geoffrey

(eds) Politics and Economics in the Eighties, Chicago: University of Chicago Press.

59.

McCubbins Moule

(2010) Making mountains of debt out of molehills: The pro-cycical implications of tax and expenditure limitations. National Tax Journal 63(3): 603–622.

60.

McGuire

Rueben

(2006) The Colorado revenue limit: The economic effects of TABOR. State Tax Notes 40(6): 459–473.

61.

Metcalf

(1989) Arbitrage and the savings behavior of state governments. Review of Economics & Statistics 72(3): 390–396.

62.

Miller

(2005) The political evolution of principal-agent models. Annual Review of Political Science 8: 203–225.

63.

Mullins

(2004) Tax expenditure limitations and the fiscal response of local governments: Asymnetric intra-local fiscal effects. Public Budgeting & Finance 26(4): 111–147.

64.

Mullins

Wallin

(2004) Tax and expenditure limitations: Introduction and overview. Public Budgeting & Finance 24(4): 2–15.

65.

Navin

(1994) An evaluation of state budget stabilization funds among Midwestern states. Growth & Change 25(4): 445–466.

66.

New MJ (2001) Limiting government through direct democracy. The Cato Institute Policy Analysis 420: 1–17.

67.

Nicholson-Crotty

Theobald

(2011) Claiming credit in the U.S. federal system: Testing a model of competitive federalism. Publius: The Journal of Federalism 41(2): 232–256.

68.

Pew Charitable Trusts (2014) Building state rainy day funds: Policies to harness revenue volatility, stabilize budgets, and strengthen reserves. Available at: www.pewtrusts.org/∼/media/assets/2014/07/sfh_rainy-day-fund-deposit-rules-report_artready_v9.pdf (accessed 13 July 2016).

69.

Poterba

(1994) State responses to fiscal crises: The effects of budgetary institutions and politics. Journal of Political Economy 102(41): 799–821.

70.

Poterba JM and Rueben KS (1999a) State fiscal institutions and the U.S. municipal bond market. In James Poterba and Jurgen von Hagen (Eds.). Fiscal institutions and fiscal performance. Chicago: University of Chicago Press.

71.

Poterba

Rueben

(1999b) Fiscal Rules and State Borrowing Costs: Evidence from California and other States, San Francisco, CA: Public Policy Institute of California.

72.

Poterba

von Hagen

(2008) Fiscal institutions and fiscal performance, Chicago: University of Chicago Press.

73.

Poulson B (2005) Grading the States’ Tax and Expenditure Limits: A Fiscal Discipline Report Card. Washington, DC: Americans for Prosperity Foundation.

74.

Preston

Ichniowski

(1991) A national perspective on the nature and effects of the local property tax revolt: 1976–1987. National Tax Journal 44(2): 123–145.

75.

Primo

(2006) Stop us before we spend again: Institutional constraints on government spending. Economics & Politics 18(3): 269–312.

76.

Rodriguez-Tejedo

(2012) The determinants of the structure of rainy day funds. Public Administration Review 72(3): 376–386.

77.

Rose

(2008) The political manipulation of US state rainy day funds under rules versus discretion. State Politics & Policy Quarterly 8(2): 150–176.

78.

Rose

Smith

(2012) Budget slack, institutions, and transparency. Public Administration Review 72(2): 187–195.

79.

Seljan

(2014) The limits of tax and expenditure limits: TEL implementation as a principal-agent problem. Public Choice 159(3–4): 485–501.

80.

Shadbegian

(1998) Do tax and expenditure limitations affect local government budgets? Evidence from panel data. Public Finance Review 26(2): 218–236.

81.

Shadbegian

(1999) The effect of tax and expenditure limitations on the revenue structure of local governments, 1962–1987. National Tax Journal 52(2): 221–237.

82.

Shadbegian

(2003) Did the property tax revolt affect local public education? Evidence from panel data. Public Finance Review 31(1): 91–21.

83.

Skidmore

(1999) Tax and expenditure limitations and the fiscal relationships between state and local governments. Public Choice 99(1–2): 77–102.

84.

Skidmore M, Ballard CL and Hodge TR (2010) Property value assessment growth limits and redistribution of property tax payments: Evidence from Michigan. National Tax Journal 63(3): 509–538.

85.

Sobel

Holcombe

(1996) The impact of state rainy day funds in easing state fiscal crises during the 1990–1991 recession. Public Budgeting & Finance 16(3): 28–48.

86.

Springer

Lusby

Leatherman

(2009) An evaluation of alternative tax and expenditure limitation policies on Kansas local governments. Public Budgeting & Finance 29(2): 48–70.

87.

Stallmann JI, Deller SC, Amiel L, et al (2012) Tax and expenditure limitations and state credit ratings. Public Finance Review 40(5): 643–669.

88.

Vasche

Williams

(1987) Optimal government budgeting contingency reserve funds. Public Budgeting & Finance 7(1): 66–82.

89.

Von Hagen

(2002) Fiscal rules, fiscal institutions, and fiscal performance. Economic and Social Review 33(3): 263–284.

90.

Von Hagen J and Harden I (1996) Budget processes and commitment to fiscal discipline. IMF working paper, No. 96/78.

91.

Wagner

Elder

(2005) The role of budget stabilization funds in smoothing government expenditures over the business cycle. Public Finance Review 33(4): 439–465.

92.

Wagner

Sobel

(2006) State budget stabilization fund adoption: Preparing for the next recession or circumventing fiscal constraints? Public Choice 126(1–2): 177–199.

93.

Wasi N and White MJ (2005) Property tax limitations and mobility: Lock-in effect of California's proposition 13. In: Gary Band Janet RP (eds) Brookings-Wharton Papers on Urban Affairs. Issue 6. Brookings Institution, Washington, DC, pp.59–97.

94.

Zahradnik B and Johnson N (2002) State rainy day funds: What to do when it rains? Center on Budget and Policy Priorities. Washington, DC. Available at: www.cbpp.org/archives/1-31-02sfp2.htm (accessed 13 July 2016).