The Financial Health and Fiscal Capacity of Municipal Governments in Michigan

by Steven B. Duprey and Lynn R. Harvey (1)

Introduction

Public Act 342, 1996 amended the state revenue sharing program by changing the formula used to distribute the statutory portion of state shared revenues and changing the source of the funds for the revenue sharing pool. A joint House and Senate Task Force was appointed to examine the issue and develop a set of recommendations to be presented to the legislature by September 1997. The findings of the Task Force were to include a new plan or formula for the distribution of state revenue sharing funds to locals. While the Task Force has yet to develop a set of recommendations to present to their colleagues, discussions have commenced as to the objectives of state revenue sharing, the methodology to be used in distributing the funds, and the distributional consequences of alternative formulas.

The $1.3 billion transferred to counties, cities, villages and townships through the state revenue sharing program in FY 1997 has varying impacts by type of governmental entity. For some local governments (townships), state revenue sharing is the single largest revenue source for general operation. While in other cases, such as counties, state revenue sharing represents a small percentage of general fund revenues. Any change to the method of distributing state revenue sharing creates winners and losers both by type of government and within each type of government. Therefore, examining the fiscal capacity and financial health of counties and municipalities provides basic information to policy makers that can be used in attempting to quantify the distributional consequences of any policy change to the state revenue sharing program.

A study was undertaken to construct financial profiles for a sample group of cities, villages, and townships, to ascertain the importance of state revenue sharing in supporting general governmental operation, and to estimate the fiscal capacity and financial health of each of the types of local governments.

Methodology

Data for all municipalities within twenty four Michigan counties (2) was assembled from the FY 1996 "F65 Local Unit Fiscal Report", an annual report filed by local governmental entities with the Michigan Department of Treasury. The purposive sample included 657 municipalities (cities, villages and townships), however, the financial reports from 32 municipalities were incomplete leaving a net sample size of 625 - Table 1. Financial profiles were not constructed for county governments since the policy discussion thus far has not proposed altering the method of distributing state revenue sharing funds to counties. Currently, counties receive their state revenue sharing allotment on a per capita basis.

Data elements from the "F65 Reports" were loaded to a data base by staff from Michigan State University, Michigan Municipal League, and Plante & Moran, LLP. The discussions and conclusions in the report are the responsibility of the authors.

Table 1A: Municipal Sample Profile

** Village population not included in total to avoid double counting

The 156 cities included in sample represent 57.14 percent of the total number of cities in Michigan. The 84 villages represent 32.19 percent of total number of home rule and general law villages in the state. The 385 sample townships represent 31.00 percent of the total number of townships organized in the state. Overall, the 625 municipalities represent 32.15 percent of all Michigan municipalities. While the selected sample was purposive by design as opposed to a random sample, the authors expect the results to fairly represent the financial conditions and fiscal capacity of all Michigan municipalities.

The total population of the municipalities included in the sample is 7,087,468 representing 77.17 percent of the state's total population. The sample cities account for 65.04 percent of the population of the municipalities in the sample. Since Detroit is the largest city within sample, deleting Detroit from the sample, changes the summary comparisons, therefore comparison tables will provide both a "with" and "without" Detroit analysis - Table 1B.

Table 1B: Municipal Sample Profile Without Detroit

** Village population not included in total to avoid double counting

General Fund Revenue and Expenditures - Summary

The 625 municipalities had combined general fund revenues in FY 1996 of $3.785 billion and GF expenditures of $3.698 billion. Total state revenue sharing payments to the 625 units in FY 1996 totaled $899.2 million or 23.75 percent of total GF revenue. Table 2A summarizes the breakdown of GF data for the sample municipalities. Table 2B provides a similar summary without Detroit included in sample. Excluding Detroit raises the state revenue sharing as a percent of total GF revenue slightly for cities (22.15 percent compared to 21.86 percent) and raises the overall average for sample municipalities from 23.75 percent to 25.17 percent. For the sample municipalities, Detroit's FY 1996 state revenue sharing payment represented 44.01 percent of state revenue payments.

Table 2A: GF Revenue and Expenditure Summary - Sample

Table 2B: GF Revenue and Expenditure Summary Without Detroit

Tables 3A&B, 4 and 5 provide a breakdown of GF revenue, expenditure, and state revenue sharing payment by type of local government sorted by taxable value per capita. Table 3B displays the summary of the sample cities less Detroit. Since Detroit's taxable value per capita is $5,980, the adjustment in Table 3B is made to the <9,999 group of cities.

Table 3A: GF Revenue/Expenditure - Cities

Table 3B: GF Revenue/Expenditure - Cities

Table 4: GF Revenue/Expenditure - Villages

Table 5: GF Revenue/Expenditure - Townships

Table 6: State Revenue As Percent of Total GF

Table 6 provides range data of state revenue as a percent of total GF revenue by type of government. All the types exhibit significant range differences with township governments having the largest range in percentages, 4.60 to 79.52, and cities the smallest range difference.

Separating the sample municipalities by government type and taxable value per capita (Tables 3-5) results in an identifiable pattern of the importance of state revenue sharing patterns. The lower the taxable value per capita the higher the percentage that state revenue sharing payments comprise of total general fund revenues. This pattern is consistent with all three types of municipalities. For example, although state revenue payment represent an average of 21.86 percent of total GF for cities, the range of taxable value groups is from a high of 25.56 percent for the $10-14,999 group to a low of 10.90 percent for cities with per capita taxable value greater than $50,000.

State revenue sharing payments to townships comprise the largest share (59.33%) of general fund revenue for township with less than $9,999 taxable value per capita with an overall sample average of 37.72 percent. The sample villages reliance of state revenue sharing payments as a percent of total GF revenue ranged from a high of 38.05 percent to a low of 5.70 percent with an average of 20.96 percent.

One explanation of the inverse relationship between taxable value and the dependence a municipality displays on state revenue sharing for general fund operation relates to the relative tax effort formula for distribution. Units with low taxable values must exert a greater taxing effort to generate sufficient revenues to support community services. Since the "relative tax effort" formula provides additional monies to units with above average taxing effort, state revenue payments will constitute a larger portion of general fund revenues.

Millage Rates of Sample Municipalities

Comparing millage levies of the municipalities in the sample sorted by taxable value supports the discussion of the relationship between low taxable value and the importance of state revenue sharing to general fund operation. Tables 7-9 display the data by type of local government.

For sample cities the low taxable value units (n=20), average millage rate is 22.24 mills and high taxable value group (n=4), levies on average 12.41 mills. The millage range for the cities is from a high of 34.74 mills (Ecorse) to a low of 5.96 mills (Kentwood). For villages, the range in millage levies is from a high of 24.36 mills (Capac) to a low of 1.0 mills (Shoreham). The sample townships also exhibit a significant range difference in millage levies. Two townships in the sample levy zero mills (Salem in Washtenaw Co. and Wells in Delta Co.). Redford Township levies the highest millage in township sample at 15.38 mills.

Table 7: Millage Rates - Sample Cities

Table 8: Millage Rates - Sample Villages

Table 9: Millage Rates - Sample Townships

Data Source: 1996 Ad Valorem Property Tax Levy Report, Depart of Treasury

Financial Health of Municipalities

The previous discussion has focused on fiscal capacity, that is, the ability or capacity of a local unit to generate general fund revenues given their property tax base as measured by taxable value and millage effort. The financial health of municipalities is also an important criteria when considering policy changes to the sharing of state funds. One measure of financial health is termed fund equity. Fund equity is determined by dividing "general fund balance" by "total general fund expenditures". The percent fund equity indicates the accumulated general fund reserves of a unit. Municipalities attempt to maintain sufficient reserves to serve as a financial safety net in the event that unforseen expenditures occur during the fiscal year. It is advisable that units maintain between 14-16 percent fund equity. As a matter of past practice, particularly townships, some jurisdictions maintain higher fund reserves compared to other jurisdictions. In part, the larger fund reserves may be attributed to accounting practices. For example, a unit may desire to accumulate sufficient reserves to fund road maintenance, buy a fire truck or engage in a capital construction project. Instead of transferring monies from the general fund to a capital project or public improvement fund, the units accounts for the accumulated savings in the general fund thus giving rise to a larger fund reserve. Units may carry additional fund reserves in other accounts such as, special revenue fund, budget stabilization fund, or public improvement fund. No attempt was made to assess fund reserves other than those posted to the general fund.

Units may as a matter of philosophy elect to fund capital project with accumulated reserves instead of borrowing the necessary capital leading to a large fund equity. How large fund reserves should be is a political question determined by local governing bodies, however, it should be pointed out that local units are not banks. If large fund balances in relation to annual GF expenditures demands cannot be justified, then a unit as a matter of principle could ro should lower millage rates and pass the savings along to taxpayers in their jurisdiction. As will be seen in Tables 10-13, some units have accumulated sizable fund reserves with townships having on average larger fund balances compared to cities and villages. The average fund equity for the 156 sample cities for FY 1996 was 16.89 percent, for the 84 villages 44.04 percent, and the 385 townships 59.37 percent. For cities (Table 10), as taxable value per capita rises, percent fund equity rises, state revenue sharing as a percent of total GF declines and millage rates decline.

Table 10: Fund Equity - Sample Cities

The relationship of fund equity, taxable value and millage rates exhibited by cities does not emerge with villages and townships. For example, townships (Table 12) with taxable value of less than $9,999 recorded fund equity percentage of 82.68 with state shared revenues comprising 59.33 percent of budget, and highest millage rate average of the sample group. On the opposite extreme, the 38 townships with taxable value per capita exceeding $50,000 also has an average fund equity of 82.33 percent lower average millage rate and state revenue sharing represented 28.58 percent of GF revenues.

Table 11: Fund Equity - Sample Villages

Table 12: Fund Equity - Sample Townships

Table 13 provides insight to the distribution of percent fund equity across the units by fund equity ranges. Each governmental type had one unit in the sample with negative fund equity, while 2 villages and 12 townships had fund equity for FY 1996 exceeding 300 percent.

Sixty five percent of the cities, 22.14 percent of the villages and 47.01 percent of the townships had fund equity of 30 percent or less. On the opposite end of the fund equity scale, 2.56 percent of the cities, 15.48 percent of the villages and 34.54 percent of the townships had fund equity exceeding 100 percent.

Table 13: Fund Equity By Range

State Revenue Sharing Policy Discussion

The method used to distribute the statutory portion of state revenue sharing to municipalities is subject to much discussion since winners and losers are created with any change in the distribution formula. Arguments have been advanced that using the relative tax effort as a basis is a perverse goal since implicit in the method is rewarding local jurisdictions with above average taxing efforts. Countering this argument is the relative tax effort method recognizes and rewards local units who utilize local taxing effort to support local services. Units with high taxing efforts, principally cities with aging infrastructure and in some cases declining populations, experience significant revenue loss if formula distribution methods are altered from the relative tax effort basis without recognition of fiscal need. To assume that communities with declining or slow population growth, mainly older cities, have declining service costs is an errant assumption since fixed costs do not change but per capita costs of community services rise.

If distribution of state revenues to locals shifts to a per capita basis or growth in population basis, jurisdictions with rapidly expanding populations and service needs are rewarded. However it can be countered that jurisdictions with declining populations or below state average rates of growth have sunk or fixed costs thus the tax price of services rises with population declines. If a corresponding decline in tax base occurs, these jurisdictions experience a "double whammy" not only do they lose state revenue sharing if a per capita distribution system is implemented but lose property tax revenues at the very time increased revenues are demanded.

Additionally, it can be argued that units with expanding populations while experiencing a corresponding increase in service demands that translate into additional costs of the unit are also collecting additional revenues from increases in the tax base and associated service fees or charges thus not only would yield more local revenue but additional revenue from state shared sources. The principal of tax substitution apply to this situation since state shared revenues are passed on to local units as a lump sum grant which translates into the local units being able to substitute state shared revenues for local taxing effort.

An often over looked implication of changing from a relative tax effort to a per capita basis of sharing state revenues is the issue surrounding land use, land fragmentation and implicit state policy impacting both suburbanization and the emergence of brownfields in central cities. State policies that reward units with increasing populations and corresponding suburbanization efforts is in conflict with stated goals of preserving farmland and lessening the impact of land fragmentation. Land fragmentation not only impacts the agriculture industry but increases services costs in the recipient units, mainly townships. Therefore state policies and the structure of programs such as state revenue sharing have a direct impact on location decisions, suburbanization, tax prices of governmental services and hastening exit from central cities.

While it is simplistic to assume that state revenue sharing is the driving force behind suburbanization since the issue is much more complex and is intertwined with social and political decisions, the fact remains that the state does indeed influence location decisions both of households and firms by the structure of state policies. Therefore the first item of business it would appear would be for the State Revenue Sharing Task Force to establish the objectives of state revenue sharing. Should the program be redistributive in nature or return monies to local in proportion to the contribution to the state from the local units? Should the state revenue sharing program reward above average population growth units or recognize units with declining fiscal capacity? Should state revenues be shared on the basis of financial need? Should state revenue sharing policy reward jurisdictions with above average taxing efforts or penalize jurisdictions with high taxing efforts? Should state revenue sharing policies reward jurisdictions engaged in land fragmentation or reward units attempting to control growth to minimize land fragmentation?

Any change in the state revenue sharing policy impacts local jurisdictions. Therefore it seems critical that fiscal capacity, taxing effort, and population be included in a distribution formula. Relying only upon one distribution factor fails to capture the interdependent nature of local governmental finances. Including fiscal capacity (taxing effort and taxable value), as a factor in the state revenue sharing formula, partially accounts for the variation that is present among the municipalities in the state. If taxable value and population are the only factors, a major determinant of fiscal capacity is missing that being local effort. Per haps using taxable value, population and capping taxing effort at some level depending on type of municipality would offer an alternative for the formula. Since county government provides services to all residents in the county, it can be argued that distributing state revenue sharing on a per capita basis is appropriate since all residents have access to the services, to some degree.

Data has been advanced in this study that points out the variance in financial health of local units of government. Townships tend to carry higher fund balances as compared to cities and appear, at least from sample data, to have a significantly higher percentage of units with fund equity greater than 100 percent. The financial health of villages is somewhat between the cities and townships. The difficulty of designing a state revenue sharing policy that incorporates the various economic condition factors is almost overwhelming. However, the research does point out the necessity to consider more than just population in the distribution formula.

(1) Research Associate, Department of Agricultural Economics; Professor and Extension Specialist-State and Local Government, Department of Agricultural Economics, Michigan State University, E. Lansing, MI 48824. March 1998

(2) Counties included in sample: Bay, Berrien, Calhoun, Delta, Genesee, Huron, Ingham, Iron, Jackson, Kalamazoo, Kent, Leelanau, Livingston, Macomb, Marquette, Monroe, Muskegon, Oakland, Otsego, Ottawa, Saginaw, St. Clair, Washtenaw, and Wayne.