RESULTS - GENERAL FUND

    A total of 160 regressions were run. The results will be presented first in general groups then in a department-by-department basis. The regressions use budgeted amounts as the dependent variable and population as the independent variable. When further analysis is required, some additional regressions were run which use independent variables other than population. An example of this is equalization in which number of parcels is a better estimate of workload than population is. The following pages detail the regressions for most county departments and offer a brief explanation of the results.

General Fund Budget

    The size of the general fund in Ingham County is quite comparable to the size of other counties’ general funds. On a per resident basis, Ingham is average. Other counties collect and allocate as little as $153 per resident (Genesee, which is 18 percent less than expected) to as much as $214 (Washtenaw, which is 19 percent greater than expected). The R-Squared shows that the size of a county’s general fund is directly linked to population. As a general point, larger counties tend to have a smaller general fund budget when examined on a per resident basis. This trend will be true for many budgetary characteristics as well. There are two factors which contribute to this. The first is economies of scale - the marginal cost per resident of supplying services generally diminishes as the size of the government increases. The second is that larger counties tend to rely more heavily on computers and automated systems, which decreases the overall costs of personnel.

    After an examination of the general fund, departmental budgets are separated into three groups which reflect the liaison committees of the Ingham County Board of Commissioners. This grouping is for the purpose of simplifying the presentation of the analysis. The departments which are included in the tables are not exactly the same as they appear in the 1997 budget. Only departments which are common among at least ten counties are included. This assumes that if at least ten counties have a particular department, then the remaining counties perform the same functions through different departments.  Ingham County departments which are not common among at least ten counties include all advisory boards, community agency grants, housing commission, conservation district, road commission, library, medical care facility, transportation millage, women's commission, and adolescent diversion.

    The following tables are abbreviated summaries of the results of the regression analyses.  Each summary includes a graph of the results for all counties, a summary table for Ingham County, and a brief explanation of the findings as they related to the specific department. Keep in mind that the difference between the actual budgeted amount and the expected amount could be due to the specific policy objectives of Ingham County. Also, the expected amount could be largely determined by the policy objectives of other counties. One should not infer that the expected amount is a value that the county should strive to reach. It is merely a guideline with which to compare Ingham County with other counties. However, a significant variance is evidence that differences exist in funding priorities or departmental structure. Due to the methodology used, a negative variance indicates that the expected budget exceeds the total budget and a positive variance indicates that the total budget exceeds the expected budget.

    A general trend of this comparison is that Ingham’s departments tend to receive slightly higher funding levels on a per capita basis. This is due in part to the methodology used - total budget is the unit of comparison instead of general fund budget. Grants, intergovernmental transfers, and special millages are included in this. Although Ingham’s total general fund is well within its expected range, the total amount used for this comparison exceeds its expected amount by 16.59 percent. Although this effect can not be corrected, the analysis will highlight the departments in which this effect is most apparent.

    The results show simply what the variation is between the amounts and whether the results are statistically significant - i.e., probably not due to chance. Significant differences could be due in part to efficiency, departmental structure, county priorities, or departmental responsibilities. The nature of this analysis is such the outcomes can not differentiate between these four possibilities. Those departments which are either higher or lower than the average were contacted for information on the differences in departmental responsibilities and structure among the counties. This information is included in the discussion of the departments.

    The R-Squared value appears for each department. If the R-Squared value is less than .7000 then the dependent variable is probably not related to population. This shows that the dependent variable is not contingent on the population characteristics of the counties but is instead assumed to be contingent entirely on the differences that exist in the preferences and objectives among the commissioners and administrations of all counties. In these cases, the expected amounts serve only a theoretical purpose. The budget per resident relationship will be the factor for developing conclusions in these cases.

    The R-Squared value and variance will be combined with an assumption variable to determine whether statistical significance exists. The assumption variable incorporates a hypothesized natural variance of ten percent for each department. In other words, exogenous factors are assumed to naturally influence a department’s budget by as much as ten percent in either direction. (This 10 percent assumption is the default estimate of the standard deviation which is used in many statistical packages. Using this estimate, one standard deviation is assumed toe be equal to 10 percent of the expected value divided by the R-Squared value.) Due to this methodology, 70 percent of departments will naturally fall within the expected range of one standard deviation while 15 percent will naturally be greater than expected and 15 percent will naturally be less than expected.

    The results of the regressions will be presented in a department-by-department basis, sorted according to the responsibilities of three liaison committees of the Board of Commissioners - the Administrative Services and Personnel Committee, which includes central services as well as some statutory services; the Human Services Committee, which includes departments that assist individuals in meeting basic needs; and, the Law and Courts Committee. Each department will include a graph to show the similarities and differences in funding among the counties as well as a chart which sums up the results of the regression as they relate to Ingham County. A brief discussion of each department and any information revealed by the regression will follow.