After we released the first Fiscal Situation of our Municipalities story, we received a couple inquiries on the data for the cost drivers. The Municipal Cost refers to the amount that each municipality must spend in order to provide a common level of non-school services such as administration, public works, and public safety. The New England Public Policy Center estimated municipal cost by using a statistical regression analysis to identify cost factors strongly related to non-school per capita spending levels.

The identified cost drivers include:

  • Population Density;
  • Unemployment Rate;
  • Total Jobs;
  • Private Sector Wage Index; and
  • Road Mileage

It is important to note, that these costs are not actual costs which are subject to the choices and actions of local officials.

Each cost driver has a different level of magnitude depending on the town. In the tables below, we show the share of each cost driver in Table 1 and in Table 2 we provide the raw data (values represent the coefficients from the regression analysis). We also added in bar charts for each of the Council of Governments as a visual to show the share of costs for each town.


Table 1: Municipal Cost Drivers as Share of Total Cost


Table 2: Municipal Cost Drivers

Private Sector Wage Index the main cost driver for all towns in Connecticut


Chart 1: Municipal Cost Drivers as Share of Total Cost for each Council of Government

It is important to note, that these costs are not actual costs which are subject to the choices and actions of local officials. The statistical analysis allows the researchers to hold the following factors constant across communities: economic resources (for example, ENGL and income), factors that may affect preferences (for example, demographic characteristics, political makeup of the town), factors that may affect operating efficiency (for example, form of government), and each town’s public safety arrangement (resident state trooper versus state troopers, and paid versus volunteer firefighters).

The NEPPC looked at poverty rate, population size, percentage of the population that is foreign born, and percentage of housing units that are older rental units as potential cost factors but found they were not statistically significant. Results of their statistical analysis can be found in the Appendix of the report.