<<  <  >  >>

and revenue departments. Each of the 46 gov-

substantial severance taxes from companies

ernors graded in the report has been in office

that extract oil and minerals. The burden of

long enough to propose at least one budget.

those taxes falls on out-of-state residents to

some extent. Furthermore, the fiscal condi-

Grading Procedure

tion of those states can improve or deteriorate

The report card consists of 23 policy vari-

dramatically in response to changes in the

ables: 8 for spending (2 of which have a

market price of commodities. Severance taxes

weight of only one-half), 9 for revenue (2 of

are a large distortion for Alaska, so its gover-

which have a weight of only one-half), and 6

nor has been excluded from this study.

for tax rates (1 of which has a weight of only

In recent years, many states have moved to

one-half). However, the scores of the gover-

reduce reliance on local property taxes as part of

nors who took office in 2005 do not include

school finance overhauls. In 1994 Michigan

two of the spending variables and two of the

passed an education finance package that

revenue variables that are based on Census

increased the state sales tax in exchange for a

Bureau data. Those data are published with a

larger dollar reduction in local property taxes.

two-year lag, making it impossible to include

Since 1994 other states have followed Michi-

them in the midterm grade calculation of

gan's lead, including Idaho, Kansas, South

new governors.

Carolina, South Dakota, Vermont, Michigan,

For each variable, we use a procedure to

Texas, Florida, and Wisconsin. In most cases,

standardize the results, such that the gover-

the changes involve a reduction in local proper-

nor with the worst score receives a zero and

ty taxes, with the state government compensat-

the governor with the best score a 100. An

ing local governments by increasing the state

equal weight is assigned to each variable

share of school funding. For the purposes of

(with the exception the variables that only

this report card, such reforms create a signifi-

have a weight of one-half), and the scores are

cant challenge. The data on state finances

averaged to obtain an overall grade for each

reflect the increased state spending and revenue

governor.

but do not reflect the reductions at the local

level. Yet because local property taxes were sub-

Policy Variable Details

stantially cut, the combined state and local bur-

To make meaningful comparisons between

den has not risen in some cases. For states that

the states, the analysis controls for differences

have implemented such school finance over-

in the sizes of state populations and economies

hauls, adjustments have been made to the

by expressing spending and tax revenue data

spending and tax variables so that governors are

for each state as a ratio of either each state's

not penalized for an increase in state-level

population or personal income. Most of the

spending when the spending was designed to

revenue and spending variables are expressed

compensate localities for a local tax cut.

in that way (i.e., per capita or as a percentage of

personal income). All variables measure state-

Appendix B:

level tax and spending, and thus the report

Report Card Methodology

does not include the fiscal activities of local

governments. All variables are measured for

only the years of each governor's current term

This study computes a fiscal policy grade

in office.

for each governor reflecting his or her relative

success at restraining the growth of taxes and

Expenditure Variables

spending. All of the tax and spending data

1. Average annual change in real per capi-

used in the study come from the U.S. Bureau of

ta spending through FY04 (measured

the Census, the National Association of State

only for the governors in office before

Budget Officers, the National Conference of

2004).

State Legislatures, and individual state budget

9