Edward J. Jepson, Jr.
Department of Urban and Regional Planning
University of Tennessee
Edward J. Jepson, Jr. is an assistant professor at the Department of Urban and Regional Planning, University of Tennessee. He received his Ph.D. in urban and regional planning from the University of Wisconsin-Madison.
January 31, 2003
EXECUTIVE SUMMARY
In this report, seven primary spending categories are identified and a comparison provided between Tennessee and the other states as to how well it balances (a) how much it spends, on the one hand and (b) how much it should be spending, on the other. Such an approach attempts to address not whether spending is low or high compared to other states but whether spending is low or high relative to its needs. For example, just because Tennessee is spending less on Education than other states doesn=t necessarily mean that it should be spending more. The low spending may be coincidental with satisfactory educational performance due to various intrinsic characteristics that exist in the state (eg., better teachers or brighter students).
For each of seven spending categories, Aneed@ is defined in terms of a number of measurable characteristics. The level of these characteristics reflects either a condition of high need or low need. For example, for Education spending, one of the measured characteristics is graduation rate; the assumption would be that a state=s system of schools with a high graduation rate would not be as Aneedy@ as one with a low graduation rate. This analysis would then look for high spending in the Aneedy@ state relative to other states. If it isn=t found, a condition of Anegative disconnect@ is identified, that is, there is a discrepancy between how much the state is spending on education and how much it should be spending.
Under the criteria of need that are utilized in this report, Tennessee ranks highest among the fifty states; that is, Tennessee is most in need of effective programs and funding to address the seven policy areas that are considered. The states with the highest need to develop effective policies and programs are identified below:
State Needs ranking
Tennessee
1
Florida 2
Michigan
Illinois
Washington 5
On the basis of 1999 budgetary data, it can be concluded that Tennessee is underspending relative to its needs across all of the seven categories of spending. When the level of disconnect is calculated for all fifty states, each state can be ranked from 1-50, with 1 being a high level of negative disconnect (low spending relative to estimated spending needs) and 50 being a high level of positive disconnect (high spending relative to estimated spending needs). The list below shows how Tennessee ranks:
Spending category Rank
Natural resources 4
Police protection 9
Education 11
Health & hospitals 13
Parks & recreation 16
Highways 18
Public welfare 20
TOTAL 4
The ATotal@ indicates that Tennessee ranked 4th in terms of how much its spending is low relative to its estimated spending needs in all seven of the categories combined. This indicates a high total Anegative disconnect@ that is only exceeded by the states of California, Texas, and Florida.
Also included in the report is information about how much Tennessee is spending relative to how much its residents can afford to spend, calculated as a function of median household income. It was found that spending in these seven program areas as a percentage of median household income is about in the middle when compared with the other forty-nine states. However, it is also noted that government spending overall and taxes in Tennessee are low in comparison to other states.
This indicates that Tennessee has the capacity to increase its spending to correct the disconnects that are reported here. Whether it will actually do so is another story. If ways cannot be found to significantly change Tennessee’s approach to spending and/or taxes so that adequate revenues can be collected from the existing base of wealth, an alternative strategy would be to identify and enact policies that have the most chance of increasing the state=s median household income. This would have the effect of simultaneously increasing spending capacity (by generating more tax revenues) while also potentially decreasing spending need (in such areas as welfare, crime, and health care). Statistical analysis indicates a strong connection between education and income: as one goes up, so does the other. This relationship may point to one possible strategic focus, i.e., having improvement in our state’s educational system – in terms of test scores, graduation rates, and college attendance - as a main policy priority.
Table of Contents
Introduction....................................................................................................................................... 1
Methodology..................................................................................................................................... 1
Findings............................................................................................................................................. 4
Summary and Discussion................................................................................................................. 10
Attachment A - State rankings by total disconnect............................................................................ 13
Attachment B - State rankings in terms of spending........................................................................... 15
Attachment C - Comparative rankings of total spending need............................................................ 17
Attachment D - Multiple regression model........................................................................................ 19
Attachment E- Comparison of total state spending as a % of private wealth……………………20
Notes.. 22
Introduction
The purpose of this analysis is to compare the extent to which the categorical spending of states in the U.S. is Adisconnected@, in the sense that they are spending too little or too much in relation to their spending needs. A Adisconnect@ is identified when the need for a particular type of spending is low and the actual level of spending is high (a positive disconnect), or vice versa (a negative disconnect). The numerical value of a disconnect is the ranking of the state in terms of spending (high, if spending is high) minus the ranking of the state in terms of need (high, if need is low). Therefore, for a state that is ranked 28th in terms of spending (out of the fifty states) and 12th in terms of spending need, the amount of the disconnect will be reported as 16 (28 minus12), which is a positive disconnect. Alternately, a state with a spending ranking of 12 (low spending) and a spending need ranking of 49 (high need) will have a disconnect of -37, indicating a negative disconnect. Generally, a positive disconnect is preferred to a negative disconnect, in that the latter indicates that it may be that needs are not being met, whereas the former is most likely a reflection of strategic preference.
Methodology
The following dimensions of 1999 state spending were analyzed[i]:
* Total
* Education
* Public welfare
* Health and hospitals
* Highways
* Police protection
* Natural resources
* Parks and recreation
* Governmental administration
* Intergovernmental transfers
The fifty states were then sorted in increasing order of per capita spending, i.e., lowest to highest.
An estimate of spending need that corresponded to each spending category except Governmental Administration and Intergovernmental Transfers was then calculated by (a) dividing the average value for all fifty states into the value for each specific state for those characteristics that are high relative to need (for example, a high poverty rate is an indicator of more need for public welfare spending) and (b) dividing a specific state value into the value for all fifty states for those characteristics that are low relative to need (a high median household income is an indicator of less need for public welfare spending). If a state=s cumulative value (the value of all measured
characteristics combined) was high relative to the average for all states, this calculation would return an indication of low need. To clarify this calculation process, the following example is provided. Table 1 lists two characteristics, median household income and % below the poverty level. Their conversion into indicators is different due to the fact that a high value for the income
characteristic indicates a condition of low need, whereas a high value for the poverty characteristic indicates a condition of high need.
Table 1 - Indicator calculation example
_____________________________________________________________________________
Median % below
Household poverty Total
State Income indicator level indicator indicator
Alabama $36,213 .91* 15.1 .84** 1.75
Alaska $51,509 1.30 7.6 1.66 2.96
Arizona $37,119 .94 12.0 1.05 1.99
Arkansas $29,762 .75 14.7 .86 1.61
California $43,744 1.10 13.8 .92 2.02
Average $39,669 12.6
* $36,213 divided by $39,213=.91
** 12.6 divided by 15.1=.84
_____________________________________________________________________________Notice in the example that in the case of median household income, the highest value ($51,509) returned the highest indicator (1.30), which corresponds to the high value calculation under (a), above, whereas the opposite occurred for % below poverty level, where the highest value (15.1) returned the lowest indicator (.84) and which corresponds to the low value calculation under (b), above. When the indicators for these two characteristics are combined, Alaska receives the highest total indicator; since high value equals low spending need, Alaska is correctly portrayed due to its high median household income and its low poverty level. These five states would be assigned the following rankings:
1. Alaska
2. California
3. Arizona
4. Alabama
5. Arkansas
The calculation of spending need is not a simple matter. It requires that characteristics be selected that are Agood@ indicators. Since it is a matter of judgement, the quality of an indicator as being a good reflection of spending need is open to question. To some extent, this analysis is data-driven, that is, the characteristics are the ones that were most readily-available, in addition to being apparently valid. As can be seen below, there was significant variation in the number of characteristics that defined spending need across the seven spending categories.
Spending category Spending need characteristics
Education: % high school graduates, graduation rates, post-secondary public school enrollment, SAT[ii] and ACT scores[iii]
Public welfare: Median household income, % below poverty, % unemployed
Health and hospitals: Cancer cases, % of low birth weights, death
and infant mortality rates
Highway (capital spending): State highway miles, urban congestion, narrow roadway miles, deficient bridges, number of registered vehicles[iv]
Police protection: Overall crime rate, violent crime rate, change in
overall crime rate 1997-99
Natural resources: Toxic releases, forest development pressure[v], number of threatened & endangered species[vi], species biodiversity, plant diversity[vii], air pollution emissions, surface water pollution[viii]
Parks and recreation: State park acreage, visitors per acre
Governmental
administration: None
Intergovernmental
transfers: None
For each of the spending categories, the fifty states were ranked from lowest to highest per capita spending and from lowest to highest estimate of spending need. In the case of all except Governmental Administration and Intergovernmental Transfers, the value of the second ranking was then subtracted from the value of the first ranking, providing a measure of Adisconnect@ for each state. These were then ranked in ascending order so that the states with the largest negative disconnect would be ranked highest and those with the largest positive disconnect would be ranked lowest. Thus, if a state had a disconnect ranking of 1, it was spending the least amount of money compared to other states relative to its need for that type of spending; alternately, if a state had a disconnect ranking of 50, it was spending the most amount of money.
Findings
In the case of per capita state spending and spending for Governmental Administration and Intergovernmental Transfers, the five states that were ranked lowest and highest are reported. For each of the seven spending categories, the states that were the most negatively and positively disconnected are reported. A list of the states with the highest negative and positive total disconnect is also provided. In addition, since this report focuses on the state of Tennessee, it is highlighted and also included if it was not among either of the top-rank groups.
State government per capita spending
States with the lowest per capita spending:
1. Texas ($2,591)
2. Florida ($2,630)
3. Tennessee
($2,719)
4. Missouri ($2,816)
5. Colorado ($2,842)
Average ($3,431)
States with the highest per capita spending:
50. Alaska ($9,436)
49. Hawaii ($4,924)
48. New York ($4,810)
47. Delaware ($4,657)
46. Wyoming ($4,525)
Average: $3,431
Governmental administration
States with the lowest per capita spending for governmental administration:
1. Tennessee
($66)
2. Texas ($67)
3. Georgia ($69)
4. Alabama ($73)
5. Indiana ($74)
States with the highest per capita spending for governmental administration:
50. Alaska ($561)
49. Delaware ($370)
48. Vermont ($240)
47. Hawaii ($229)
46. Connecticut ($207)
Average $113
Intergovernmental transfers
States with the lowest per capita spending for intergovernmental transfers:
1. Hawaii ($129)
2. New Hampshire ($398)
3. Rhode Island ($600)
4. South Dakota ($644)
5. Maine ($685)
7. Tennessee
($761)
States with the highest per capita spending for intergovernmental transfers:
50. California ($1,760)
49. New York ($1,670)
48. Alaska ($1,660)
47. Michigan ($1,625)
46. Wyoming ($1,588)
Average $1,002
Analysis of disconnect
An important overriding finding is that Tennessee is the highest scoring state in terms of its need for effective policies and programs relative to the seven policy areas. In other words, there is no other state that has a greater challenge in terms of meeting the educational, health, safety, and welfare requirements of its people, building and maintaining its transportation system, and protecting the integrity of its environment. A complete listing of the state’s scores in this regard is provided as Attachment C. The highest scoring states in terms of overall need are presented below:
State overall policy effectiveness need
States with highest need for policy effectiveness:
1. Tennessee (28)
2. Florida (31)
Michigan (31)
Illinois (31)
5. Washington (33)
The following tables list the order of the fifty states as a combined function of (a) their amount of spending and (b) their estimated need for effective policy relative to the seven policy areas under consideration in this report.
Education
States with the least spending for education relative to estimated spending need (negative disconnect):
1. Arizona (-42)
2. California (-39)
3. Florida (-38)
4. Texas (-35)
5. Nevada (-29)
11. Tennessee
(-15)
States with the most spending for education relative to estimated spending need (positive disconnect):
50. Vermont (+38)
49. Hawaii (+31)
48. Montana (+29)
(tie) Minnesota (+29)
47. Iowa (+28)
Public welfare
States with the least spending for public welfare relative to estimated spending need (negative disconnect):
47. Montana (-36)
48. Arizona (-34)
49. California (-33)
50. Idaho (-32)
(tie) Texas (-32)
20. Tennessee
(-3)
States with the most spending for public welfare relative to estimated spending need (positive disconnect):
50. Connecticut (+39)
49. Minnesota (+34)
48. Arkansas (+33)
47. Vermont (+31)
46. Massachusetts (+30)
Health and hospitals
States with the least spending for health and hospitals relative to estimated spending need (negative disconnect):
1. West Virginia (-38)
2. Indiana (-27)
(tie) Oklahoma (-27)
4. North Dakota (-26)
5. Pennsylvania (-24)
13. Tennessee
(-14)
States with the most spending for health and hospitals relative to estimated spending need (positive disconnect):
50. Hawaii (+41)
49. Utah (+37)
(tie) Washington (+37)
47. New Mexico (+32)
46. Massachusetts (+31)
Highways
States with the least spending for highways relative to estimated spending need (negative disconnect):
1. California (-49)
2. Maryland (-41)
3. New York (-40)
4. Rhode Island (-37)
(tie) Hawaii (-37)
(tie) Michigan (-37)
18. Tennessee
(-13)
States with the most spending for highways relative to estimated spending need (positive disconnect):
50. Alaska (+48)
49. Wyoming (+46)
48. West Virginia (+45)
47. South Dakota (+42)
46. Montana (+41)
Police protection
States with the least spending for police protection relative to estimated spending need (negative disconnect):
1. Colorado (-43)
2. Texas (-41)
3. West Virginia (-32)
4. Alabama (-28)
(tie) Missouri (-28)
9. Tennessee
(-23)
States with the most spending for police protection relative to estimated spending need (positive disconnect):
50. Vermont (+37)
49. Maine (+33)
48. Wyoming (+32)
47. New Hampshire (+28)
46. Connecticut (+26)
Natural resources
States with the least spending to protect natural resources relative to estimated spending need (negative disconnect):
1. Virginia (-42)
1. Texas (-38)
3. Ohio (-36)
4. Tennessee
(-35)
5. Illinois (-34)
States with the most spending to protect natural resources relative to estimated spending need (positive disconnect):
50. Alaska (+44)
49. Wyoming (+41)
48. South Dakota (+37)
47. Vermont (+35)
46. North Dakota (+33)
Parks and recreation
States with the least spending for parks and recreation relative to estimated spending need (negative disconnect):
1. New Hampshire (-42)
2. California (-34)
3. Connecticut (-29)
(tie) Washington (-29)
4. Illinois (-24)
16. Tennessee
(-12)
States with the most spending for parks and recreation relative to estimated spending need (positive disconnect):
50. Louisiana (+47)
49. Rhode Island (+44)
48. Kentucky (+33)
47. Alaska (+30)
(tie) Mississippi (+30)
DISCONNECT TOTAL
States with the highest total negative disconnect (lowest spending relative to estimated spending need):
1. California (-154)
2. Texas (-149)
3. Florida (-136)
4. Tennessee &