Economics 151 The Economics of the Public Sector: Expenditure

1
Economics 151 The Economics of the Public
SectorExpenditure

• Professor Nora Gordon
• Fall 2004
• Lecture 20

2
Testing the Tiebout model sorting

• Prediction people sort into communities based
  on taste for G
• Evidence Gramlich and Rubinfeld (1982), survey
  people on their demand for public spending
• In smaller communities, less variation in
  reported demand

3
Capitalization

• Tiebout model predicts two types of
  capitalization under property tax (not head tax)
  finance
• If towns A and B have same tax rates and differ
  in school quality, homes cost more where quality
  is better.
• ? Black (1999) tests this.
• If towns A and B have same school quality and
  differ in tax rates, homes cost more where tax
  rate is lower.
• ? Rosen (1982) tests this.

4
Testing the Tiebout model capitalization

• Prediction houses will cost more in towns which
  produce more/better G for the same cost
• Evidence Rosen (1982), looks at house prices
  around San Francisco just before and after Prop.
  13
• Prop. 13 imposed local property tax limits
• Rosen found 1 reduction in prop tax?7 increase
  in house value

5
Interpreting Rosens finding

• 1 reduction in prop tax per year?7 increase in
  house value
• What is NPV of prop tax reduction?

• Whats missing here?

6
Home prices and school quality

• If school quality is a normal good, demand
  increases with income.
• But price may remain constant, fall, or decrease
• Demand for unobservable house or neighborhood
  characteristics will rise also.

7
Blacks (1999) solution to unobservable
neighborhood characteristics

• Compare prices of home in the same geographic
  neighborhood, but in different school attendance
  areas
• Neighborhood characteristics no longer omitted
• We dont need to know all about your neighborhood
• Just need to know which one you are in

8
Map of Melrose
9
Optimal fiscal federalism

• Which types of goods should be provided locally?
• Those with strong tax-benefit linkages
• Those that benefit a majority of residents

10
Fiscal federalism and inequality

• There is much inequality in level of public
  spending across localities.
• Current spending per elem/sec pupil
• Canton MS 4,119
• Cambridge MA 17,286
• Athol-Royalston MA 6,777
• US average about 7,898
• Why might we care about this?
• Failures in Tiebout mechanism
• Externalities across jurisdictions

11
Intergovernmental grants

• Intergovernmental grants allow redistribution
  across jurisdictions, while still retaining some
  local qualities of Tiebout model.
• Grants go from higher level of govt to lower.
• Federal ? state
• Federal ? local
• State ? local

12
Matching grants

• Amount of grant depends on amount of local
  contribution
• One-to-one match is simple case local govt
  spends 1, gets 1 from state.
• Match can be less than or greater than one.
• Match affects the price of local spending (slope
  of towns budget constraint).

13
Towns budget constraint with and without
matching grants
Exp on pvt goods
Y total income in the town
Y
BC with no matching grant
Exp on educ
Y
14
Income and substitution effects of matching grants
Exp on pvt goods
Y total income in the town
Y
BC with no matching grant
Exp on educ
Y
3Y
E1
E2
15
Block grants
Exp on pvt goods
Y total income in the town
YBG
Y
Exp on educ
E1
E2
Y
YBG
16
Conditional block grants
Exp on pvt goods
Y total income in the town
YBG
In this case, conditional block grant has same
effect as unconditional.
Y
Exp on educ
E1
E2
Y
YBG
Entire change in E is pure income effect
17
The flypaper effect
Exp on pvt goods
Y total income in the town
YBG
Y
If spending on E goes up by full amount BG, then
new E3.
Exp on educ
E1
E2
E3
Y
YBG
Income effect predicts new E2.
18
The flypaper effect

• We expect to see new education spending E2
• Much research finds levels closer to E3 than to
  E2
• The grant disproportionately sticks to
  education spending.
• Why might this be?