Transcript Diapositiva 1

Government Expenditure
Composition and Growth in Chile
January 2007
Carlos J. García
Central Bank of Chile
Santiago Herrera
World Bank
Jorge E. Restrepo
Central Bank of Chile
Organization of the presentation:
1.
Introduction and Stylized facts
2.
Model.
3.
Calibration of the model.
4.
Policy Experiments.
5.
Conclusions.
1. Introduction
• The purpose of this paper is to examine and
quantify the impact on growth of alternative
budgetary compositions.
• We use a model that captures some of the
specific stylized facts of the Chilean Economy.
• One of the specific targets is to test the effects of
larger social security payments. This is relevant
given that the pension system in Chile is being
reformed.
Total Expenditure and GDP per capita Growth
80s
% of GDP
GDP p.c. Growth(%)
28.3
0.6
% of GDP
GDP p.c. Growth(%)
37.9
1.8
% of GDP
GDP p.c. Growth(%)
25.2
0.1
% of GDP
GDP p.c. Growth(%)
25.6
3.5
90s
AFR
27.2
-0.6
ECA
33.5
1.0
LAC
21.7
1.8
SAS
24.7
3.3
2000s
80s
23.1
0.7
25.1
2.5
29.4
5.5
36.8
2.3
21.9
0.2
34.9
-0.5
90s
EAP
22.3
2.5
INL
37.1
2.0
MNA
30.0
1.2
2000s
21.8
2.4
29.2
1.9
29.0
2.1
27.3
2.0
AFR – Africa, EAP – East Asia and Pacific, ECA – East Europe and Central Asia, INL – Industrialized Countries,
LAC – Latin America and Caribbean, MNA – Middle East and North Africa, SAS – South Asia.
Source: World Bank World Development Indicators and IMF Government Finance Statistics
•
Total public spending as a share of GDP has evolved through time and across
regions with little or no relationship with growth rates.
•
The public spending ratio has decreased, while growth rates show diverse
behavior.
Different Regions: Composition of Total Expenditure (%, consolidated central government)
1980
Agriculture
6.7
Education
12.8
Health
4.2
T&C
7.3
Social Security
3.4
Defense
8.8
Order & Safety
0.0
Fuel & Energy
1.3
Mining & Manuf. 2.0
Public Service
18.2
Housing
1.5
Recreation
1.3
Other
32.3
AFR
1998
4.8
13.9
5.3
2.3
5.8
9.1
4.6
0.4
2.6
14.7
1.6
0.6
34.4
2001
4.6
14.7
8.4
6.1
21.4
0.8
7.2
0.1
0.7
9.8
6.6
1.3
18.3
1980
9.7
14.1
5.4
11.7
3.2
18.8
0.7
2.0
2.5
14.5
2.7
0.7
13.8
EAP
1998
4.9
12.5
5.6
7.9
6.7
13.6
4.8
0.9
0.4
10.3
5.4
2.8
24.2
2001
2.4
13.4
7.1
4.5
12.1
8.3
7.0
1.0
0.5
10.2
3.9
2.1
27.6
1980
3.1
5.7
1.4
4.0
7.7
23.0
0.0
2.0
14.1
10.3
1.5
0.1
27.1
ECA
1998
3.3
6.6
8.6
4.1
32.2
6.1
5.8
0.6
1.2
5.7
1.3
1.8
22.7
2001 1980
2.9 3.8
6.4 8.5
8.8 9.0
4.2 5.9
35.0 34.9
5.2 8.5
6.0 0.5
0.8 0.8
0.4 1.5
6.4 7.3
1.1 2.4
1.6 1.0
21.0 15.8
INL
1998
2.6
7.9
10.3
3.6
34.2
6.6
2.8
0.5
1.0
5.9
2.5
1.3
20.8
2001
1.4
6.0
12.1
2.1
34.2
12.9
2.9
0.1
0.6
6.1
2.9
0.9
17.8
1980
5.5
12.5
8.0
7.4
14.0
6.5
0.4
1.8
1.9
15.4
2.9
0.7
22.8
LAC
1998
2.3
15.7
9.1
3.9
23.8
4.5
5.7
0.9
0.7
8.5
3.4
0.5
21.0
2001
1.5
17.4
10.9
3.9
24.1
4.2
6.5
0.4
0.6
7.8
3.2
0.6
18.8
1980
4.9
12.8
4.7
5.1
6.1
17.7
0.0
5.1
3.8
11.8
5.4
3.0
19.5
MNA
1998
5.3
15.0
5.2
2.9
9.5
14.7
7.8
5.8
1.4
10.1
3.1
2.1
17.1
2001
0.5
13.3
7.8
2.7
7.5
15.2
12.2
0.3
0.3
29.5
3.7
0.5
6.3
1980
8.1
5.2
3.1
18.2
3.6
11.8
0.0
3.9
3.6
11.8
3.6
0.6
26.7
SAS
1998
4.6
10.3
6.1
7.9
3.8
12.3
4.1
6.5
0.8
9.9
5.3
0.4
28.1
2001
3.8
9.3
5.0
6.3
3.6
11.1
3.9
9.0
0.5
12.1
4.3
0.2
31.0
Source: Calculated using data from IMF Government Finance Statistics
•
The composition of public expenditure has varied significantly with clear patterns
across regions and through time
•
A notable trend is the rising importance of social security payments.
•
Agriculture spending and transport and communication are decreasing in
importance within central government budgets
Chile: Composition of Central Government Expenditure
(% of GDP)
1990
1996
2005
Agriculture and others
1.2
1.2
0.9
Defense
2.3
1.5
1.3
Education
2.3
2.8
3.3
Environment
0.0
0.1
0.1
Order & Law
0.9
1.0
1.3
Health
1.9
2.4
2.9
Housing
0.9
1.1
1.0
Public Service
2.8
1.4
1.3
Recreation
0.1
0.1
0.1
Social Security
7.4
6.4
5.8
Transportation
0.8
1.6
1.7
Others
0.1
0.0
0.1
Source: Estadísticas de las finanzas públicas Ministerio de HaciendaDIPRES several issues.
•
Health and education are increasing in importance within central government budgets
•
Social security payments is decreasing but important as % of GDP.
2. Model: The Framework of General Equilibrium
• The model is overlapping generation model was developed by
Glomm-Rioja (2004) for Brazil, but this version include additional
types of expenditure (maintenance of public capital) and changes in
the calibration parameters.
• The building blocks of the model are defined by the preferences, the
technology, and the resource constraints.
• Three crucial features are:
– Consumption and leisure decisions are made by agents differentiated
by their generation: they study when young, work in adulthood, and
receive transfers (social security) payments when old.
– Government expenditure is productive (in infrastructure and education)
and unproductive (transfer payments to the old), affecting production
and consumption decisions.
– Interest rates depend on the size of public debt.
2. Model: Preferences
•
•
•
•
Each generation of households lives for three periods: youth, adulthood and
retirement.
Each individual, when young, is endowed with one unit of time which can be
allocated to learning or leisure.
During adulthood the individual supplies labor inelastically, and allocate labor
income between current consumption and savings.
When retired the individual lives on transfers and returns on savings.
Specifically, preferences are given by
ln(1  nt )  ln ct ,t   ln ct ,t 1
•
(1)
The evolution of human capital follows the rule below:
ht  Bnt Et1ht1 , 0   ,  ,   1,
B0
(2)
2. Model: Preferences
•
The utility maximization problem is solved recursively, starting with the problem
faced by adults:
max
ln ct ,t   ln ct ,t 1
ct ,t  st  (1  L,t )wt ht
ct ,t 1  (1  (1  K ,t )rt 1)st  Tt 1
s.t.
given
•
(3)
( wt ,rt 1 , L ,t , K ,t ,Tt 1 ,ht ) ,
First order conditions yields the savings decisions given by
st 
Tt 1

(1   L, t ) wt ht  1
1 
1  1 (1 K ,t 1 ) rt 1
(4)
2. Model: Preferences
•
Replacing the optimal savings (equation (4) ) into the objective function in the
consumer’s problem (3) yields an indirect utility function for the adult

~ h  T   ln(1   )(1  ~
IUFt  ln(1  nt )  (1   ) ln(1  ~
rt 1 ) w
rt 1 )   ln(
)
t t
t 1
1 
•
(5)
The problem for the young is hence to maximize (5) with respect to learning
time, subject to the law of motion for human capital in (2). The solution to this
problem is defined by the following nonlinear equation
(6)
~ Bn E  h   T   (1   )(1  ~
~ Bn 1 E  h 
(1   (1   ))(1  ~
rt 1 )w
r
)
w
t
t
t 1 t 1
t 1
t 1
t
t
t 1 t 1
2. Model: Production
•
The aggregate production technology for the single non-storable consumption
good is given by
 1
Yt  AG
, 0   ,   1,
t Kt H t
•
(7)
Public infrastructure capital evolves according to
Gt 1  (1   G (mt ))Gt  I G,t
•
A0
(8)
The private physical capital evolves according to
Kt 1  (1   K )Kt  I K ,t
(9)
2. Model: Production
•
The representative firm maximizes profits, taking as given the market factor
prices. Perfect competition dictates that the followings first order conditions:
wt  (1   ) HYtt
qt  
•
Yt
Kt
(10)
 (1   K )
The firm’s profits maximization conditions in (10) imply that private physical
capital will evolve according to the following path
K t 1  (A)
1
1 

1
Gt 1 H t 1 (rt 1   K )
1
 1
(11)
2. Model: Fiscal Policy
•
The government provides public goods, which is financed either by tax revenue
or by borrowing
•
The government expenditure as percent of GDP is distribute on investment in
infrastructure, on maintenance, on education, on transfers, and on other
general public services (non-utility enhancing).
•
The government collects taxes on labor income at rate and on capital (interest)
income at rate . It can also choose to raise debt to finance spending.
•
Formally, the government budget constraint is given by
Dt 1   L,t wt H t   K ,t rt K t 
( G ,t   M ,t   E ,t   T ,t   P ,t )Yt  (1  rt (1   K ,t ))Dt
(12)
2. Model: Competitive Equilibrium
A macroeconomic equilibrium is defined by the following system, where
uppercase letters indicate aggregate variables.
1. The household utility maximization problem is solved. That is, conditions (4) and
(6) hold.
2. The representative firm’s profits maximization problem is solved. That is,
condition (10) holds.
3. The government budget constraint (12) is satisfied.
4. The goods market clears: Ct  St  Taxt  Yt  (1   K ) K t
5. The competitive input market for human capital (labor) clears: H t   ht
6. The interest rate is determined as suggested by Schmitt-Grohe and Uribe (2003),
with a debt elastic interest rate as follows:
rt  r *  R( DYtt )
(13)
3. Model Calibration for Chilean Economy
Table 1 Benchmark Parameter Values
Discount Factor (  )
Total Factor Productivity (  )
Human Capital parameter ( B )
Capital’s Share of GDP (  )
Public Capital Elasticity ( )
Public Education Expenditure Elasticity (  )
Interest rate sensitivity to public debt (  )
Learning Time Elasticity (  )
Parental Human Capital Elasticity (ρ)
Depreciation parameter - public capital (  )
Depreciation rate - private capital (  K )
(.973)30
13.0, calibrated to get balanced growth
3.87, calibrated to get balanced growth
0.5
0.3
0.1
0.04
0.137, calibrated to get n  .15
0.75, calibrated to get balanced growth
4.0, to match 10% depreciation per
annum
10% per annum
Tax Revenue as a fraction of GDP (Tax)
Transfers as a fraction of GDP (  T )
Public Education Expenditure (  E )
Public Capital Expenditure (  G )
Public Expenditure on Maintenance ( M )
Non-utility enhancing Public Expenditure (  P )
Labor income tax rate (  L )
Capital income tax rate (  K )
20.5%
8%
6.3%
1.0%
1.0%
4.2%
20%
World Interest Rate ( r * )
5.58%
17%
4. Policy Experiments I: increase in expenditure (1% of
GDP)
GDP growth rate after a permanent increase in expenditure
5.90
5.90
5.70
5.70
5.50
5.50
5.30
5.30
5.10
5.10
4.90
4.90
4.70
4.70
4.50
4.50
1
2
3
Benchmark
Transfer
Infrastructure
Maintenace
4
5
Education
1
2
3
Benchmark
Transfer
Infrastructure
Maintenace
4
5
Education
4. Policy Experiments II: increase in expenditure (1% of
GDP)
GDP growth rate after a temporary increase in expenditure
6.10
5.90
5.90
5.70
5.70
5.50
5.50
5.30
5.30
5.10
5.10
4.90
4.90
4.70
4.70
4.50
4.50
1
2
3
Benchmark
Transfer
Infrastructure
Maintenace
4
5
Education
1
2
3
Benchmark
Transfer
Infrastructure
Maintenace
4
5
Education
• 5. Conclusions…
• We think this is a useful first step in quantifying the impact on long
run growth and income of alternative budget compositions capturing
key elements in the Chilean economy.
• The paper’s results provide quantitative evidence supporting the
hypothesis of the importance of public investment in achieving
higher income in the long run.
• Even though the preliminary results of the simulations show that
there is a cost, in terms of growth, of increasing social security
payments, this cost is low.
• 5. …and future work
• It would be more realistic to model a non-linear elasticity of public
investment such that productivity of public capital decreases as the
amount of investment increases.
• It could be useful to model a function of “efficiency” of investment,
such that not all public investment is transformed into public capital.
• We could also consider increasing administrative costs of taxation.
• Future extensions of this paper could include public health
expenditures that enhance human capital.
Government Expenditure
Composition and Growth in Chile
January 2007
Carlos J. García
Central Bank of Chile
Santiago Herrera
World Bank
Jorge E. Restrepo
Central Bank of Chile