emmanuel abuelafia, sergio berensztein, miguel braun and luciano di gresia

Centro de Implementación de Políticas Públicas para la Equidad y el Crecimiento

Who decides on Public Who decides on Public Expenditures? A Political Economy Expenditures? A Political Economy

Analysis of the Budget Process: Analysis of the Budget Process: the case of Argentinathe case of Argentina

Emmanuel Abuelafia, Sergio Berensztein, Miguel Braun and

Luciano di Gresia

Av. Callao 25, 1° • C1022AAA Buenos Aires, Argentina - Tel: (54 11) 4384-9009 • Fax: (54 11) 4371-1221 • [email protected] • www.cippec.org


Outline of the presentation

1.The PMP in Argentina and literature review

2.A (very brief) overview of the formal budget process

3.The actual workings of the budget process

4.Conclusions and recommendations


The PMP in Argentina

Institutional Legacies

1930-1983

Constitutionand electoral

rules

•Short horizonsfor policymakersand interest groups

•Weak arenasfor intertemporaldeals

•Congress

Policyinstability

Lack of coordination

Lack of cooperation

Deep determinants

Nature of the PMP

Outer features of policies


Implications for the budget process

1.The President is strong vis a vis Congress

2.Provincial governors are powerful informal actors

3.Weak and unstable bureaucracy, which limits budget efficiency

4.Instability in budget outcomes


Other papers

- Jones (2001): Congress has an important role in the budget process.

- Rodriguez and Bonvecchi (2005): the role of Congress is weak.

- The evaluation of the role of Congress depends on the universe of public expenditure that one analyzes.


The National Budget 2004

34%

34%

13%

11%

8%

.

48%

6%

37%

9%

55%

7%

18%

13%

7%

Total Revenues: $108.080

Municipal’s Revenues

NationalAdministrationRevenues

Coparticipation

PE, FF y Other entities

Provincial’s Revenues

Total Expenditure:

$104.371

Primary Surpluss: $12.645 = 3% PBI

Municipal Expenditure


Provincial’s Expenditure

NationalAdministrationExpenditure

Other Expenditures

Payroll

Debt’s Services

Transfers

Pensions

NationalAdministration Budget

Law: $59.712

Argentinean Public Sector. 2004

34%

34%

13%

11%

8%

.

48%

6%

37%

9%

55%

7%

18%

13%

7%

Total Revenues: $108.080

Municipal’s Revenues

NationalAdministrationRevenues

Coparticipation


Provincial’s Revenues

Total Expenditure:

$104.371

Primary Surpluss: $12.645 = 3% PBI

Municipal Expenditure


Provincial’s Expenditure

NationalAdministrationExpenditure

Other Expenditures

Payroll

Debt’s Services

Transfers

Pensions

NationalAdministration Budget

Law: $59.712

Argentinean Public Sector. 2004


Degree of Exogeneity

We constructed an index that reflects how much the public expenditure deviates from its trend.

dSit = β0 + β1 dSi(t-1) + β2 dGDPt + β3 dPOPt +

+β4 dPRIt+ β5 dEMPt + β6 dDEBt + + wt (1 + β7 dGDPt + β8dPRIt) + uit

Zit = Abs(uit ) / Sit


Discretionality of Public ExpenditureYear Z

19811982 16.6%1983 2.9%1984 10.3%1985 17.4%1986 5.8%1987 13.8%1988 13.2%1989 0.9%1990 4.5%1991 6.3%1992 4.4%1993 9.2%1994 9.6%1995 4.3%1996 6.0%1997 3.8%1998 2.3%1999 7.7%2000 4.1%2001 5.3%2002 2.1%2003 7.1%

0%

5%

10%

15%

20%

25%

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003


A brief history of fiscal outcomes


The formal budget process

APPROVAL

LEGISLATIVO

CongressDiscussion and

approval. Posiblemodiifications

Sets the debt limit

ExecutiveSanction or veto

FORMULATION

Executive

National Budget Office

Agencies and Entities

CUENTA DE INVERSION

Congress

Controlling Agencies

PUBLIC ADMINSTRATION

Implements the budget

IMPLEMENTATION

and CONTROL

Budget Proposal

BUDGET LAW

APPROVAL

LEGISLATIVO



Sets the debt limit



Sets the debt limit



FORMULATION

Executive



FORMULATION

Executive Executive





CUENTA DE INVERSION

Congress




CUENTA DE INVERSION

CongressCongress







IMPLEMENTATION

and CONTROL

Budget ProposalBudget

Proposal

BUDGET LAWBUDGET LAW


The actual workings of the budget process

1.The President is the key actor, and he maximizes his goals subject to the constraints set by the rigidity of the budget, fiscal rules and agreements with IFIs.

2.He also has to deal with macroeconomic shocks and other powerful actors. (governors, lobbies, IFIs)


The actual workings of the budget process

What are the President´s goals?

- Obtaining reelection (political coalition)

- Favorable public opinion (macro stability, national public goods)

- Helping his home province (pork)


Formulation Approval Implementation

President is key actor

-Strategic use of forecasts-President defines allocation of marginal resources-Allocation of funds to home province

-Limited changes (pork)-Congress lacks technical capability

-Changes in both size and allocation-Changes not approved by Congress

-Delegation. New superpowers!!!

Constraints -Rigidity-Macro conditions-Fiscal rules-IMF agreements-Lobbies (eg: Carpa blanca)

-Need to underexecute if low growth-Rigidity limits adjustment-Bureaucratic learning


Strategic use of revenue forecasts

41,1

69

41,7

32 45,2

13

45,7

03

42,8

65 46,4

13

39,8

95

62,2

68

37,8

15

35,8

82

40,2

64

41,2

16

40,3

89 43,1

89

37,1

76

43,2

11

57,7

33

42,9

80

-

10,000

20,000

30,000

40,000

50,000

60,000

70,000

1995 1996 1997 1998 1999 2000 2001 2002 2003

$ m

illio

ns

Revenues' forecast

Collected Revenues


Changes made by Congress and Executive

-15,00%

-10,00%

-5,00%

0,00%

5,00%

10,00%

15,00%

1995 1996 1997 1998 1999 2000 2001 2002 2003

CongressExecutive


Δ Composition by Congress and Executive Changes in expense's composition

(before the crisis: 1995-2001)

-

0.00005

0.00010

0.00015

0.00020

0.00025

0.00030

0.00035

0.00040

0.00045

0.00050

1995 1996 1997 1998 1999 2000 2001

Stage 1: Play Congress Stage 2: Play Executive branch Stage 3: Efficiency in the execution


Implementation

High volatility of ministers and secretaries. (Average duration: 2.36 years).

The institutional capacities are undermined by– Lack of incentives faced by permanent

bureaucrats– Lack of experience of parallel bureaucrats

The purchasing mechanism requires a high institutional capacity.

Then, the degree of execution will be lower the less experienced the minister is.

The bureaucratic work may involve a learning process.


Bureaucratic learningTable 1: Regression results for bureaucratic learning

Dependent Variable Program's degree

of execution

Program's degree of execution relative to agency's execution

Consumption and usage

goods execution

Coefficient Standard

Error Coefficient Standard

Error Coefficient Standard

Error

Agency's Degree of Execution 0.591 0.043 0.733 0.127 % of Consumption and Usage goods -0.098 0.014 -0.086 0.015 0.234 0.035 % of Payroll 0.058 0.006 0.071 0.006 0.234 0.023 % of earmarked taxes 0.028 0.008 0.038 0.009 0.141 0.026 % of external transfers -0.564 0.048 -0.601 0.053 -0.268 0.066 % of external credit -0.280 0.019 -0.306 0.019 -0.249 0.054 % execution external transfers 0.054 0.025 0.076 0.027 0.115 0.054 % execution external credits 0.128 0.005 0.133 0.007 0.057 0.019 Ministry's experience 0.003 0.001 0.003 0.001 0.010 0.002 Ministry's experience squared -0.00004 0.00001 -0.00004 0.00001 -0.0001 0.00003


An exception:

The Ministry of Economy.– The average seniority of the Ministry of Economy

employees is around 23 years, 15% higher than the average seniority of all the SINAPA´s employees (20 years). Compared with the other agencies, the Ministry is in the top of the employees´ seniority distribution.

– The educational level of the Ministry of Economy’s bureaucracy is above the average. 38% of the total Ministry’ s employees have tertiary education, two times more than the Ministry of Internal Affairs and 20% more than the SINAPA´s average.

– The Ministry of Economy has a higher proportion of upper echelons. 44% of its employees are in the top 3 categories, compared with the system average of 38%.


Extra Budgetary Funds

Number ofFunds created

2005́ s Bugdet

1994 1 7.2 1995 1 732.1 1996 1 12.0 1997 2 20.6 2000 2 23.3 2001 4 2,643.7 2002 1 126.8 2003 2 138.6 2004 1 0.6

15 3,705

Agencies that consolidates in the Public Sector Budget:

AFIP, INCAA, INNSJyP

Entities and Funds that do not consolidate:

Fondo Especial del Tabaco, National Universities, Yacyreta, Fondo solidario de Redistribucion.


Evaluation and Control

SIGEN: depends on the President. AGN: depends on Congress.

Evaluation is weak. Audits are scarce, not timely, only legal

compilance and not an effective tool.


Conclusions

1.Actual workings of the BP are consistent with Spiller-Tommasi description of Argentine PMP

- Executive as key actor

2.Some surprises: The bureaucracy that deals with the BP is stable and competent.


Recommendations

1.Strengthen coordination mechanisms between the federal and provincial governments

2.Capacity building in Congress

3.Strengthen expenditure evaluation

4.Use of private forecasts for budgeting

emmanuel abuelafia, sergio berensztein, miguel braun and luciano di gresia

Documents