Public Expenditure Analysis for Education

Deon Filmer

DECRG

January 2004

Key questions

What are the basics of public expenditures?– How much, what are the trends, what is public

money spent on, etc…

How can public money be put to its best use?– What should government finance?– What is the distributional impact of public spending?– How can the system be improved in order to

maximize the impact of public spending?

The basics of public expenditures

How much does government spend– As a share of GDP– As a share of total public expenditures

And how has that changed over time

Education spending as a percent of GDP

Haiti 1.1

Cambodia 1.9

China 2.9

Senegal 3.2

United States 4.8

South Africa 5.5

Denmark 8.2

Cuba 8.5

South Asia 2.5

Sub-Saharan Africa 3.4

World 4.4

Source: World Development Indicators Database. Data are for 2000

Public education spending as a percent of government expenditures

0

10

20

30

Eas

t A

sia

and

Pac

ific

Eur

ope

and

Cen

tral

Asi

a

Latin

Am

eric

aan

d C

arib

bean

Mid

dle

Eas

t an

dN

orth

Afr

ica

Sou

th A

sia

Sub

-Sah

aran

Afr

ica

Low

-inc

ome

coun

trie

s

Mid

dle-

inco

me

coun

trie

s

Hig

h-in

com

eco

untr

ies

Mean

25th percentile

75th percentile

Note: Of the 135 countries included, 52 have data for 2000, 8 for 2001, 30 for 1999, 17 for 1998. The remaining 28 have data from earlier in the 1990s. Source: World Development Indicators database.

Some things to keep in mind

Planned vs. actual expenditures Real vs. nominal expenditures Consolidated budget:

– all sources of public money– all expenditures for the sector

Spending on what …

Type of spending– Capital vs. Recurrent

Functional allocations – budget shares by level of education

Economic allocations – inputs—e.g., teachers, textbooks

Spending on teachers in Sub-Saharan Africa – as a share of recurrent spending (primary level)

Source: WDR 2004

But don’t get lost in the budget numbers

Frequently, more public money and better outcomes are not strongly related (or related at all)– Public Expenditure Analysis is an opportunity to

reflect on how to make money work to improve outcomes

Primary completion rate and public spending on education across countries (conditional on GDP per capita)

Source: WDR 2004

Changes in the primary completion rate and public spending on education within countries

Source: WDR 2004

Education expenditures and learningSpending and median math test scores

Source: TIMSS

AustraliaBulgaria

Colombia

Czech Republic

Denmark

Germany

Greece

Hong Kong

Iran

Japan

Korea

Lithuania

Netherlands

Norway

Portugal

Romania

Singapore

Slovak Republic

South Africa

Spain

Sweden

Switzerland

Thailand

United States

300

400

500

600

700

0 2 4 6 8

Public spending on pre-primary, primary, and secondary education (% of GNP)

Me

dia

n m

ath

sc

ore

Why public intervention ?

Public responsibility for education motivated by– Equity (Human Rights?)– Market failures– Social cohesion/Nation-building

Equity How unequally is education distributed – and where are the problems?

Source: WDR 2004

Percent aged 15 to 19 completing each grade or higher

Equity Is the current allocation of expenditures pro-poor – benefit incidence analysis

Source: WDR 2004

Equity Education as an anti-poverty program

If spending on education is justified as an anti-poverty program, then it needs to be assessed as such

Market failures Externalities

Productivity– e.g. spread of adoption of green revolution

technology in India (Foster and Rosenzweig, JPE 1995)

Social outcomes– Not just for the person making the education

investments but others as well

Externalities – social outcomesPercent of children with all immunizations by mother’s education

0

20

40

60

80

100

Nigeria 1999  Ethiopia 2000  Paraguay 1990  Haiti 2000  India 1998/99  Philippines 1998  Indonesia 1997  Peru 2000  Brazil 1996 

Source: Analysis of Demographic and Health Survey data

None

Primary

Secondary

Market failures

Capital market imperfections– Education is a long-term investment process: financial

institutions unwilling to take risk– Difficult for providers to borrow against future revenue stream– Difficult for students to borrow against future income stream

Asymmetric information– School quality is hard to assess– Student quality is hard to assess

Public provision and public finance … no single model

Voucher schools; regulated private schools (Chile basic, Netherlands, Belgium)

Public schools with little/no cost recovery (many)

Unsubsidized privatesector (Philippine, tertiary level)

Public facilitieswith cost recovery(US, Korea, Chile universities)

More public provision

More public finance

Public responsibility for education

A key question

Is public spending allocated in such a way as to address the equity and efficiency issues that motivate public involvement?

The share of private spending also varies substantiallyPrivate expenditure as % of total education expenditure

Source: Psacharopoulos and Nguyen 1995 “Fighting Poverty: the role of government and the private sector” World Bank.

0% 50%100%

S. AfricaMalaysia

BoliviaFranceUK

VenezuelaGhanaUSA

IndonesiaGermany

PeruUgandaSierra Leone

Netherlands

And government intervention need not be public provision or even finance

Regulation– Can regulate without financing or providing– Should regulate whether provision is public or private – If provision is public, regulator and provider should not be the

same

Information– To inform student and provider choice– To create incentives for policymakers and providers to deliver

Public provision is widespread in education, especially at basic levels

Buying a sandwich vs. getting educated– Supply and consumption of educational services are

not a simple market transaction with direct feedback from the customer

Just as there are market failures, there are government failures

Accountability in the delivery of education services … the WDR 2004 framework

Students/Parents

Policymakers

Schools/teachers

Provider organizations

Building accountability into the systemVoice

Students/Parents

Policymakers

Enhancing citizen voice to avoid:– Resources to political constituents and

voting blocks such as teachers’ unions (political patronage)

– Resources to personal gain (corruption)– Resources for less relevant learning

outcomes (delinked from private sector and labor markets)

Improving the compact(s) to ensure that providers have the incentive to serve poor people (well):

– Need to balance the autonomy of schools and teachers with performance assessment

– Schools and school systems must be able to manage for performance, particularly, to train and motivate teachers

Building accountability into the systemCompact

Policymakers

Schools/Teachers

Provider organizations

Building accountability into the systemCompact – Staff absence

Percent of staff absent in primary schools

0

10

20

30

Ecuador India Indonesia PapuaNew

Guinea

Peru Zambia Uganda

Building accountability into the systemCompact – Show me the money

Flow of recurrent expenditures for community and primary schools

Leave fares

"Grants to provinces" (Transfer)

Education subsidy (Q2,Q4)

Education subsidy (Q1 and Q3) Education subsidy (Q1,Q3)

Funds for maintainence and infrastructure

Infrastructure and Maintenance grant (Developed with districts for grades 7 & 8): only 46 schools in 2001

Bank Account

School

Contractor

Contractor

Bank Account

Non-teaching staff

Parents

Bank Account

Treasury

NDOE

District ?

LLG ?

Education payroll

Province

Teacher

The money trail in Papua New Guinea

Improving client power directly to improve services:– Choice (e.g. Girls’ scholarship program

Bangladesh, with school subsidies for girl’s enrollment)

– Participation (e.g. EDUCO in El Salvador with parent committees responsible for education delivery)

Building accountability into the systemClient power

Schools/Teachers

Students/Parents

Building accountability into the systemClient power – Parent participation in EDUCO

EDUCO promoted parental involvement…

…which boosted student performance

Source: Adapted from Jimenez and Sawada 1999

Key elements of PEA

Basic budget information Data on outcomes

– assessment system– household surveys

Links between spending and outcomes– Household surveys– School surveys– Administrative data

Impact evaluation evidence on programs What does this say about the adequacy of spending? What does this say about how to improve the effectiveness of

spending?