How data can inform policy? Some examples…. 1. Data from public budgets Public expenditures and revenues are telling a lot about policy (and government.

How data can inform policy? Some examples…

1. Data from public budgets

• Public expenditures and revenues are telling a lot about policy (and government efficiency)– Accountability– Budget transparency– Allocative efficiency

Line-item vs. programme budgetMinistry of Health

Line-item Budget Programme Budget

Salaries & wages

6 000 General Administration

462

Overtime

150

Building expenses

800 Primary health care & health promotion

4 326

Transport

750

Equipment 400

Shipping 125 Hospital services 2 817

Water & lights

15

Telephone 25 Training & medical research

692

Printing 20

Consumables 12

TOTAL 8 297 TOTAL 8 297

Public Financial Management

Fiscal Discipline(gov’t budget balance)

Operational efficiency(implementation)

Allocative Efficiency(public expenditure

planning)

Tax & aid policy(revenue planning)

2. Data from Household Surveys

• Descriptive statistics – together they can be powerful– Focus on the big picture of “issues and policy responsiveness”– Can be used for highlighting vicious and virtuous policy cycles

(multidimensional model of child poverty)

• Exploring causality with multivariate statistics– What is the role of certain factors (e.g. parental education) in

child outcomes– Why certain policies work or do not work

Percentage of children experiencing severe deprivations in East Asia

Source: MICS/direct communication with Bristol University

Income Poverty DynamicsPercentage of households with less than Rf. 15 per person per day, Atolls

60%

23% 11%

9% 4% 5% 3%

40%

37% 29%

14% 7% 32% 27%

'poor'

'non-poor'

1997

2004

2005

Figure 1.3: Income Poverty Dynamics 1997, 2004 and 2005

Income poverty dynamics in the Maldives, 1997, 2004 and 2005

Source: Dr. Fuwad Thowfeek, Statistics Maldives

Intergenerational income mobility: your father earns 100 per cent more than mine

- what per cent impact will that alone have on our earning differences?

0 20 40 60 80 100

Ecuador

Brazil

Peru

Malaysia

United Kingdom

United States

Pakistan

Nepal

France

Germany

Sweden

Finland

Canada

Source: Dr Miles Corak Statistics Canada

MATERIALLY POOR

POOR HEALTH OUTCOME

NOT ATTEDING PRE-SCHOOL

20.2% 17.9%

1.3%

11.8%

15.8% are not poor, have access to preschool, clean water and are in good health

2.9%

22.7%

7.3%

Albania: % of children 3-5 yrs old materially poor with poor nutritional Albania: % of children 3-5 yrs old materially poor with poor nutritional outcomes and not attending pre-schooloutcomes and not attending pre-school – Venn diagrams

Source: 2002 LSMS. Note: Total number of children 450.

Angela Baschieri and Jane Falkingham (University of Southampton)

Multidimensional child poverty concepts broaden policy focus

Anthropometric failure and breastfeeding practices in Tajikistan

0

5

10

15

20

25

30

35

Wasted Underweight Stunted

Exclusively or partiallybreastfed

Fully weaned

Source: MICS 2005 and Baschieri and Falkingham, 2007

Nutritional status by breastfeeding pattern for children less than 18 months

Breastfeeding practices

• Most women in Tajikistan stop exclusively breastfeeding and switch to a mix feeding pattern relatively early– Amongst children aged 6-23 months under 5 percent are either

‘exclusively’ or ‘almost exclusively’ breastfed.

• As a result many children are exposed to the risk of poor nutrition and associated adverse developmental consequences.

Is family land ownership an effective policy against child malnutrition?

(results of multivariate analysis)

• We control for children age (months), region, mother education, wealth quintile, ethnicity, sanitation, household access to land, ownership of livestock

• We found that children living in a households with access to land have higher probability of being underweight that those without access to land

3. International comparisons

• Can be helpful for “big policy ideas”• Highlighting policy coherence and/or policy efficiency• Can stimulate policy transfer• Advocacy value

I n c o m e in e q u a l i t y a n d e c o n o m ic o u t p u t : p a t h s o fd e v e lo p m e n t

U nited K ing d o m

ItalyC anad a

F rance

F R G

U S A

J ap an

S w ed enN etherland s

B U L9 4

C Z E 9 2S LK 9 4

HU N 9 4

P O L9 4

R O M 9 4

R U S 9 4E S T 9 4

B ras il

C hile

C o lo mb ia

P anama

C o s ta R ica

B U L8 9

C Z E 8 9S LK 8 9

HU N 8 9

P O L8 9R O M 8 9 R U S 8 9

E S T 8 9

K o rea, R ep

S ing ap o re

LIT 8 9

LIT 9 4

M O L8 9

M O L9 4

1 0

2 0

3 0

4 0

5 0

6 0

7 0

0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 1 0 0

P P P e s t im a t e s o f G D P p e r c a p it a ( U S A = 1 0 0 )

Gini

coef

ficien

t

W e s te rn e c o n o m ie s (e a rly 9 0 s )

S o u th - A m e ric a n e c o n o m ie s (e a rly 9 0 s )

C E E 1 9 8 9

C E E 1 9 9 4

V enez uela

M o s t d e v e lo p e d F a r-S o u th e rn E a s t e c o n o m ie s (e a rly 9 0 s )

S o u r c e : C h i ld r e n a t R is k in C e n t r a l a n d E a s te r n E u r o p e : P e r i l s a n d P r o m is e s , R e g io n a lM o n i to r in g R e p o r t - N o . 4 , U N I C E F 1 9 9 7 .

Challenges in using statistics to inform policy

• Existing concepts, data and availability• Sensitivity analysis, robustness

– child focus– thresholds– economy of scale/equivalence of scale (income data)

• Design causal analysis: Need hypotheses plus data to test them

• Overlaps of income and non-income dimension: limitation