REGIONALANDCOUNTRY REPORTS OF THEASEANASSESSMENT ONTHE SOCIAL IMPACT OFTHEGLOBAL FINANCIAL CRISIS

7/31/2019 REGIONALANDCOUNTRY REPORTS OF THEASEANASSESSMENT ONTHE SOCIAL IMPACT OFTHEGLOBAL FINANCIAL CR

1/68

REGIONAL AND COUNTRY

REPORTS OF THE ASEAN

ASSESSMENT ONTHE SOCIAL IMPACT OF

THE GLOBAL FINANCIAL CRISIS

with the support of:


2/68

The Association of Southeast Asian Nations (ASEAN) was established on 8 August 1967. The Member States of the Association

are Brunei Darussalam, Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, Philippines, Singapore, Thailand and Viet Nam.

The ASEAN Secretariat is based in Jakarta, Indonesia.

For inquiries, contact:

The ASEAN Secretariat

Public Outreach and Civil Society Division

70A Jalan Sisingamangaraja

Jakarta 12110

Indonesia

Phone : (62 21) 724-3372, 726-2991

Fax : (62 21) 739-8234, 724-3504

E-mail : [email protected]

General information on ASEAN appears online at

the ASEAN Website: www.asean.org

Catalogue-in-Publication Data

Regional and Country Reports of the ASEAN Assessment on the Social Impact

of the Global Financial Crisis

Jakarta: ASEAN Secretariat, July 2010

332.4959

1. Monetary Systems - Protection Policy

2. Social impact - ASEAN

ISBN 978-602-8411-37-0

The text of this publication may be freely quoted or reprinted with proper acknowledgement.

Copyright ASEAN Secretariat 2010

All rights reserved


3/68

REGIONAL AND COUNTRY

REPORTS OF THE ASEAN

ASSESSMENT ON

THE SOCIAL IMPACT OF


ith the support o:


4/68

iv

This volume is a product resulting rom a project jointly implemented by the sta o the International Bank or Reconstruction and Development/

the World Bank and the ASEAN Secretariat, with nancial support o the Australian Government. The ndings, interpretations, and conclusions

expressed in this paper do not necessarily refect the views o the Executive Directors o the World Bank, the governments they represent, the ASEAN

Secretariat, the Australian Government and/or ASEAN Member States. The World Bank, the ASEAN Secretariat and the Australian Government do

not guarantee the accuracy o the data included in this work. The boundaries, colors, denominations, and other inormation shown on any map in

this work do not imply any judgment on the part o the World Bank concerning the legal status o any territory or the endorsement or acceptance

o such boundaries.


5/68

v

The global nancial crisis in 2008 combined with ood and uel crises highlighted the vulnerability o people who are

poor and near-poor in the ASEAN region. While economic recovery is now underway, all ASEAN Member States

experienced a signicant slowdown in their growth, with some alling into recession. Newspapers across the region

reported on actory closures, the return o laid-o workers to their villages, salary cuts and decreased remittances

home and amilies struggling to make ends meet. Many ASEAN Member States endeavoured to respond in a timely

ashion oten with limited inormation available both on the social impacts o the downturn and the eectiveness

o existing social protection instruments.

This turn o events has made policy makers keenly aware o the need to not only understand better the social

consequences o economic volatility but also to strengthen social protection and labour market policies, both

during crisis and normal periods. This creates an opportunity or a new dialogue on the role o social protection

and labour market policies and programmes in ASEAN Member States.

To support ASEANs mission to strengthen the oundations or a region o lasting peace, security and stability,

sustained economic growth, shared prosperity and social progress, the ASEAN Secretariat partnered with the

World Bank to put together this Assessment o the Social Impact o the Global Financial Crisis. The project was

agreed by the ASEAN Ministers on Rural Development and Poverty Eradication at their Sixth Meeting in May 2009

in Ha Noi, Viet Nam.

This report provides a rapid assessment o the social impacts o the crisis and oers a stocktaking o the range o

social protection measures adopted in ASEAN Member States in response to the crisis and in the context o the

pre-existing programmes available. The report also highlights a number o areas which deserve policy makers

attention, such as social impact monitoring, programme monitoring and evaluation, targeting and programme

coverage, as well as institutional coordination.

This report, which is supplemented with country case studies available in the attached CD, was prepared with

the participation o the relevant Ministries and ASEAN sectoral bodies, including the Senior Ocials Meeting onRural Development and Poverty Eradication (SOMRDPE), the Senior Ocials Meetings on Development Planning

(SOMDP), the Senior Labour Ocials Meeting (SLOM), the Social Welare and Development (SOMSWD) and the

ASEAN Heads o Statistics Oces Meeting (AHSOM). ASEAN Member States had contributed their inputs to

the report.

FOREwORD


6/68

vi

This report is the rst in this emerging cooperation between the ASEAN Secretariat and the World Bank East Asia and

Pacic Region in the area o social protection and labour markets. It is a sign o our shared interest in the welare and

the uture o the people o ASEAN. This joint eort is expected to also contribute to realising an ASEAN Community that

is people-oriented and socially responsible.

Appreciation is due to the Australian Government or the nancial support provided or this project.

We hope that this report will be a useul reerence or policymakers and relevant stakeholders across the region in

undertaking their social protection policies and programmes.

DR. SURIN PITSUWAN JAMES W. ADAMS

Secretary-General o ASEAN Regional Vice President

East Asia and Pacic Region

World Bank


7/68

vii

This report is a joint product o the ASEAN Secretariat and the World Bank East Asia and Pacic Region. It was made

possible by the nancial support o the Australian Government. The report is part o an eort to monitor and understand

vulnerability and labor markets in ASEAN Member States during the global economic crisis. A snap-shot report was

initially produced as background material or the 15th ASEAN Summit in Cha-am Hua Hin, Thailand, on 23-24 October

2009, based on in-country consultations in ASEAN Member States during August and September 2009, as well as

on social impact monitoring o the global nancial crisis by the World Bank. The snapshot version o the report

was elaborated ollowing the 15th ASEAN Summit into this regional report, together with detailed country studies.

The country studies are presented as attachments to this regional report. They have been nalized based on urther

consultations with Member States in December 2009. Special thanks go to ASEAN Member States or acilitating

consultations and providing the teams with essential inormation to produce this report. A list o counterparts who

contributed to country consultations is provided in Annex 1. Thanks also go to the ASEAN Secretariat and World Bank

Country Oces or coordinating country visits and providing guidance throughout the process.

ACKNOwLEDGEMENTS


8/68

viii


9/68

ix

Frerd v

Acnledgements vii

List Areviatins x

Executive Summar 1

Sectin 1: Mnitring te Scial Impacts te Glal Financial Crisis 5

I. The Global Economic Slowdown in ASEAN Member States 5

II. Labor Market Impacts 7

III. Migration and Remittances 12

IV. Health Outcomes and School Enrollment 13

V. Poverty and Welare 14

Sectin 2: Scial Prtectin Plic Respnses t te Crisis 19

I. Introduction 19

II. Recent Evolution o Social Protection Systems in ASEAN 20

III. Social Protection Responses to the Crisis 23

Overall Social Protection Responses 23

Types o Social Protection Responses 26

Cross-cutting Issues in Social Protection Policies and Programs in ASEAN Member

States34

IV. Conclusions and Recommendations 38

Reerences 43

Annex 1: ASEAN Memer State Cnsultatins: List persns met 46

Attacments: Cuntr Reprts (availale in CD)

TABLE OF CONTENTS


10/68

x

ACSS ASEAN Community Statistical Systems

AHSOM ASEAN Heads o Statistics Ofces Meeting

ALMP Active Labor Market Policies

AMS ASEAN Member States

ASEAN Association o Southeast Asian Nations

ASEC ASEAN Secretariat

AusAID Australian Agency or International Development

BLT Bantuan Langsung Tunai Program o Indonesia

CCT Conditional Cash Transers

CLEEP Comprehensive Livelihood and Emergency Employment Program

CMR Crisis Monitoring and Response

CPF Central Provident Fund

FDI Foreign Direct Investment

FFF Food, Fuel and Financial

GDP Gross Domestic Product

GFC Global Financial Crisis

HEF Health Equity Funds

ILO International Labor Organization

IMF International Monetary Fund

KDP Kecamatan Development ProjectLFS Labor Force Surveys

MOLISA Ministry o Labour, Invalids and Social Aairs o Vietnam

NFA National Food Authority

OECD Organization or Economic Cooperation and Development

OFw Overseas Filipino workers

PNPMProgram Nasional Pemberdayaan Masyarakat Mandiri (National Program or Community Empoerment

o Indonesia)

PMT Proy Means-Test

RIP Rural Inrastructure Program

SI Social Insurance

SP Social Protection

SLOM Senior Ofcials Meeting on Labour

SOMDP Senior Ofcials Meeting on Development Planning

SOMSwD Senior Ofcials Meeting on Social welare and Development

SOMRDPE Senior Ofcials Meeting on Rural Development and Poverty Eradication

UCT Unconditional Cash Transers

UPP Urban Poverty Project

wB world Bank

wTO world Trade Organization

LIST OF ABBREVIATIONS


11/68

1

ExECUTIVE SUMMARY

Te eects te glal ecnmic sldn n peples emplment and earnings are at te cre its scial

impacts, maniested largel trug increased j curning and reduced rer earnings in crisis-aected

sectrs, as ppsed t idespread pen unemplment. However, the primary labor market impacts have had a range

o secondary eects on household welare beyond urban areas and the ormal sector through actors such as reduced

remittances. Although there are recent signs that the slowdown appears to have bottomed out in AMS, the collapse

o aggregate demand is still working its way through the global economy, and the labor market and social impacts o

the crisis are likely to linger or some time. Evidence rom previous crises, including rom other regions, indicates that

social and poverty indicators oten rebound more slowly than economic indicators.

Te glal ecnmic sldn as een transmitted t ASEAN ecnmies trug several cannels, including:

decreased global demand or the Member States manuactured exports; declines in key commodity prices; lower

levels o oreign direct investment (FDI); ewer tourist arrivals; and reduced remittance receipts rom overseas

migrants, as described below. Migrant and low-skill laborers in these sectors, who generally have little access to

ormal social saety nets, appear to be most aected. The impact on health and school enrollment remains largely

unknown, but evidence rom the 1997-98 nancial crisis in East Asia suggests a range o possible negative impacts

on human capital investment, which can have long-term consequences on peoples uture productivity and inter-

generational transmission o poverty. The challenges o getting data on a range o social indicators during the crisis

has highlighted the need or strengthening the capacity o national and regional statistical systems to generate timely

and good quality inormation to monitor social impacts and inorm policy responses.

Te cumulative impacts te d, uel and nancial crises ave given increased impetus t scial prtectin

plicies and prgrams in ASEAN. Social protection (SP) interventions have had a signicant role in many o the

stimulus packages. Given that the primary transmission channels have been through employment and earnings, labor

market interventions to cushion the impact on employment have been an important crisis-response instrument in most

ASEAN Member States. In higher and some middle income ASEAN Member States, the ocus has been on active labor

market interventions, including wage and tax incentives to retain or hire workers, training subsidies, etc. In lower and

some other middle income Member States, decentralized stimulus measures such as small inrastructure investmentsand community grants have been used to generate jobs and act as a productive saety net. Apart rom labor market

interventions, programs to supplement household incomes have also been adopted in a number o Member States,

with cash transers o various types playing an important role in the SP policy response. While ood subsidies have

played a role in the global nancial crisis (GFC) response, overall there has been somewhat less reliance on them,

partly due to the enhanced role o cash transers. Social insurance, on the other hand, has not or the most part been

as signicant a crisis response tool as other pillars o social protection, though it has played a useul role in Member

States which had signicant pre-existing coverage o social insurance.


12/68

2

Te glal sldn als demnstrates pre-existing SP prgrams sape te set Memer States easile crisis

respnse ptins. Where Member States have robust SP systems in place pre-crisis, it is easier to inject incremental

unding and to target specic groups in crisis response. However, where SP programs have limited coverage and

delivery systems are under-developed, ramping them up when crises hit is very challenging. This highlights the

importance o having solid blocks upon which to build the SP crisis response, particularly in areas such as targeting

mechanisms, institutional capacity and coordination, and adequate social impact monitoring.

Te glal ecnmic sldn and te preceding d and uel crises ave rugt me tat scs are a act

lie in ASEAN, and SP sstems ill increasingl need t deal it ecnmic vlatilit even in nn-crisis perids.

For Member States which place a premium on competitiveness, the social risk management and eciency enhancing

unctions o social protection systems are central, in addition to their equity unction. What remains to be seen is the

extent to which the increased priority given to social protection programs in crisis response represents an emerging

structural shit in their uture role in ASEAN economies. The nature o responses in some ASEAN Member Statessuggests that this could be the case, particularly with respect to the role o cash transers.


13/68

3

SECTION 1


14/68

4


15/68

5

I. ThE GLobAL ECoNoMIC SLowDowN IN ASEAN

In cntrast t te East Asian crisis a decade ag, te current glal ecnmic sldn riginated nt in te

regin ut in te igl industrialized cuntries. OECD countries were not only the rst but also the hardest hit

by the crisis, with GDP growth or advanced economies expected to contract by 3.4 percent on average in 2009.

As a region, East Asia has ared relatively well during the global slowdown, led by high growth in China which is

projected to grow by 8.5 percent or 2009 as a whole. For developing Asia, growth is expected to slow somewhat

rom 7.6 percent in 2008 to 6.2 percent in 2009 beore rebounding to 7.4 percent in 2010. For the ASEAN-5, the eects

o the GFC are more pronounced, with average growth alling rom 4.75 percent in 2008 to only 0.72 percent in 2009,

but also rebounding in 2010 (IMF, 2009).

Te extent t ic ASEAN Memer States (AMS) ave een aected te crisis varies cnsideral. While

Member States such as Indonesia and Vietnam are maintaining positive (although lower than pre-crisis) growth

rates, Singapore, Thailand, Malaysia, and Cambodia have been aected more signicantly (Figure 1) 1. Although

recent signs indicate that the global slowdown may have bottomed out and that most economies in the region have

begun to rebound, the World Trade Organization (WTO) has cautioned that the world economy remains ragile and

the economic outlook remains uncertain. The collapse o aggregate demand is still working its way through the

global economy, and the social impacts o the crisis will most likely linger or some time, even i macroeconomic

indicators show signs o recovery (WTO, 2009).

Te glal ecnmic sldn as een transmitted t ASEAN ecnmies trug several transmissin

cannels. Among the most important are: decreased global demand or the Member States manuactured exports;

declines in key commodity prices; lower levels o oreign direct investment (FDI); ewer tourist arrivals; and reduced

remittance receipts rom overseas migrants, as described below. In addition to the impacts o the GFC itsel, natural

disasters have been compounding shocks in a number o AMS.

Manuacturing exprts. In the rst seven months o 2009, manuacturing exports declined by 30 percent in

Vietnam and by almost 40 percent in Indonesia, compared to the same period in 2008. Demand or semiconductorsand electronics manuactured in the Philippines also declined as much as 60 percent in the rst quarter o 2009.

In Malaysia, the contraction in manuacturing continued, albeit at a slower rate o minus 14.5 percent year-

on-year in the second quarter o 2009, compared with minus 17.9 percent in the rst quarter. In Singapore,

electronics exports contracted by 14 percent in August 2009, ater a 15 percent decline in the previous month

(year-on-year). In contrast, in a recent rm survey in Lao PDR, manuacturers indicated that the nancial crisis

1 Cambodia is not shon in Figure 1 due to lack o quarterly data on GDP groth. Hoever, recent projections by the world Bank and other multilateral

organizations suggest that the Cambodian economy is likely to contract by beteen 1.0 and 1.5 percent in 2009.

MONITORING THE SOCIAL IMPACTS OF



16/68

6

only mildly aected their business, and roughly 60 percent expected an increase in revenues or the next

nancial year.

Cmmdit prices. The average prices o crude petroleum, copper, palm oil, coee, and rice have allen by

15-30 percent since the start o the crisis and through the second and third quarters o 2009. Myanmar has

experienced a decline in the value o exports o natural gas due to decreasing global demand and alling

commodity prices. The value o natural gas exports to Thailand, its biggest trade partner, is expected to drop by

50 percent year-on-year.

Freign Direct Investment. In Cambodia, FDI decreased by nearly 50 percent in the rst quarter o 2009 compared

to the same period in 2008, resulting in signicantly lower investment in the construction sector. Thailand aced

a 22 percent contraction in FDI in the rst quarter o 2009 (year-on-year).

Turism. During the rst seven months o 2009, oreign tourist arrivals ell by 14 percent in Indonesia and 20

percent in Vietnam, compared with the same period in 2008. In Cambodia, the number o tourists declined by 2.2

percent in the rst two months o 2009 compared to a year earlier.

Remittances. Worldwide, international remittances are estimated to have allen by 7-10 percent (World Bank,

2009b). In Indonesia, remittances are expected to decline to US$3 billion in 2009, rom US$6 billion in 2007

(Koser, 2009). Qualitative assessments o migrant workers rom Myanmar and the Philippines also suggest that

the requency and size o remittances sent home were declining in recent months, though recent estimates


17/68

7

indicate that overall remittance growth in the Philippines remained low but positive in the rst hal o 2009

(BSP, 2009).

Te lling sectins prvide an vervie te scial impacts te glal sldn in ASEAN, using

inrmatin rm existing data surces and nel cmmissined rapid surves in selected ASEAN Memer

States (AMS). Existing data collection systems only imperectly capture the social impacts o the crisis. While data

collection systems or macroeconomic indicators in most AMS are well-developed, systems and methodologies

or collecting social indicators generally lag behind. There are issues with the range o indicators collected, the

periodicity and timeliness o such data, and in many cases with the quality o social data. This section brings

together the best available evidence rom existing micro surveys, administrative data, simulation analyses, and

qualitative assessments undertaken by various agencies since the start o the global slowdown. It rst reviews the

overall labor market impacts o the slowdown, then ocuses on the impacts on migration and remittances, which

have had a pronounced eect on households in the region. It then examines the impacts o the slowdown on healthand education and on poverty and welare.

II. LAboR MARkET IMPACTS

Te eects te glal sldn n emplment and earnings, ic are at te cre its scial impacts, are

ccurring largel trug increased j curning and reduced rer earnings in crisis-aected sectrs rater

tan trug lesale j lss and pen unemplment. In most AMS, workers are acing increased prospects

o underemployment, lower earnings and income, shits rom ormal to inormal employment, greater job insecurity,

and in some cases, higher open unemployment. Qualitative studies also indicate that households are experiencing

greater hardships due to lost jobs and incomes. Even in AMS that on average are still acing positive growth rates,

the shit in sources o growth rom outward- to more domestic-oriented sectors is aecting laid-o workers, since

nding employment in new sectors is otentimes dicult. This higher churning in and out o dierent jobs, which

can have serious impacts on worker and household welare, is only imperectly captured by standard employment

and unemployment statistics, which oten show little up and down movement.

Accrdingl, natinal-level unemplment gures s little cange in mst Memer States. For example,

between June 2008 and June 2009, Malaysias unemployment rate increased rom 3.5 to 3.6 percent, and Thailands

open unemployment changed rom 1.2 to 1.8 percenta 50 percent increase, although the level o unemployment is

still low compared with unemployment levels in the region and elsewhere. Unemployment is however sometimes

a poor indicator o labor market perormance, particularly in low- and middle-income Member States with large

inormal sectors and where enorcement o regulations is limited, as appears to be the case or most o ASEAN.

hever, particular pcets cuntr ppulatins ave een aected mre severel. These pockets are oten

concentrated in export-oriented sectors. The manuacturing, mining, and tourism sectors, as well as estate arming,

have all been hit hard. Declines in FDI have also hit sectors such as real estate and construction. Qualitative

studies show that workers in aected sectors oten cope with short-term unemployment by taking on part-time

jobs elsewhere with inerior pay, engaging in sel-employment activities, or sometimes by going back to rural areas.


18/68

8

Accordingly, the share o sel-employed workers and casual laborers has increased relative to ormal, contracted

wage employees since the start o the global slowdown.

Migrant and l-silled larers, r n cntract r as da larers, appear t e aected mst te

sldn. The rapid qualitative assessments suggest that enterprise managers are very reluctant to re skilled

workers, anticipating costs and delays in recruiting skilled labor when production will eventually increase, although

skilled workers have commonly ound themselves working ewer hours during the crisis. At the same time, sel-

employed workers in AMS aected by the slowdown have seen their revenues decline signicantly. Small rubber

plantation owners in Indonesia, or instance, are experiencing signicant income declines due to the sharp all

in rubber prices, which has had a visible impact on local shop owners and truck drivers who provide an array o

services to plantations.

h te crisis as aected men and men depends n te extent emale and male rers invlvement in

crisis-aected sectrs. In Cambodia, both the garment and construction sectors have been aected signicantly.

While the majority o garment sector workers are emale, most construction workers are male. The shares o

male and emale workers in crisis-aected sectors are not that dierent: 6.7 percent o jobs held by women,

compared to 9.9 percent o jobs held by men, are in garments, construction, and tourism, the three most vulnerable

sectors. However, while the crisis may have aected male and emale workers similarly, laid-o women may

ace greater risks. Qualitative studies indicate that some women in crisis-aected sectors have ound work in

the entertainment sector or other orms o employment that entail a high risk o sexual harassment. The risk o


19/68

9

moving toward more hazardous orms o employment highlights the need to develop eective saety nets targeted

at vulnerable groups.

Mst served j lsses in te regin ave een in manuacturing, ile mst j creatin as een in te

inrmal sectr (e.g., sales and trade) r in te pulic sectr. Workers who lose their jobs in the manuacturing

sector, especially those with low skills, are unlikely to nd jobs in well-paid sectors. Nonetheless, the data suggest

that most laid-o workers appear to have ound other work, even i those jobs are less remunerative. In Thailand,

about 600,000 jobs were lost in the manuacturing sector, but almost 800,000 people ound work in wholesale and

retail trade and in the accommodation and catering sectors, where inormal activities dominate (Figure 3). In the

Philippines, about 100,000 people lost their job in the manuacturing sector, but almost 800,000 people ound work

in the agriculture and wholesale/retail sectors.2

Te emplment impacts te sldn are als ten lcalized gegrapicall and var accrding t te

lcal ecnmic envirnment. In Indonesia, or example, the shocks are most likely to be elt among a ew regions

that produce commodities such as estate crops, or which world prices have dropped signicantly. Regions

where export-oriented actories, such as garment actories, are located are also more likely to eel the impacts

(Figure 4).

2 Hoever it has to be noted that the statement that orkers ho lost their jobs in the manuacturing sector ound ne employment in another sector cannot

be proved decisively ithout panel data. Evidence rom previous crisis suggests that more people enter the labor orce as unpaid amily orkers to support the

amily income. On the other hand, laid-o orkers might leave the labor orce as discouraged orkers. The degree o churning - i.e. orkers changing status

o employment, the sector o employment etc. - can only be quantifed using panel data.


20/68

10

In additin t aecting te qualit js, te glal sldn is als aecting real ages in manuacturing

sectrs. In most AMS, wages in manuacturing sector declined to some degree around the time that the global

slowdown began, in the ourth quarter o 2008 (Figure 5). For Thailand and Malaysia the eect on wages in the rst

quarter was very modest and wages quickly bounced back to pre crisis levels and even higher in the second quarter

o 2008. Wages in the manuacturing sector o Cambodia declined since mid 2008 but seem to have recovered in

2009. However, real wage declines may have been due to high infation in 2008 caused by the ood price crisis that

aected Cambodia particularly hard. Qualitative assessments suggest that rms have been implementing a range

o temporary arrangements to lower their overall wage costs, including eliminating overtime and oering workers

ewer hours; placing workers on leave without pay; and reducing bonuses and non-wage benets. In some sectors,

rms have also cut wages. Qualitative assessments also reveal that daily rates or skilled construction workers

and porters in Thailand had declined in May by about 60 percent compared to one year prior, while piece rates or

sewing workers had declined by 25 percent.3

overall, adjustments in emplment are similar t at East Asia experienced during te crisis a decade ag.

Although the nature o the crisis was dierent, it is instructive to look at some o the outcomes and compare

it to the recent economic crisis. Lucas and Fallon (2002) summarize the eects o the East Asian crisis 1997/98

on Indonesia, Thailand and Malaysia. Their results are airly similar to what we can observe in East Asia right

now. Total employment rose in almost all countries even during the crisis, but there appears to be a considerable

dierence between sectors. For example employment in the construction and manuacturing sectors decreased

considerably whereas more people ound jobs in agriculture and some o the service industries. This is similar to

what is happening today, particularly or the services sector. In the case o agriculture, the shit is not always as

evident. During the 1997-1998 crisis, the sudden depreciation o the real exchange rate may have provided a bigger

incentive to increase production o cash crops. While data are not yet available to determine precisely what is

happening in the agriculture sector ollowing the recent crisis, qualitative evidence suggests that agriculture is

3 Follo-up ork has been commissioned to see i ages have rebounded since then.


21/68

11

acting more as a saety net in those ew cases where urban migrant workers chose to return, providing unemployed

urban workers and their rural amilies with additional means to cope with the reduced household income. However,

reverse migration seems to have been less important in this crisis than previously. Evidence rom Indonesia also

suggested that the degree o churning a decade ago was higher in rural than in urban areas. In particular, only

41 percent o rural men and 50 percent o women who had reported not working in 1997 were still out o work in

19984, suggesting that many o those aected by the East Asia crisis had ound alternative albeit inerior earning

opportunities.

Unlie te current crisis, ig infatin during te East Asian crisis 10 ears ag signicantl erded real ages.

During the East Asian crisis, Indonesia experienced high levels o infation. As a result, real wages especially

or employed workers ell by close to 44 percent during the crisis period. Although the decline in real wages in

Malaysia was not as severe as in Indonesia, it ollowed a period o strong real wage growth and thereore also elt

in signicant ways by workers. In particular, deep real wage cuts seem to have contributed to limit the impact oninequality ollowing the 1997-1998 crisis in Indonesia, whereas in countries such as Thailand, higher unemployment

rates and smaller wage cuts led to higher inequality 5. While infationary pressure is not a major eature o the crisis

today, the qualitative evidence above suggests that many employers have ound ways to cut real wages in order to

lower production costs without laying o workers, so that the end eect could be similar to that o ten years ago.

4 Fallon and Lucas (2002).

5 Ibid, p.34


22/68

12

III. MIGRATIoN AND REMITTANCES

Migrants are ten amng te rst t e it ecnmic dnturns, and tis crisis is n exceptin. Both overseas

and domestic migrants report reductions in hours worked, incomes, and related benets. In Thailand, workers rom

Myanmar report signicant reductions in earnings and ewer available jobs. Families o Oversees Filipino Workers

(OFW) stated in ocus group discussions that their relatives in Taiwan, Japan, the Middle East, and Singapore

have suered rom reduced overtime pay, benets, and working hours (Pulse Asia, 2009). Declines in incomes

vary according to the countries in which migrants work. Middle Eastern countries have not been as badly hit as

Singapore and Japan, or example.

Reduced remittances are aving in sme cases detrimental eects n t useld elare and te lcal

ecnm. Worldwide international remittances are expected to decline by 7-10 percent in 2009, and the downward

trend is also aecting East Asian countries: In Indonesia, or instance, remittances are projected to decline to

US$3 billion in 2009 rom US$6 billion in 2007 (Koser, 2009). Interviews with urban/industrial workers in Vietnam,Thailand, and Cambodia highlight the hardships aced by those receiving lower remittances. These workers report

that reducing or eliminating remittance payments can sometimes reduce receiving amily incomes by as much as

75 percent. This reduction not only aects daily consumption but also makes it harder or amilies to save money or

insure themselves against various risks (health, agricultural production, etc.). Eliminating these payments can thus

have important longer-term impacts on amilies livelihoods and well-being. Moreover, declining remittances can

also aect the local economy by reducing household purchasing power.

wile te glal sldn as led t reduced fs remittances, it des nt appear t ave generated

extensive fs return migrants. During the 1997/98 East Asia crisis, many workers coped with job and income

losses by returning rom urban areas to their amilies and to rural areas, though systematic evidence on internal

migration ollowing crises is scarce. Qualitative evidence suggests that migrants now preer to look or alternative

employment around their current residence rather than to return to their amilies. Focus group discussions in

several countries indicate that workers see return migration as a solution o last resort, preerring to seek work in

sectors that continue to experience growth.

hever, te glal sldn culd induce a sit rm legal t illegal crss-natinal migratin, as sme

Gvernments ave initiated prgrams t limit migratin as part teir crisis respnse measures. Some AMS

are taking measures to reduce the infow o migrant workers rom abroad. In Thailand, where an estimated 1.8

million workers rom neighboring countries are currently employed, no additional work visas will be issued in

2009, nor will existing visas be renewed. Reports rom Malaysia also suggest that the Government has reduced

the number o available work permits, reduced the duration o work permits or some workers rom six to three

months, and encouraged rms to retrench oreign workers and hire unemployed domestic workers instead. Such

programs could result in increased numbers o illegal oreign workers while potentially decreasing the number o

legal migrants (Fix et al, 2009).


23/68

13

IV. hEALTh oUTCoMES AND SChooL ENRoLLMENT

Altug te impacts te glal ecnmic sldn n ealt utcmes and scl enrllment remain largel

unnn, evidence rm past crises suggests tat te ptential impacts culd e serius. A general lack o high-

requency data on social sector indicators makes it dicult to determine the impacts o the current slowdown

on health and education at this stage. However, evidence rom the 1997-98 nancial crisis in East Asia points to a

range o possible impacts on human capital investment, at least in the most aected countries. The evidence on

changes in household spending behavior is limited, but an analysis o the Indonesia Family Lie Survey ollowing the

1997-98 crisis showed that amilies reduced health care expenditures as a coping strategy in both rural and urban

areas6. Also in Indonesia, school enrollment ell among the poorest ater 1997, and inant mortality was ound to

have increased by over 3 percentage points. Such losses in human capital can become irreversible i appropriate

policy actions are not taken. Health and education outcomes should thereore be monitored closely in vulnerable

regions and among vulnerable groups.

Mst AMS appear t ave een ale t prtect levels pulic spending n te scial sectrs, elping preserve

access t and utilizatin asic scial services. While data on public sector spending is not yet available or all

AMS at this stage, comparison with the 1997-98 situation oers useul insights. A decade ago, public expenditure

on health changed little relative to GDP in Indonesia, Malaysia and Thailand, indicating that in absolute terms

spending was lower at a time where households disposable income was also shrinking and in Indonesia, use

o health acilities ell dramatically between 1997 and19987. Public expenditure on education actually ell relative

to GDP in Indonesia and Malaysia, but increased in Thailand. Again, in Indonesia, enrollment rates declined and

dropout rates increased, particularly among youths aged 13-19. During the current crisis most Government seemed

better prepared to protect real social spending, due to better scal stances and lower infation. In Indonesia, the

Government included increased education sector spending as part o its stimulus package and in Thailand the

provision o ree education was extended rom 12 to 15 years. In the Philippines, the Government planned additional

investments or primary and secondary hospital acilities and allocated resources to recruit additional health care

workers or these hospitals.

Accrdingl, existing assessments did nt nd majr impacts n ealt and educatin utcmes, altug

sstems ave een put under strain. In Indonesia, some crisis-aected amilies have started to go to public clinics

or ree treatment even i such clinics are arther rom their homes than the private clinics they visited previously.

They are also using more over-the-counter medicines rather than seeking proessional treatment. In Cambodia,

cyclo drivers, mostly rom rural areas, have no health insurance, and the cost o health services is beyond their

means. They conrmed in FGD that, as a result, i they become ill, they commonly return to their homes in rural

areas where they treat themselves with traditional medicines.

6 Ibid., p.39

7 Ibid., p.37


24/68

14

V. PoVERTy AND wELFARE

As a result te glal ecnmic sldn, te pace pvert reductin in AMS, tug still psitive, is

expected t all (see Figure 6)8. Projections indicate that 14 million ewer people will be pulled out o poverty in 2009

than originally anticipated beore the slowdown. The poverty reduction orgone, measured in numbers o people,

is expected mostly in the emerging middle-income countries. In the most aected AMS, the absolute number o

poor people is actually expected to increase: Cambodia, and Thailand will likely ace absolute increases in poverty

due to a contraction in per capita income. With its higher poverty rates Cambodia is the country with the largest

projected increase in poverty, which is expected to rise by slightly more than 200,000 people, raising the headcount

ratio rom 28.6 to 29.6 percent based on the $1.25 poverty line.

wile te glal ecnmic sldn ill ave serius impacts n te elare millins peple in te

regin, it is unliel t reverse te prgress made AMS in reducing pvert in recent ears. Since 1990, the

share o East Asias population living in absolute poverty (less than US$1.25/day) has declined rom 55 percentto less than 10 percent o the population. Notably, the more severe 1997-98 nancial crisis, although it aected

temporarily the pace o poverty reduction, did not aect subsequent progress in poverty reduction. However, the

speed with which AMS will be able to recover rom the current slowdown remains questionable and may still aect

the regions poverty reduction eorts.

Neverteless, it is imprtant t interpret it cautin te estimates aggregate pvert impacts en

cnsidering te ider elare eects te crisis in ASEAN. Firstly, because the slowdown is aecting people

o all income levels, the relatively small impacts on poverty are likely to underestimate the overall welare impact

o the slowdown. Secondly, the poverty headcount estimates do not refect the ull reality o the social impacts

on a range o non-income indicators. Thirdly, as noted above, the impacts on specic groups o the population,

locations, and sectors (e.g., garments, construction, and tourism) are ar more pronounced and hidden in aggregate

poverty estimates.

8 These estimates must be interpreted ith caution. The poverty headcount as estimated based on groth projections, hich do not take into account sectoral

eects o employment that are important during crises. In-depth analyses using recent and upcoming household survey data ill be needed to shed more light

on the poverty and social impacts o the slodon.


25/68

15

During te East Asian crisis 10 ears ag, pvert reductin rates sled dn trugut te regin and unced

ac immediatel t pre crisis levels. Between 1997 and 98 the pace o poverty reduction in East Asia slowed

down considerably. However, in line with the observations o the recent economic crisis, dierent countries were

aected quite heterogeneously. In Indonesia, poverty rates in both rural and urban areas almost doubled between

1997 and 1998, whereas in Thailand the increase in poverty was elt mainly in rural areas. Also because o the strong

currency depreciations in East Asia during and ater the crisis a decade ago, exports rom the region boomed. This

was one o the major reasons why GDP growth and the pace o poverty reduction bounced back immediately.


26/68

16


27/68

17

SECTION 2


28/68

18


29/68

19

I. INTRoDUCTIoN

Te main scial impacts te crisis in AMS t date ave trug j curning and reduced rer earnings,

it secndar impacts n remittances and useld incmes. Government responses through stimulus

packages and additional SP programs thereore largely refect the need to generate/preserve employment and

protect vulnerable populations. The previous section has shown how the eects o the global economic slowdown

on peoples employment and earnings are at the core o the social impacts o the crisis. The nature o a shock

determines to a large extent which social protection measures are best suited to address the social impacts o the

crisis. Price shocks, such as those experienced during the ood and uel crisis, and employment shocks such as

those described in this note, are best addressed through dierent program responses. So countries that had set

up programs recently to respond to the ood and uel crises, did not necessarily have programs in place to address

the employment shock associated with the current crisis. This section summarizes the various social protection

measures undertaken in the AMS in response to this labor market shock and in the context o the pre-existing

programs available.

Te plicies and prgrams used AMS and discussed in tis sectin refect a rad denitin scial prtectin.

Governments responses varied in size and scope, oten as a result o capacity or scal constraints, but tended

to draw rom a relatively common menu o options. Overall, these policies and programs may be categorized as

protective (to protect those already in or alling into poverty), promotional (ocused on enhancing employment and

livelihoods and promoting sustained movement out o poverty) and preventive (mainly social insurance measures

providing ex ante protection against both unpredicted and lie cycle shocks), as described in Box 1. The discussion

does not ocus on disaster/relie interventions, although these are clearly important in ASEAN.

SOCIAL PROTECTION POLICY RESPONSES TO CRISIS


30/68

20

II. RECENT EVoLUTIoN oF SoCIAL PRoTECTIoN SySTEMS IN ASEAN

Despite sme variatin acrss Memer States, pre-crisis SP appraces in ASEAN sared sme cmmn

eatures tat ere quite distinct rm tse in ter regins. Historically, AMS had a vision o the role o the State

in social protection that diered rom that in other regions such as Latin America or Eastern Europe and resulted in

signicantly lower SP spending as a share o GDP (Figure 7). Although SP systems did vary across AMS refecting

the diversity o country income levels, administrative capacity, poverty proles, labor market eatures, and other

actors (Box 2), common eatures were:

Astrongemphasisonself-reliancethroughemploymentandonfamilyandotherinformalsupportnetworks.In

this context, the state was viewed as the provider o last resort.

Publicspendingthatfocusedmoreonhumancapitalacquisitionthanwelfaretransfers.

Concentrationofsocialinsurancecoverageinthecivilserviceandformalprivatesectors,contributingtolow

coverage in most ASEAN economies where inormal employment was dominant. FormostMemberStates,fairlyunder-developedornascentsocialassistanceprograms,witha historically

signicant role or ood subsidies in a number o them. More recently, there has been greater divergence o

social assistance program coverage, as a result o some AMS increased reliance on cash transers.

over te past decade, mst AMS ave undertaen rerms t strengten teir SP sstems. General trends in SP

reorms in ASEAN include:

Initiatives to expand coverage o core social insurance (SI) programs. Some Member States (e.g., Thailand

and to some extent Vietnam with health insurance) have achieved notable success in expanding specic types

o social insurance to a wide population in both the ormal and inormal sectors. Other AMS (e.g. Indonesia)

are aiming or wide coverage o SI but are still developing strategies to achieve the goal. Others remain at a

more nascent stage, with policy still ocused on the civil service and ormal sector workers (e.g., Cambodia

and Lao PDR), but with parallel interventions to increase access to basic services or the poor (e.g., Health


31/68


32/68

22

interest in income generating programs, including skills development and small business credit/interest

subsidy programs.

Adoption o anti-poverty programs that utilize an integrated rural development approach, particularly in

low-income and rural settings. Programs oten have a strong inrastructure emphasis and requently give

communities a role in prioritizing program activities with a ocus on multi-dimensional support to poor areas

rather than on targeting poor people within those areas. Notable examples include programs in Lao PDR,

Indonesia, Thailand, and Vietnam. The policy balance between targeting poor areas and poor people is

one that is evolving in a number o AMS, as it has also in neighboring countries such as China. While the poor

area approach to poverty alleviation has appeal in an environment where household level targeting is under-

developed, it is likely to have signicant limitations also, especially or the emerging poor such as those in

urban areas, or who live outside designated poor areas but are subject to localized shocks.

Increased use o saety nets such as scholarships, ee waivers and conditional cash transers (CCTs) to

protect access to education. Indonesia has a relatively new school assistance program that provides grants to

schools in exchange or their reducing or eliminating ees or the poor. Thailand has been promoting compulsory

ree education through educational subsidies rom kindergarten to secondary school. There have also been

eorts in lower income AMS such as Cambodia through its Scholarships or the Poor program or secondary

education. These are refections o wider interest in ASEAN countries in strengthening the linkages between

social protection programs and human capital acquisition. The CCT program o the Philippines introduced in

2007 is a good example o the use o such instruments, with a ocus on health, nutritional and educational

conditionalities.

Greater eorts to ensure health coverage or the poor. Recognizing the importance o health shocks as a actor

driving households into poverty, or keeping them there, most AMS have sought to expand the coverage o

aordable health services or the poor, both through insurance and other models. In 2001, Thailand introduced

a health insurance scheme guaranteeing health insurance to all people without social security benets, at a

cost o 30 Thai baht per treatment. Indonesia distributed health cards entitling 16 million amilies to ree health

care at public clinics and hospitals in 2005, and has committed also to expanded health insurance coverage in

a 2004 law. In Cambodia and Laos, Health Equity Funds (HEFs) are being used to improve access or the poor.

HEFs are managed through cooperative arrangements between the Government, development partners, and

local NGOs.

Increased emphasis on improving targeting systems. Eorts to improve targeting to ensure better protection

o the poorest, control scal outlays on SP programs, and improve transparency o program administration

have included proxy means testing (e.g., Indonesia BLT; Philippines CCT) and participation o communities in

identiying the poor (e.g., Cambodias IDPoor program, the MOLISA poverty listing in Vietnam).


33/68

23

III. SoCIAL PRoTECTIoN RESPoNSES To ThE CRISIS

overall Scial Prtectin Respnses

Te Asian crisis 1997-98 revealed a numer limitatins in te scial prtectin plic ramers and respnse

capacities AMS ic te ave addressed t varing degrees in te intervening ears. The comprehensiveness

and robustness o social protection mechanisms at the onset o this crisis varied signicantly and the available

instruments did not always ocus on those most immediately aected by the current crisis, including the urban poor

and vulnerable, laid-o workers, and more broadly those in the inormal sector. O particular note were the limited

coverage o social insurance in most AMS and the low prevalence o a core social assistance program which was

broadly accessible to the poor, independent o location, social category, or other actors.

Scial prtectin interventins ave ecme mainstreamed as part te arsenal t cunteract te current crisis

and teir apprpriateness seems less cntested tan in previus crises9

. In most AMS, the cumulative impactso the recent ood, uel, and nancial crises have given increased impetus to SP policies and programs, albeit to

varying degrees depending on the country context. For AMS that have introduced economic stimulus packages,

SP interventions are a signicant part o the overall packages in most cases, though both the size o the overall

packages relative to GDP (ranging rom around 1.5 percent o GDP in Indonesia and Thailand to over 8 and 9 percent

in Vietnam and Malaysia, respectively) and the share o SP as a share o the total stimulus vary substantially.

For example, Vietnam in its initial package and Indonesia have dedicated the bulk o stimulus spending to tax

reductions and interest subsidies o dierent orms, complemented by a sizeable injection o unds into household

and community based SP measures. Singapore has ocused on active labor interventions and tax and credit

programs to help rms weather the crisis, while Thailand ocused the rst wave o stimulus on direct transers

and incentives to sustain human capital investments. Most stimulus packages have also included inrastructure

investments which vary in labor intensity and the degree o direct employment generation. Some Member States

such as Thailand, Vietnam and Malaysia have had more than one round o stimulus, with the relative importance o

SP interventions shiting between rounds.

wile nt all AMS ave intrduced explicit stimulus measures in respnse t te GFC, cncern it te scial

impacts te recent crises as underscred te imprtance SP. To date, several AMS (including Lao PDR,

Myanmar, Cambodia, and Brunei Darussalam) have not introduced explicit stimulus packages in response to the

global nancial crisis, but have nonetheless strengthened SP programs to mitigate its impact.

Te e SP interventins t pre-crisis and incremental measures in respnse t te GFC acrss AMS are

summarized in Tales 1 and 2. AMS with concerted stimulus eorts have drawn rom a airly common menu o SP

options, but with diering emphasis between types o interventions. As noted, given the sequential nature o the

recent crises unolding in the AMS, it is not always clear whether policies should be categorized as responses to the

global nancial crisis specically or as carryover policies rom the uel and ood crises. Thereore, SP responses to

9 It is also important to note that draing a precise line beteen hat can be characterized as social protection or other types o responses is not alays

possible (e.g., or area-based stimulus measures).


34/68

24

the ood, uel, and nancial crises should be viewed as an evolving response rather than entirely distinct measures.

In addition, Table 1 does not attempt to distinguish the relative degrees o coverage and adequacy in dierent

programs, so there is considerable heterogeneity across Member States listed as having certain programs (e.g.

or cash transers, ranging rom social assistance benets accessible to all poor citizens to benets conned to

specic social groups or locations).

Despite te radl cmmn menu SP respnses utlined, tere ave een ntale variatins acrss AMS

su-grupings in te relative empasis n dierent plicies and te intensit scal ert. The menu o

response options outlined has been to a signicant degree constrained by the availability o specic SP programs

and coordination mechanisms in place in individual Member States when the crisis hit. Because o the limited

reaction time when crises hit, existing programs are oten the natural vehicle or immediate response and thereore

infuence the menu o policy options. The easibility o dierent SP responses is also linked to the pre-existing

degree o ormality o labor markets in dierent AMS. While there are a number o commonalities in SP crisisresponse by AMS, some notable sub-regional patterns are also evident, including:

the upper-middle to higher income AMS (Singapore,Malaysia, and Thailand, though notably not Brunei

Darussalam) have placed a relatively greater reliance on direct labor retention and job stimulation measures,

relying primarily on tax and subsidy systems. Such emphasis has been greatly acilitated by enterprise and

individual tax systems which have wider coverage o workers, and hence allow or adjustments at the margin

in tax/subsidy treatment and social insurance entitlements on a temporary basis.

themiddleincomeAMS(Indonesia,Vietnam,andtosomeextentPhilippines)haveplacedarelativelygreater

reliance on community-based and livelihood approaches, refecting greater inormality o labor markets, higher

rural shares in total workorces, and oten under-developed household level targeting systems which may limit

the capacity or ne-tuned responses such as those available to higher income AMS.

thelowincomeAMS(inparticularLaoPDR,CambodiaandMyanmar)haveforthemostpartnothadamajor

response to the crisis in terms o new types o interventions nor major scal stimulus towards existing programs.

This is a mixed unction o relatively minor impacts or some low income AMS rom the crisis to date, limited

administrative capacity to scale up SP program delivery rapidly, and other actors.

acommonfeatureacrossAMSexceptthelowincomemembershasbeentheprevalenceofcashtransfersas

part o the crisis response. These have taken a variety o orms (see Table 2 or summary o the diversity o cash

transer options adopted), ranging rom one-o to more sustained but time bound, to more structural reorms

(e.g. Philippines). The prevalence o cash transers as part o crisis response is notable in a region which has

not historically placed a strong reliance on cash transers. In contrast, there has been a relative lack o reliance

on incremental support to pre-existing in-kind transer programs, in some cases due to earlier reorms o in-

kind programs which had limited their coverage (e.g. Indonesia), and in others due to policy dierentiation by

the authorities in the SP response between the GFC and the ood crisis (e.g. Philippines).


35/68

25

An imprtant crss-cutting pint in AMS SP crisis respnse as een te level institutinal crdinatin acrss

agencies implementing SP related prgrams. This is a particular challenge in the social protection eld, where

multiple agencies are oten involved in policy development and program delivery. There has been considerable

diversity o approach across AMS in grappling with this challenge. Some countries such as Singapore were

already well-positioned pre-crisis due to the institutional consolidation o social insurance (under CPF) and social

assistance/welare programs (under the ComCare Fund). Others such as Philippines have used the crisis to promote

greater institutional coherence in SP institutions, while other AMS has not made specic institutional coordination

or consolidation eorts. While it is too early to reach clear conclusions, AMS where coordination across agencies

was stronger, or where coordination mechanisms have been put in place rapidly, appear to have been able to react

more quickly and implement a more integrated set o SP responses. A brie discussion o coordination challenges

as well as promising examples o institutional coordination can be ound below.


36/68

26

Tpes Scial Prtectin Respnses

Given te imprtance GFC impacts n emplment, tere as een a natural cus n lar maret interventins

dierent rms. In several middle and high income AMS, a variety o active labor market interventions have been

implemented which largely target ormal sector workers in crisis-impacted rms. In other middle and lower income

AMS, where inormal sectors dominate employment, support to employment has been ocused on incremental

stimulus unding or more decentralized small inrastructure and community-based livelihood initiatives (e.g. in

Philippines, Thailand, Vietnam and Indonesia), or increased ocus on existing labor intensive public works programs

(e.g. in Cambodia and Lao PDR). The prevalence o interventions directly ocused on job retention and employment

generation seems appropriate given the transmission channel o the GFC.

At te same time, tere remain a numer crisis-aected grups are unliel t enet rm direct

emplment-related measures, and r m ter instruments ave een used t mitigate direct and secndar

crisis impacts. Groups that are less likely to be included in the ambit o labor market interventions include those in

the inormal sector, migrant workers, or those with low education levels who may not meet minimal pre-conditions


37/68

27

or participation in training. Even or small inrastructure and community-based works, there will be groups who are

less likely to be covered, including those with limited work capacity like elderly and some people with disabilities,

or groups such as women who may generally have lower participation rates in works programs. For these groups,

SP interventions like social assistance transers which address the wider population o poor and vulnerable

populations are more relevant, but some o them also ace challenges in accessing saety net programs (e.g.

internal migrants). The discussion below outlines the ways in which AMS have struck a balance between dierent

SP instruments, with a key nding being the importance o blending instruments to respond to dierent types o

crisis impacts, the location and sectoral prole o impacted groups, and the pre-existing delivery mechanisms.

(a) Active Lar Maret Plicies (ALMPs). Given tat te primar transmissin cannel r te GFC as een

trug te lar maret, ALMPs dierent tpes ave een an imprtant cus SP respnses in a numer

AMS. While it is dicult to know how the situation may have evolved in the absence o intervention, it

appears that the relatively low levels o open unemployment during the GFC are at least partly attributable toadjustments in the labor market on wages, work time, bonuses, and so on, in some cases supplemented by

public incentives to retain or hire workers. Overall, ALMPs have tended to play a more signicant role in crisis

response in middle-high income AMS, where the more ormal nature o labor relations and social insurance

systems provide easier channels or directing dierent orms o public subsidies to both employers and workers.

The philosophy behind the approach taken in several AMS is captured in the phrase cut costs to save jobs, not

jobs to save costs. Labor market interventions across ASEAN include:

Providing wage subsidies and/or preerential tax treatment or employers to encourage worker retention

and rehiring o retrenched workers. Singapore has used retention incentives to good eect during the

crisis, while AMS such as Malaysia have created tax incentives to rehire. There has also been use o

direct wage increases or selected groups, such as Cambodias increase in salaries or civil servants,

teachers and garment workers, and Lao PDRs 20 percent increase in 2009 in the minimum wage or ormal

sector workers.

Use o the unemployment beneft system. This has included extending the duration o unemployment

insurance benets in AMS and or sectors where available (as seen in Thailand, with the extension o

unemployment insurance rom six to eight months or ormal sector employees under the SSF scheme), and

tax relie on unemployment benets or retrenched workers in Malaysia.

Subsidizing training to protect jobs or provide alternative activity, while enhancing worker productivity.

Examples rom various AMS are provided in Box 3. The attraction o (re)training workers at a time when

the opportunity cost o participation in training is low is potentially signicant, although should obviously be

relevant to labor market demand.


38/68

28

Expanding public sector employment as a cushion. As seen in Section 1, public sector employment has

been an area o net job creation during the crisis, particularly in middle income AMS such as Malaysia,

Philippines and Thailand.

Public acilitation o social dialogue between employers and workers to promote worker-riendly

management o employment impacts o the crisis. Malaysia provides an example o an integrated emphasis

on tripartite mediation and consultation between employers, workers and government through the Ministry

o Human Resources.

wile te utcmes lar maret interventins t date appear psitive in terms stening te negative

emplment impacts te crisis, designing active lar plicies tat can eectivel address crisis-induced

emplment impacts is callenging. A rst point to note is the limitations o some types o ALMPs in AMS

with higher shares o inormal sector employment, as incentive measures delivered through the tax or social

contribution systems will only be eective or ormal sector workers. A second issue is the potential mismatch


39/68

29

between workers in sectors most negatively impacted by the crisis (i.e. manuacturing, mining, tourism, estate

arming) and the sectors with net job growth (i.e. public sector, sales and trade), with the latter requiring higher

qualications, more entrepreneurial skills, or willingness to shit rom the ormal to inormal sector. A third issue is

the spatial dimension o impacts and responses when impacts are localized, such as with estate arming, but the

supply side is constrained in impacted areas in its ability to respond. Another issue is the appropriate time horizon

or emergency stimulus measures in the labor market, given the risks o sustaining them too long and articially

prolonging employment in sectors or rms where structural change and shiting comparative advantage would

have led to declining employment even in the absence o the crisis.

() Pulic rs. Unlie previus crises in ASEAN and ter regins, te GFC respnse as nt seen te launc

ne fagsip pulic rs prgrams. hever, incremental unding t small inrastructure and cmmunit-

ased scemes is liel t ave een prviding signicant temprar emplment r inrmal rers,

particularl in rural areas. The absence o major new works programs may be due to the muted eects to dateon employment levels in AMS and the inclination among laid-o workers to continue with job search in urban

areas and more lucrative sectors, as described in Section 1. Nonetheless, a number o stimulus packages (e.g.

in Indonesia, Philippines, Vietnam, and Thailand) have included signicant allocations to small inrastructure

programs which in many cases are community-based approaches to create temporary employment

opportunities and livelihood options and are intended to be relatively labor-intensive. The incremental unding

or such community-based works has been acilitated in a number o Member States by pre-existing programs

or channeling public unds or decentralized asset creation and employment generation. Examples o such

approaches include Thailands Suciency Economy Community Project, Indonesias Community Empowerment

Program, and the Comprehensive Livelihood and Emergency Employment Program in the Philippines, all o

which are decentralized stimulus measures (see Box 4). The Indonesian CLEEP is notable or the earmarking in

2009 o unds or the most crisis-aected districts, within the wider context o signicant increases in overall

allocations to the program. In some cases, the programs evolved and grew rom labor-intensive responses to

previous crises, and most combine geographic targeting with sel-targeting o poor households through the

wage rate.

Ler-incme AMS suc as Camdia and La PDR ave als cntinued partnersips it dnrs and NGos

n cas- and d-r-r prgrams, and pulic rs seem particularl ell-suited t suc ler capacit

envirnments, t as crisis perid saet nets and as prductive saet nets at ter times. Experience

rom AMS and beyond suggests that labor-intensive public works can be very eective in crisis response,

particularly where there are large inormal sectors not easily amenable to the range o ALMP interventions,

and where targeting mechanisms unction poorly and the sel-targeting o public works through the wage rate

oers a viable alternative. They have proven to be programs which can be put in place more quickly than other

types o SP interventions such as social insurance or de novo targeted social assistance programs. They also

have the attraction in Member States without signicant coverage o unemployment insurance o perorming a

mixed saety net unction o general income supplementation and fexible entry/exit or individuals in the ace

o unemployment or acute under-employment which provides a consumption smoothing eect.


40/68

30

(c) Cas transers. Te current crisis, and te d and uel crises ic preceded it, as signicantl increased

te rle cas transers itin te SP plic mix in man AMS. As noted in Section 1, the relatively mild

impacts o the GFC on poverty headcount estimates in AMS do not refect the magnitude o the social impacts on

the regions population due to increased un- and underemployment, lower earnings, incomes and remittances,

and reduced consumption. In this context, the potential role o social assistance is obvious. The GFC response

has seen a signicant increase in the priority given to cash transers in a number o AMS, though it remains

to be seen to what extent this represents a structural shit in priorities which will outlive the crisis response,

or will revert to the pre-crisis situation. Several AMS have implemented incremental measures to supplement

household incomes, mostly through cash transers o various types (i.e. conditional, unconditional, one-time).

A key distinction is between AMS that are combining crisis response and structural reorms o cash transers,

and those that are using crisis-specic transer programs without necessarily planning to continue them asa longer-run saety net. In the ormer case (as exemplied by the Philippines with the introduction o a large

scale CCT targeted through proxy means-testing), the crisis has served as an important catalyst or reorms that

were already contemplated prior to the crisis. In other cases, such as Malaysia, it remains to be seen whether

the enhancement o social assistance programs within the crisis response will be sustained as a medium to

long run jump shit in the priority o cash transers or the poor and vulnerable. A urther interesting example


41/68

31

is Indonesias targeted BLT cash transer program, which has been introduced and withdrawn twice, rst in

response to the uel crisis (in parallel with reduction in universal uel subsidies), and more recently in response

to the GFC. Examples o responses to the GFC through cash transers are provided in Box 5.


42/68

32

An incipient tug ptentiall signicant trend in cas transer plicies in several AMS as een te

intrductin cnditinal cas transers in recent ears. Given the long-standing ambiguity around social

welare transers in a number o AMS, the linkage between transer policies and human capital acquisition

through CCTs may appeal as a orm o social investment which can have positive personal and societal

economic impacts. Indonesias Program Keluarga Harapan (PKH) was introduced on a pilot basis in 2005

and provides benets to poor households with children 0-15 or pregnant women who are targeted through

a proxy means-test (PMT). Benet eligibility is subject to compliance with education, health and nutrition

related conditions. In contrast to BLT (which is specically crisis-related), the program is intended to remain

as a permanent eature o the Indonesian SP system. The program aced initial implementation challenges, but

appears to be consolidating delivery systems as it matures. Philippines has rapidly expanded its CCT program

known as the 4 Ps, which is targeted to poor households with children 0-14 or pregnant women, also based

on health, education and nutrition related conditionalities. It also employs a PMT to target. The program has

expanded rapidly rom 20,000 households covered and P.50 million o spending in 2007 to a target o around 1million households in 2009 (or around 20 percent o total poor households in the country) and a budget o P. 15

billion. While ocused only on education, the Scholarship or the Poor program in Cambodia is also a targeted

conditional transer or poor children to incentivize secondary school attendance, while has been shown in a

recent evaluation to increase attendance by around 20 percent (Filmer and Schady, 2009).

(d) Fd susidies. A numer AMS ave prvided incremental resurces t d-ased prgrams in respnse

t te d price crisis and te GFC, tug tis as een less idespread r prnunced tan migt ave

een expected (see bx 6). The ood saety net story in ASEAN has been an interesting and varied one over

the course o the ood, uel, and nancial crisis. The most straightorward examples have been Member States

such as Malaysia and Cambodia, which signicantly increased their ood subsidies in late 2008 and early 2009

as part o the GFC/ood crisis response. In contrast, in Indonesia, a targeted cash benet (a revamped version

o the BLT cash transer which had been used in 2005 to mitigate the social impact o uel price reorms) was

introduced in 2008 as a response to the ood crisis. While the rice subsidy program has been retained, the

marginal increments in spending in social assistance crisis response have been through cash. A nal example is

Philippines, where rice subsidy spending was ramped up in 2008 in response to the ood crisis, but then reduced

and oset by a targeted cash benet in 2009 during the GFC response. While increasing unding to existing ood

subsidy programs is a viable response option, particularly or low income Member States without signicant

alternative programs or incremental SP spending, overall it is unclear whether generalized increases in ood

subsidies in response to the GFC are the most ecient or eective SP response or several reasons. Firstly,

as evidence rom several AMS shows, they are likely to spread benets to a signicant share o the non-poor,

diluting the poverty mitigation impact. Secondly, there may be more signicant transactions costs than or

cash benets where such programs involve ood distribution and require supply side capacity in logistics than

may be under-developed. Thirdly, ood will inevitably be less ungible than cash and may not provide aected

households with as much fexibility in coping with crisis impacts.


43/68

33

(e) Scial insurance. Te ptential rle scial insurance as a crisis respnse tl is largel determined te

pre-existing sstem cverage, and as a result its rle in GFC respnse as een limited in mst AMS. Social

insurance programs, particularly those seeking to extend coverage to the inormal sector, require substantially

more time to design and roll out than other SP instruments such as public works. As such, they are unlikely

to be eective tools in crisis situations unless already operational prior to the onset o a crisis. This inherent

constraint is refected in the GFC crisis response in AMS, where social insurance has not been as signicant

as other pillars o social protection. However, it has played a useul role in more ormalized economies o


44/68

34

ASEAN that had signicant pre-existing coverage o social insurance, through measures such as extension o

UI benets (noted above with respect to Thailand) and more lenient treatment o social insurance contributions.

An example o the latter is the reduction in EPF employee contributions or Malaysian workers rom 11 to 8

percent rom February 2009 to end 2010. Interestingly, around hal the workers have chosen to sustain their

original rates o contributions to date. Similarly, Singapore has introduced a range o measures through its tax

system to protect household incomes, including a 20 percent personal income tax rebate, and doubling o the

Goods and Services Tax Credit benets, the two measures estimated to costs over S$1 billion in 2009.

() Prgrams t prmte cntinued access t scial services. Facilitating sustained access t scial services

as een a part te crisis respnse in a numer AMS. The employment and secondary impacts o the

GFC raise the obvious risk o detrimental coping strategies by households such as withdrawing children rom

school and avoiding necessary medical treatment. Analysis rom the region also suggests that these risks are

more pronounced or girls and women. In a number o AMS, ree access policies or health and education thatwere already in place, while in other cases, ree access policies have been extended to wider groups (e.g., in

Thailand, including 12-15 year olds under the ree education policy). School eeding programs have also been a

tool in particularly low-income Member States such as Lao PDR and Cambodia, and have received incremental

nancing in response to the FFF crises.

Crss-cutting Issues in Scial Prtectin Plicies and Prgrams in ASEAN Memer States

(a) Targeting and program coverage. A e crss-cutting issue r all AMS seeing t scale up existing transer

prgrams r intrduce ne transer prgrams in respnse t te GFC as ell as te d and uel crises as

een t identi pr eneciaries and tse speciall impacted te crisis. Where robust identication

processes were already in place, it has been relatively easy to inject additional unds and/or expand eligibility

to the near-poor or those particularly aected by the crises who were not previously beneciaries. This can

be seen in the GFC response o Member States such as Vietnam, Singapore and Malaysia. However, as noted

in the discussion o pre-crisis SP programs, many AMS did not have a pre-existing generalized poverty/social

assistance benet, but ocused on specic categories o the population. I broad-based targeting mechanisms

were not already in place, it has proven dicult to introduce new targeting mechanisms mid-crisis which are

very precise, and AMS have had to rely on easily veriable categorical individual or household indicators (such

as old age, which was used as an indicator in Thailand), or on existing area-based unding channels.

Te erts AMS t imprving targeting mecanisms in SP prgrams ave nt een an utcme te FFF

crises per se, ut te crises ave cused te attentin plicmaers n targeting sstems in a a nt seen

previusl. Disentangling what in targeting reorm is crisis response and what is accelerated but incremental

reorm o SP systems is not straightorward, but there is a growing body o experience in AMS in this area, and

at all levels o country income. Box 7 gives a summary o some o the recent eorts o low and middle income

AMS to improve their systems or identication o the poor or the purposes o SP program eligibility.


45/68

35

Cmplexities arise en tse mst directl aected te crisis are nt tpicall included in regular

targeting sstems. As noted in Section 1, this has been the case in the ood and nancial crises to some extent

when direct impacts have been elt more in urban areas, or among ormal sector workers, since regular targeted

SP programs tend to be ocused more on the rural poor. There may also be gender dimensions or specic

challenges related to migrant workers in Member States where employment impacts have been concentrated


46/68

36

among these groups. In such cases, tensions may develop between short-term crisis response objectives and

long-term SP system goals. Some reorms being introduced in AMS such as proxy means testing can provide a

vehicle or simulating the impacts o crises using household survey data, and the proxy means testing ormula

can be adjusted to acilitate targeting in response to crisis. Equally, where capacity is lower or systems are not in

place, the sel-targeting nature o labor-intensive public works provide benets which have been demonstrated

in previous crises.

one grup tat AMS at all incme levels ave und particularl callenging t deal it in teir SP crisis

respnses as een migrant rers, t tse cming rm ter AMS and internal migrants. As noted in

Section 1, migrant workers have been among the rst aected during the crises. However, the SP ramework

or dealing with migrant workers is generally underdeveloped. Since many migrants work in the inormal sector,

they are unlikely to have SP coverage, either because they are oreign nationals, or because eligibility in saety

net schemes or nationals is linked to residence and ID systems that link the migrant worker to the place oorigin rather than working location.

() Institutional coordination: Te GFC als igligts te callenges rapid and crdinate scial plic

respnses en SP prgrams are spread acrss varius agencies, ten respnsile r dierent segments

te ppulatin. Not all AMS have a central agency responsible or coordinating SP programs. Rather,

implementation o SP programs is oten the responsibility o separate Ministries and ad hoc program units.

In practice, this has resulted in the use o dierent targeting approaches or various SP programs, oten with

separate and unlinked beneciary databases, and parallel monitoring o program implementation and outcomes.

The challenges o coordination across programs are amplied when there are multiple unding sources or

programs, and donors and agencies providing technical and nancial support ace their own coordination

challenges. The urgency o crisis response and the high-level political ocus on SP programs as part o broader

stimulus packages can act as catalysts or increasing institutional coordination and coherence. Despite these

challenges, ASEAN provides promising examples o institutional consolidation and coordination, both pre-crisis

and during the crisis-response period. Box 8 provides details on two country examples.


47/68

37

(c) Monitoring the Impact o SP Programs:Mst AMS are still seeing t put in place timel and reliale sstems

t mnitr nt nl te impacts crises n sci-ecnmic indicatrs, ut als eective M&E sstems

t assess te elare impacts teir SP prgram respnses. While timely mobilization o resources and

programs to respond to crises has naturally been the top priority o AMS, there remains a strong need in most

countries or better program monitoring and understanding o the program eectiveness. Indonesia provides

an example o eorts to remedy this situation. Under the lead o the national planning agency BAPPENAS, the

Government is establishing a Crisis Monitoring and Response (CMR) system which is designed to gather the

necessary data to monitor the eects o the GFC on Indonesians, the coping mechanisms being adopted, and

the socio-economic consequences o these developments. The CMR will draw both on existing data sources

and on a new quarterly household survey which is designed to be relatively inexpensive and representative

at the district level. This is to be complemented by rapid qualitative w

Documents

REGIONALANDCOUNTRY REPORTS OF THEASEANASSESSMENT ONTHE SOCIAL IMPACT OFTHEGLOBAL FINANCIAL CRISIS