View
219
Download
0
Embed Size (px)
DESCRIPTION
Social benefits can potentially play an important role in protecting the poor and minimizing the impacts of an economic crisis. While many studies estimate the impacts of a crisis, there is little evidence of the actual response of social safety nets to systematic shocks. This study traces the response of social benefits during the 2008-10 global crisis for 14 countries in ECA. The study first sets out a framework for defining the “expected” response of social benefits covering an assessment of pre-crisis preparedness of social benefits and the severity of the crisis for all countries in the ECA region to provide the context; and then develops a typology of all countries categorized by expected response.
Citation preview
Printed on recycled paper
About this series...
Social Protection & Labor Discussion Papers are published to communicate the results of The World Bank’s work to the development community with the least possible delay. The typescript manuscript of this paper therefore has not been prepared in accordance with the procedures appropriate to formally edited texts.
The findings, interpretations, and conclusions expressed herein are those of the author(s), and do not necessarily reflect the views of the International Bank for Reconstruction and Development / The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The author(s) attest(s) that the paper represents original work. It fully references and describes all relevant prior work on the same subject.
For more information, please contact the Social Protection Advisory Service, The World Bank, 1818 H Street, N.W., Room G7-803, Washington, DC 20433 USA. Telephone: (202) 458-5267, Fax: (202) 614-0471, E-mail: [email protected] or visit us on-line at www.worldbank.org/sp.
N o v e m b e r 2 0 1 2
Abstract
Social benefits can potentially play an important role in protecting the poor and minimizing the impacts of an economic crisis. While many studies estimate the impacts of a crisis, there is little evidence of the actual response of social safety nets to systematic shocks. This study traces the response of social benefits during the 2008–10 global crisis for 14 countries in ECA. The study first sets out a framework for defining the “expected” response of social benefits covering an assessment of pre-crisis preparedness of social benefits and the severity of the crisis for all countries in the ECA region to provide the context; and then develops a typology of all countries categorized by expected response. Using this typology the study analyzes the monthly administrative data on the observed patterns within social benefit programs. Main findings indicate that actual responses were largely in line with expectations. Pre-crisis preparedness clearly influenced the ability of social benefits to respond to the crisis. Unemployment benefits were generally the first line of response in countries that have them, while social assistance programs also expanded coverage during the crisis. Lessons learned from the 2008-2010 global crisis (such as the importance of structural reform, design, and implementation which affect the success of social benefits programs in crisis response) are also presented. The study concludes with some policy recommendations to help ECA countries prepare for future crises.
Do Social Benefits Respond to Crises?
Evidence from Europe & Central Asia During the Global Crisis
Aylin Isik-Dikmelik
D I S C U S S I O N P A P E R NO. 1219
Do Social Benefits Respond to Crises?
Evidence from Europe & Central Asia During
the Global Crisis1
Aylin Isik‐Dikmelik
November 2012
Human Development Sector Unit
Europe and Central Asia Region
1 This study was authored by Aylin Isik‐Dikmelik with inputs from Yulia Smolyar and Syon Bhanot; and a core World Bank team who were instrumental in data collection and the preparation of the country briefs. This core team comprises of: M. Ihsan Ajwad (Latvia), Anatassia Alexandrova (Kyrgyz Republic), Rajna Cemerska (FYR Macedonia), Lire Ersado (Armenia), Ivan Drabek (Croatia), Boryana Gotcheva (Bulgaria, Montenegro, Serbia), Ufuk Guven (Albania), Victoria Levin (Armenia), Nino Moroshkina (Georgia), Yulia Smolyar (Ukraine), and Emil Tesliuc (Bulgaria and Romania). Data on Hungary and Lithuania was kindly provided by the European Commission. Victoria Strokova and Syon Bhanot provided excellent research assistance. The study benefitted from discussions and comments from: M. Ihsan Ajwad, Mehtabul Azam, Paul Cahu, Mame Fatou Diagne, Ariel Fiszbein, Ufuk Guven, Boryana Gotcheva, Jesko Henstchel, Natsuko Kiso, Johannes Koetll, Jennica Larrison, Victoria Levin, Kathy Lindert, Isil Oral, Indhira Santos, Anita Schwarz, Owen Smith, Ramya Sundaram, and Asta Zviene. The work also benefitted greatly from the peer reviewers: Isabelle Maquet, Margaret Grosh, and Cem Mete. Funding from Crisis Contingency Fund and Trust Fund for Environmentally and Socially Sustainable Development (TFESSD) is gratefully acknowledged. All errors are the responsibility of the author.
2
Abstract Social benefits can potentially play an important role in protecting the poor and minimizing
the impacts of an economic crisis. While many studies estimate the impacts of a crisis, there
is little evidence of the actual response of social safety nets to systematic shocks. This study
traces the response of social benefits during the 2008‐10 global crisis for 14 countries in
ECA. The study first sets out a framework for defining the “expected” response of social
benefits covering an assessment of pre‐crisis preparedness of social benefits and the
severity of the crisis for all countries in the ECA region to provide the context; and then
develops a typology of all countries categorized by expected response. Using this typology
the study analyzes the monthly administrative data on the observed patterns within social
benefit programs. Main findings indicate that actual responses were largely in line with
expectations. Pre‐crisis preparedness clearly influenced the ability of social benefits to
respond to the crisis. Unemployment benefits were generally the first line of response in
countries that have them, while social assistance programs also expanded coverage during
the crisis. Lessons learned from the 2008‐2010 global crisis (such as the importance of
structural reform, design, and implementation which affect the success of social benefits
programs in crisis response) are also presented. The study concludes with some policy
recommendations to help ECA countries prepare for future crises.
JEL Classification: H53; I38; J68 Keywords: social safety nets; crisis readiness; crisis response; unemployment benefits; Europe and Central Asia
1
Table of Contents I. OVERVIEW: CONCEPT AND METHODOLOGY ....................................................................... 3
Response of Social Benefits to Crisis in Practice (Overview of Methodology) ................................ 4
II. EXPECTED SOCIAL BENEFITS RESPONSE: BROAD FRAMEWORK ......................................... 7
Pre‐Crisis Preparedness of Social Benefits Systems (pre‐2009) ...................................................... 8
The Severity of the Crisis (2009‐2010) .......................................................................................... 12
Expected Crisis Response: Pre‐crisis Preparedness + Severity of the Crisis .................................. 18
III. OBSERVED SOCIAL BENEFIT RESPONSES ........................................................................... 20
IV. LESSONS LEARNED: THE IMPORTANCE OF ACTION ........................................................... 30
V. CONCLUDING REMARKS AND POLICY RECOMMENDATIONS ............................................ 36
REFERENCES .................................................................................................................................. 40
Annex 1 ‐ Methodology on Pre‐Crisis Preparedness of SB Rankings ............................................ 43
Annex 2‐ Individual country briefs ................................................................................................ 50
List of Figures:
Figure 1: Social Assistance Programs (as a share of GDP) ............................................................. 8
Figure 2: The Pre‐Crisis (before 2009) Preparedness of Social Benefit Programs in ECA
Countries ....................................................................................................................................... 10
Figure 3: GDP growth (annual percentage change) ..................................................................... 13
Figure 4: Severity of Crisis: Change in GDP & Unemployment Rate 08 to 09‐‐ ECA Countries ... 14
Figure 5: Wide Variation in Severity of the Crisis Across the ECA Region (SBO Plus Sample) ..... 15
Figure 6: Fiscal Balance: ECA Countries 2007‐2009 ..................................................................... 16
Figure 7: Severity of Crisis vs. Fiscal Constraints: GDP Growth 2009 & Change in Fiscal
Balance 2007 to 2009‐‐ SBO Plus Sample ..................................................................................... 18
Figure 8: Disability Benefits Indexed Number of Beneficiaries ................................................... 29
Figure 9: Both Ukraine and Latvia had Room for Improving the Quality of Social Safety Net
Spending ....................................................................................................................................... 32
2
List of Tables:
Table 1: Do We Expect Social Benefits to Respond? ................................................................... 19
Table 2: Observed & Expected Responses of Social Benefits to the Global Crisis, ECA .............. 20
Table 3: Timing and Duration of Responses of Social Benefits to the Global Crisis, ECA ............. 21
List of Boxes:
Box 1: Other Social Protection Instruments: Government Measures and Potential Behavioral
Responses ..................................................................................................................................... 22
Box 2: Response of Social Benefits to the Crisis in the EU ........................................................... 26
Box 3: The Response of Social Benefit Programs in the US during the Crisis ............................... 35
3
I. OVERVIEW: CONCEPT AND METHODOLOGY
The global economy experienced one of the most severe economic downturns of recent
times starting in 2008 and continuing through 2009. The Europe and Central Asia Region
(ECA) was more directly affected than any other region by the recent global financial crisis,
though the extent of this impact varied significantly by country. Although there were signs
of recovery in 2010, food prices increased again in 2011, highlighting the importance of
social benefits as a crisis response mechanism. As a result, it has become imperative to
understand the role that social benefits may have played in mitigating the welfare impacts
of the global financial crisis.
Social benefits2 – and social protection more broadly – are among the mechanisms that
are called upon to respond to systemic economic crises. It is widely accepted that social
benefits are likely to play an increasingly important role in mitigating the negative impacts
of crises through prevention (providing security to the vulnerable), protection (providing
adequate support to the poor), and promotion (expanding opportunities for higher
productivity and income, especially as the crisis wanes).3 Conceptually, some safety net
programs (in other words, means‐tested entitlement programs) are supposed to work as
“automatic stabilizers” in that they are designed to expand to meet increased need (helping
households to smooth their consumption, mitigating the adverse impact of the shocks) and
to contract as the economy recovers. Governments can expand existing programs (with
well‐proven track records) – either in coverage or the value of benefits – to channel
additional resources to the poor quickly and cost‐effectively. Alternatively, they can create
new programs as a crisis response, though this takes time. If effective, social benefits can
alleviate increases in poverty and inequality by enabling families to continue to access food
and other important services for health and education, thus preventing irreversible losses to
their human capital.
While this countercyclical role of social benefits is intuitive, it is not always
straightforward in practice. Counter cyclicality can be constrained by design features,
2 Throughout this report, social benefits refer to non‐contributory cash or near cash transfers (such as last resort social assistance programs, child/family allowances, and disability allowances) and unemployment benefits. Although unemployment benefits tend to be contributory in nature, we included them in our definition of social benefits because this crisis manifested itself mainly through the labor market and the automatic stabilizer role that they are likely to play in an external macro shock. 3 The World Bank (2012b). “Resilience, Equity, and Opportunity: The World Bank’s Social Protection & Labor Strategy, 2012‐2022. Robert Holzmann and Steen Jorgensen (February 2000). “Social Risk Management: A new conceptual framework for Social Protection and beyond.” World Bank Social Protection Discussion Paper No. 0006.
4
implementation capacity, institutional capacity, and fiscal pressures. Thus, systemic
evidence on how social benefits actually respond to crises in practice is minimal. While
some studies4 have simulated the role that social benefits could be expected to play after a
systemic shock, there is not much evidence on how social benefits have actually responded
to systemic shocks in practice. Only a few studies have examined actual responses to crisis.
In terms of actual responses, the literature mostly focuses on how governments have used
social benefits as response to crises (for example, the expansion of the safety net in Korea
during the East Asia crisis and the expansions of social pensions in Thailand), but the
literature does not say much about the actual increases or magnitudes of the benefits
provided. Two reports by the Social Protection Committee of the European Commission5
examine the responses to the global crisis in EU member states, using administrative data
(but do not cover ECA countries more broadly). We draw significantly on this paper and
incorporate the underlying data for EU new member states (part of ECA region) for our own
analysis.
Response of Social Benefits to Crisis in Practice (Overview of Methodology)
This study traces the actual response of social benefits during the global crisis using
monthly administrative data in a “Social Benefits Observatory” (SBO). We presumed that
as the global crisis unfolded, a decrease in household incomes and employment would
create an increased demand for benefits. The degree to which social benefits programs
respond to an increase in applications can be an indicator of their flexibility (counter‐
cyclicality) as crisis response mechanisms. Furthermore, the degree to which information on
numbers of applicants, beneficiaries, and spending on such programs is readily available can
be an indicator of administration capacity, which if low, can signal the need for structural
improvements in the medium term. The SBO was developed to monitor the program
response.
The actual response of social benefits during the global crisis was examined for 14
countries in the ECA region. Specifically, monthly administrative data were collected in the
4 Please see Leventi et al (2010) and Ozdemir et al (2010). Though these studies do not directly estimate the role of social benefits, they do simulate the distributional impact of the crisis as well as the austerity measures which has implications for the demand for social benefits. Also see Narayan and Sanchez‐Paramo (2012), for a macro‐micro simulation model estimates the distributional impact of the global crisis for 4 case study countries: Bangladesh, Mexico, Philippines, and Poland. 5 Council of the European Union (2010), “Joint Assessment by the Social Protection Committee (SPC) and European Commission of the Social Impact of the Economic Crisis and of Policy Responses.” Council of European Union (2011), “Third Report on the Social Impact of the Economic Crisis and Ongoing Fiscal Consolidation.”
5
SBO for the period from 2008 (pre‐crisis benchmark) to 2010 for over a dozen countries in
the ECA region: Albania, Armenia, Bulgaria, Croatia, Georgia, Hungary, the Kyrgyz Republic,
Latvia, Lithuania, FYR Macedonia, Montenegro, Romania, Serbia, and Ukraine. Countries
were selected on the basis of availability of data from administrative information systems,
with the goal of covering a broad range of countries in diverse contexts across the region.
The original SBO sample included monthly administrative data for non‐EU members
(Albania, Armenia, Croatia, Georgia, the Kyrgyz Republic, FYR Macedonia, Montenegro,
Serbia, and Ukraine), along with data for Bulgaria, Romania, and Latvia (EU member states).
The sample was expanded to cover Hungary and Lithuania (EU members) with data from a
study by the Social Protection Committee of the European Commission.6 As such, the “SBO
Plus” sample covers all 14 countries, as reported in this study.
The study focuses on the response of key social benefits, including social assistance cash
transfers plus unemployment insurance. Specifically, social benefits included in the sample
comprise non‐contributory cash assistance benefits, such as Last Resort Social Assistance
programs (LRSA), child/family allowances and disability allowances, plus unemployment
benefits. Although unemployment benefits tend to be contributory in nature, we included
them in the definition of social benefits for the SBO because the effects of the crisis were
transmitted mainly through the labor market, and as such, one would expect
unemployment benefits to respond to the crisis. This is not to disregard the potential role
of other social protection instruments or programs in responding to the crisis, such as
pensions, public works programs, and so forth.
In terms of variables, the study focuses mainly on the response of “coverage” and
“monthly program expenditures” to the crisis. The Social Benefits Observatory collected
monthly administrative data on the number of applications, rejections (to the extent
available), beneficiaries (coverage), and spending for the main social benefits programs. In
the final analysis, we focus our analysis and findings on the coverage indicator, because it is
the most complete data series across time (months) and countries (in the SBO sample). The
main source of these data was government management information systems, provided by
national Ministries of Labor and Social Protection. In addition to these monthly quantitative
data, the SBO also compiled data on changes in the policy environment from various
sources (official reports, government ministries, reform information in Development Policy
Operations (DPOs), World Bank staff working on the particular countries, and so forth). It
should be noted that throughout this paper, the term “response” is used to reflect a change
6Council of the European Union (2010). “Joint Assessment by the Social Protection Committee (SPC) and European Commission of the Social Impact of the Economic Crisis and of Policy Responses.”
6
in the relevant indicator compared to the pre‐crisis levels of the same indicator – for
example, an increase in coverage (or spending) as compared with the pre‐crisis benchmark.
Although we have attempted to document other factors besides the crisis that may have
influenced such changes, we acknowledge that the “response” terminology points to an
association with the crisis, rather than to direct causality.7
It is also important to note that existing beneficiaries of social benefits programs will also
be protected to some extent from the impacts of the crisis (e.g., via a more certain income
stream from the transfers) – and the maintenance of existing coverage can also be
considered as a crisis response. Furthermore, improvements in targeting within the existing
level of coverage can also constitute a crisis response. For example, in Armenia, the
Government implemented a proactive strategy to reduce fraud and error and improve
targeting so as to increase the coverage of the poor within existing coverage and budget
limits.8 Such a response to the crisis was identified in our qualitative information on policy
responses, but not picked up in the overall quantitative measures of changes in coverage or
spending under the SBO. Finally, the creation of a new targeted social benefit program
could also constitute a crisis response, particularly if one did not exist prior to the crisis.
While beyond the scope of our sample and study, two countries in the ECA region
introduced new mechanisms to channel benefits to the poor as part of their crisis response:
Moldova (which is starting to scale‐up a pilot targeted program) and Tajikistan (which is
starting to pilot the development of a registry for targeting social benefits to the poor)9.
Actual (observed) responses of social benefits were then contrasted with the expected
response, taking into account the severity of the crisis in each country and the “pre‐crisis
preparedness” of their social benefit systems. As a broad framework, the degree to which
social benefits respond to a systemic shock (crisis) depends on: (i) the severity of the crisis
(and the nature of the shock and the transmission channels for crisis impacts); and (ii) the
extent to which social benefits systems are ready to respond (pre‐crisis preparedness); as
7 It should be noted that it is difficult to quantify the magnitudes of expected social benefits responses. While some studies have estimated the impact of certain types of crises on poverty and/or overall distributional outcomes (Tiongson et al 2010; Leventi et al 2010; Ozdemir et al, 2010; Narayana & Sanchez‐Paramo, 2012), none seems to have systematically translated these changes in welfare into expected demand for certain social benefits by simulating the rules of benefits (assuming there are no rule changes). While this is an interesting task for future researchers, it is beyond the scope of the current paper. 8 In addition, a study of the impacts of crisis in Armenia, World Bank (2010), indicated that coping strategies, most important of which was support provided by the government (via pensions and targeted family benefits), may have averted substantial increase in poverty in Armenia during the crisis. 9 ILO/WB (2012) provides a comprehensive inventory of policy responses in labor markets and social protection to the global crisis.
7
well as other considerations, such as the influence of fiscal and political economy
constraints. The paper classifies the 14 ECA countries according to the severity of the crisis
(GDP, unemployment, fiscal shocks). It also uses data from household surveys plus
administrative information to classify countries according to the pre‐crisis preparedness of
their social benefit systems.
The paper is organized into five sections. The broad framework for defining the
“expected” response of social benefits in all ECA countries is developed in Section II to
provide the context in which to evaluate the actual responses. This covers: (a) an
assessment of pre‐crisis preparedness of social benefits for countries in the ECA region; and
(b) the severity of the crisis, which varied substantially in countries across the ECA region
(and within the SBO Plus sample). Section II compares these key factors governing the
expected response, developing a typology of countries categorized by expected responses
according to the severity of the crisis and pre‐crisis preparedness. Section III presents our
main findings based on our analysis of the administrative data under our Social Benefits
Observatory (SBO Plus sample). Section IV, addresses the lessons learned from our analysis,
while the final section summarizes the main conclusions and presents some policy
recommendations.
II. EXPECTED SOCIAL BENEFITS RESPONSE: BROAD FRAMEWORK
The extent to which governments can use social benefits to respond to a crisis depends
very much on the capacity of their core delivery systems pre‐crisis. These core social
benefits systems include targeting mechanisms, reliable household registries and
management information systems, mechanisms for making payments and delivering
services, and tools for basic monitoring, oversight, and control. When well‐targeted and
well‐functioning programs with a reasonably high coverage of the poor exist, they can be
scaled up during a crisis to alleviate the negative impact on household welfare.
Most countries in ECA operate a mix of social benefits programs similar to that in OECD
countries, with at least one targeted safety net program (otherwise known as a last‐resort
social assistance program, LRSA). Many countries operate some type of unemployment
benefit scheme, though these tend to have low coverage due to the high degree of informal
labor in many ECA countries. There is also an emphasis on family allowances (including
child allowances and birth grants) followed by disability benefits and heating/housing
8
allowances. In many countries, there is a proliferation of multiple programs, which
frequently results in fragmentation of the social assistance sector (Figure 1).
Figure 1: Social Assistance Programs (as a share of GDP)
Source: Sundaram and Strokova (forthcoming) and ECA Social Protection Database
Experience from past crises shows that countries with an effective program mix and core
systems are well‐positioned to respond. A broad safety net covering multiple vulnerable
groups facilitates crisis mitigation. In addition, when core systems are in place, countries
can scale them up quickly, whereas if they are not in place, this limits the number of options
available to respond effectively.
Pre‐Crisis Preparedness of Social Benefits Systems (pre‐2009)
We developed a typology to categorize the “pre‐crisis preparedness” of social benefits in
ECA countries based on available information. Pre‐crisis preparedness draws on a
composite score for each country for their social benefits system prior to 2009 based on: (i)
the pre‐crisis performance of any last resort social assistance program (LRSA) in terms of
targeting accuracy (the percentage of benefits that went to the poorest quintile), coverage
(the percentage of the poorest quintile covered by the program), and generosity estimated
from household surveys; (ii) the pre‐crisis performance of any child allowance program in
9
terms of targeting accuracy, coverage, and generosity estimated from household surveys;
(iii) the coverage of any unemployment benefit (as proxied by the proportion of formal
sector workers among the employed, which indicates an upper bound); and (iv) information
capacity as an indicator of the efficacy of the administration of benefit programs (especially
since information monitoring systems would be needed to monitor the crisis and process an
increase in applications during a crisis). While these indicators cover a good portion of the
relevant information, the resulting index is not a precise measure. There are many other
factors (such as quality of implementation, administrative capacity to handle increase in
applications, design issues such as ease of application etc.) that are difficult to measure and
therefore cannot be included. Moreover, composite indices can be constructed in many
different ways (simple average, weighted average etc.). However, we sorted countries
based on the available social benefits information, and the index used in this study reflects
our best judgment. Annex 1 presents the detailed methodology used for this categorization
exercise along with the combination of programs and their performance.
Overall, social benefit systems in most ECA countries were generally well prepared
compared to countries in other regions pre‐crisis (as of 2008). Almost all countries10 in the
ECA region operate at least one well‐targeted LRSA program that could be scaled up if
needed, though coverage of the poor is low and is not in line with the total fiscal effort on
safety nets as a whole.11 Most countries operate child and family allowances, though, if not
targeted, these may not be cost‐effective vehicles for channeling assistance to the poor. In
addition, most countries have an unemployment benefit scheme, though the coverage is
mostly for the formal sector workers.12
Nonetheless, the ECA region is diverse, and there was a gap between the most prepared
countries and the least prepared. Figure 2 presents the countries ranked according to their
pre‐crisis preparedness.
10 Two exceptions were Moldova, which is now scaling up a targeted program, and Tajikistan, which is now piloting an initial targeting and registry system to channel benefits to the poor. 11 Sundaram and Strokova (forthcoming) and ECA Social Protection (SP) Database. 12 The existence and coverage of unemployment benefits and a poverty‐targeted social assistance program would be one way of categorizing countries with respect to their pre‐crisis preparedness. However there is not enough variation across ECA countries to result in meaningful categories for pre‐crisis preparedness. Hence, more detailed indicators were used to construct the composite index to rank countries with respect to their pre‐crisis preparedness of social benefits.
10
Figure 2: The Pre‐Crisis (before 2009) Preparedness of Social Benefit Programs in ECA
Countries13
Source: World Bank staff; Countries in red refer to SBO Plus sample countries.
Category A – Well‐Prepared as of 2008. Those in Category A score high on the
composite score for all factors (the existence and performance of LRSA programs,
the existence and performance of child allowance programs, existence and coverage
of unemployment benefits, and availability of information). These countries include
Armenia, Croatia, and Montenegro. All three countries in this category have very
13 The pre‐crisis preparedness ranking presented here is specific to the ECA region, reflects the pre‐global recession systems (2008 and before) and employs a different methodology with more detailed data (not available globally) than used in World Bank (2012a), hence resulting in a finer categorization of the countries in the region. This categorization is different from the global ranking presented in World Bank (2012a), which encompasses the 2008‐2010 period, and reflects the overall strength of the SSNs in ECA compared to other regions.
11
well‐targeted LRSA and child allowance programs with moderate to good coverage.
In addition unemployment insurance exists in all these countries with moderate to
high formality in the labor force, indicating moderate to high coverage. In addition,
the information on social benefits programs was available on a monthly basis with a
relatively small lag in data.
Category B – Reasonably Prepared as of 2008. Category B groups countries with a
reasonable ability to respond prior to the crisis, though reforms could improve their
capacity to address the crisis. Estonia, Georgia, Hungary, Poland, and Ukraine are
included in this group. In general, these countries have moderate to well‐targeted
LRSA programs, but with low coverage and low generosity in most cases (with the
exception of Georgia where coverage of LRSA is moderate and generosity is good).
In addition, with the exception of Georgia (where family allowance is being phased
out by not accepting new applicants since 2007), they all have moderate to well‐
targeted child allowance programs with good coverage. On the other hand, all
except Georgia have some type of unemployment benefit scheme with moderate to
high formality among the employed. The relatively less preparedness is due to low
coverage and generosity of LRSAs for most countries in this group and in the case of
Georgia, mostly due to lack of an unemployment benefit scheme.
Category C – Less Prepared as of 2008. However, Category C countries presented
significant weaknesses in their pre‐crisis preparedness, often due to extremely low
coverage of their LRSA programs (due to marginalization over time), and/or the
relatively weak performance of their child allowance programs (when they exist)
compared to the countries in Group B. This group includes Albania, Azerbaijan,
Bulgaria, FYR Macedonia, Lithuania, Romania, and Turkey. Moreover, while
unemployment benefits exist in all of these countries, the coverage is moderate to
low. Finally, overall information capacity is low in most of these countries due to a
lack of national/harmonized registries with most (e.g. FYR Macedonia, Albania,
Romania) still using paper‐based benefits administration prior to crisis.14 A
concerted strategy of actions and reforms would be needed, especially in
strengthening benefits administration and increasing coverage of LRSA program, to
enhance capacity to respond to a crisis more effectively and efficiently to protect the
poor.
Category D – Least prepared as of 2008. Finally, there is a significant group of
countries in which an overhaul of the social assistance system would be needed.
14 Prompted by the fiscal pressures during the crisis and in some cases due to an ongoing reform effort, some of these countries are currently in the process of moving from paper‐based to automated management information systems to improve implementation efficiency.
12
These countries include Belarus, Bosnia and Herzegovina, Kazakhstan, Kosovo,
Kyrgyz Republic, Latvia, Moldova, Russia, and Tajikistan. Certain countries, such as
Belarus, Moldova, Russia, and Tajikistan did not have a scalable poverty‐targeted
program before the crisis and hence would require significant institutional capacity
building. Since then, however, both Moldova and Tajikistan have taken steps to
introduce a targeted program, build institutional capacity and /or consolidate social
benefits. On the other hand, while Bosnia and Herzegovina and Kazakhstan had well‐
targeted LRSA programs, their coverage was extremely low. In the case of Latvia, the
poverty‐targeted program was only moderately well‐targeted with low coverage and
very low generosity.
Note that some of the ECA countries that fall under Categories C or D still have a moderate
base for response as defined by “operating one or more progressively targeted programs to
build on.”
The more prepared a country is with respect to its social benefits system, the more likely
the system is to be effective in mitigating the effects of a systemic shock. Hence,
assuming everything else (such as the severity of the crisis, how it manifests itself, and any
political economy factors) is constant; we would expect to see an effective social benefits
response in countries in category A, while we would expect a much less effective response
in countries in category D.
The crisis may also present an opportunity for the government to introduce structural
reforms by introducing a well‐targeted program, rationalizing and consolidating the existing
safety net, or strengthening the system in some way. Even though designing and fully
implementing a new program can take at least 10 to 18 months (often more), the medium‐
term rewards cannot be ignored, particularly in terms of increasing the system’s
preparedness for a potential future crisis. The previous food and fuel price increases
prompted the governments of both Moldova and Tajikistan to introduce structural reforms
of their safety nets.
The Severity of the Crisis (2009‐2010)
Conceptually, the more severe the impact of a crisis in a country, the more it can be
expected that social benefits will be needed in response. In this context, we examine the
severity of the recent global financial crisis in ECA countries to identify how social benefits
might have been expected to respond.
13
The ECA region was hit harder than any other region by the recent global crisis in terms of
GDP contraction, but the extent of this impact varied significantly from country to
country. While global GDP contracted by around 2.2 percent in 2009,15 in ECA countries, the
average contraction was more than 5.0 percent (Figure 3). Eastern European countries were
particularly affected, with Estonia, Latvia, Lithuania, and Ukraine experiencing startling
economic contractions of 14 to 18 percent (Figure 4). The severity of the crisis varied,
however, with most Central Asian countries seemingly escaping the worst of the crisis.
Simulations indicate that by the end of 2010 there might have been 10 million more poor
people in Europe and Central Asia than were estimated in pre‐crisis baseline projections.16
This increase in poverty would correspond to 20 percent of the ECA population who rose
out of poverty between 1998 and 2006 (Tiongson et al, 2010).
Figure 3: GDP growth (annual percentage change)
15 This was the first contraction in global GDP since World War II. 16This number appears to be quite large. However it is important to note that poverty in ECA tends to be shallow, with large groups of individuals clustered around the poverty line and hence likely to fall into poverty even after only a moderate shock to their income. Recent poverty estimates indicate that, contrary to expectations poverty remained mostly unchanged in 2009 and slightly declined in 2010. This is partly due to the role of social assistance and pensions in mitigating the reduction in household labor income. Please see “Poverty in ECA: Understanding recent trends” April 2012, for more details.
14
Source: Staff calculations using IMF World Economic Outlook (WEO) Database April 2012
Note: Only low and middle income countries included based on World Bank classification of regions. LAC –
Latin American and the Caribbean, MNA – Middle East and North Africa, EAP – East Asia and the Pacific, and
SA – South Asia, SSA – Sub‐Saharan Africa.
The countries that were most affected with respect to GDP also experienced the highest
increases in unemployment (Figure 4).17 In most countries the decline in employment was
proportional to the magnitude of the GDP contraction.18 However, there were notable
exceptions, such as Latvia, Serbia, and Estonia, where the decrease in employment was
higher than the corresponding decline in GDP.19
Figure 4: Severity of Crisis: Change in GDP & Unemployment Rate 08 to 09‐‐ ECA
Countries
Source: Staff calculations using IMF WEO April 2012 and ILO
Note: No data available for Azerbaijan, Kyrgyz, Tajikistan, Turkmenistan and Uzbekistan (all of which had
positive GDP growths in 2009 and are part of the mildly affected group).
Using both measures, the countries of ECA can be categorized broadly into a typology of
three groups according to the severity of the crisis ‒ those who were severely affected,
17 It is important to note that overall unemployment numbers do not take into account other labor market adjustments such as increased part‐time employment, administrative leave, wage arrears, and temporary contracts. 18 According to World Bank, (2011), a 1 percent decline in GDP on average reflected a 0.5 percent decline in employment. 19 World Bank (2011a).
LV
UKR LTHARMEE
SI RUSRO HUMLDCRTMNT BG
SK TR CZ GRG SRB BIH
MKDBLR
KZK PLKOSALB
LV
UKR
LTH
ARM
EE
SIRUS
ROHUMLD
CRTMNT
BGSK TR
CZ
GRG
SRB
BIH
MKDBLR KZK
PL
KOS
ALB
‐20
‐15
‐10
‐5
0
5
10
15
GDP growth 2009 Change in unemployment rate (2008 to 2009)
15
those for whom the impact was moderate to high, and those who felt no impact. The Baltic
countries were among the severely hit, with unemployment rates in 2009 as high as 12 to
16 percent compared with only 6 to 8 percent before the crisis in 2008. There were also
large increases in unemployment in Armenia and Ukraine. On the other hand, most
countries in Central Asia and some in Eastern Europe experienced only limited effects of the
recent global financial crisis.
The SBO Plus sample includes a mixture of countries with respect to how severely they
were affected by the crisis (Figure 5). Four of the five severely hit countries and eight of the
fourteen moderate to high impact countries are included in the SBO Plus sample. The
sample only includes three out of the eleven countries that experienced a limited impact,20
but given that our focus is on how social benefits responded to the crisis any under‐
coverage of these countries in the data was not likely to be a problem for our analysis.
Figure 5: Wide Variation in Severity of the Crisis Across the ECA Region (SBO Plus Sample)
Source: IMF WEO, April 2012 and ILO
20 While it would have been more efficient to monitor only those countries who experienced significant (severe or moderate to high) crisis impacts, since the crisis monitoring was started in the early stages of the crisis it was not clear which countries would be impacted significantly.
LTV
UKR LTH ARM
ROM HNGCRT MNT BLG
GRG SRB
MKD
ALB KRG
LTV
UKR
LTH
ARMROM
HNGCRT
MNTBLG
GRG
SRB
MKD
ALB
‐20
‐15
‐10
‐5
0
5
10
15
GDP growth 2009 Change in unemployment rate (2008 to 2009)
16
As the recent global financial crisis progressed, immense fiscal pressures emerged in many
European and Central Asian countries. Public finances worsened in 2009, with fiscal deficits
increasing on average by 3.7 percent of GDP (Figure 6).21 These rising fiscal deficits have
limited the ability of governments to respond to the crisis. There is not much systematic
region‐wide analysis of political economy issues (and is thus on the Bank’s future knowledge
agenda in social protection), and these will be discussed later in the context of the actual
social benefits response in ECA countries.
While government finances decreased in most ECA countries, there were discernible
differences between countries. While fiscal balances worsened significantly in Russia,
Latvia, Lithuania, Montenegro, and Georgia with an average increase in the fiscal deficit of
10 percent of GDP between 2007 and 2009, Croatia, and Serbia appear to have weathered
the crisis without a huge deterioration in their finances (Figure 6).
Figure 6: Fiscal Balance: ECA Countries 2007‐2009
Source: IMF WEO April 2012
The proportion of the current fiscal balances that are structural versus cyclical is a vital
factor in whether or not each country needs short‐term fiscal consolidation. This in turn
21 World Bank (2011a)
‐10
‐5
0
5
10
15
20
25
2007 2008 2009
17
will affect how best to mitigate the negative effects of the crisis on household welfare and
the corresponding expected social benefits response. For instance, on average, structural
deficits account for more than three‐quarters of projected 2012 deficits in OECD
countries.22 In other words, in these countries, even with increased revenues in 2012,
spending will still be much higher than before the crisis. It is difficult to analyze the
structural proportion of the fiscal deficits in ECA countries. In any case, severe fiscal
pressures have limited the ability of ECA governments to respond to the crisis. There is little
to no room for expansionary fiscal policy, which has forced governments to execute a series
of cuts in expenditures. However, in most countries, social assistance spending was
protected (and increased in some cases) for allocative efficiency with the ultimate goal of
protecting existing and emerging vulnerable groups. Furthermore, certain social assistance
programs were frozen and/or eliminated as a first step towards the consolidation of
programs to improve the quality and increase the effectiveness of spending on safety nets.
In most cases, countries that were severely affected by the recent financial crisis also
faced the largest fiscal deteriorations (Figure 7). However, there are some notable
exceptions. For instance, while Montenegro and Georgia only experienced relatively
moderate GDP contractions, they experienced significant fiscal deterioration. In addition, a
large proportion of these deficits is structural. Hence, while countries may be fiscally
constrained overall, this is not necessarily due to the crisis alone, but also to a structural
component best addressed with structural reforms (such as expenditure rationalizations or
the consolidation of social benefits) than by across‐the‐board cuts in expenditures.
22 Organization for Economic Cooperation and Development (2011)
18
Figure 7: Severity of Crisis vs. Fiscal Constraints: GDP Growth 2009 & Change in Fiscal
Balance 2007 to 2009‐‐ SBO Plus Sample
Source : IMF WEO April 2012
Expected Crisis Response: Pre‐crisis Preparedness + Severity of the Crisis
Juxtaposing the degree of severity of the crisis with each country’s pre‐crisis preparedness
reveals important patterns for “expected crisis response.” Considering all the key factors
outlined above, ECA countries can be grouped into three categories in terms of what could
be expected of their social benefits response (the colors of the group labels correspond to
those in Table 1). These three categories set the stage to compare the expected response
of social benefits in each country with the actual response employed by each country.
Group 1 ‐ Response Needed & Expected. The first group consists of countries
where a social benefits response was very much needed and expected, but in some
countries (such as Georgia, Montenegro, and Armenia) worsening finances may have
constrained the expected social benefits response. This group corresponds to
countries that were moderately to severely hit by the crisis (with respect to GDP
contraction and unemployment), and whose social benefits systems were well
prepared to respond.
LTV
UKR LTHARM
ROM HNGCRT MNT BLG
GRGSRB
MKD
KRG ALB
LTV
UKR
LTH
ARMROM
HNG
CRT
MNT
BLG
GRG
SRBMKD
KRG
ALB
‐20
‐15
‐10
‐5
0
5
GDP growth 2009 Change in fiscal balance (2007 to 2009)
19
Group 2 ‐ Response Needed – but Less Prepared. The second group consists of
those countries which were severely‐to‐moderately affected by the recent global
financial crisis and were in need of a response, but whose social benefits systems
were not very well prepared to respond to the crisis. Hence, in this group of
countries, the social benefits system needed reforms before it could be expected to
respond. Fiscal pressures faced by this group of countries may have also reinforced
the need for reforms to improve the quality of spending.
Group 3 – Limited Response Needed or Expected. The last group of countries
consists of those that were affected only slightly by the recent crisis, so that only a
limited (if any) social benefits response was expected, regardless of pre‐crisis
preparedness.
Table 1: Do We Expect Social Benefits to Respond?
Typology of ECA Countries According to Severity of Crisis and Pre‐Crisis Preparedness
Source: Author
20
III. OBSERVED SOCIAL BENEFIT RESPONSES
Based on our analysis of the actual responses of the social benefits systems in ECA
countries, we created a typology to illustrate the observed vs. expected responses with
four main dimensions. The first dimension is the composition of the social benefits
response taking into account the type of programs that expanded during the crisis
(unemployment benefits, social assistance benefits). The second dimension is the
magnitude of the response as measured by the expansion in social benefits programs during
the crisis. The third dimension is the timing of the response, taking into account any lags in
the response by social benefits programs. The fourth and final dimension is the
sustainability of the response from the onset of the crisis to recovery, taking into account
the fact that the effects of a crisis tend to persist for a while. All of these dimensions have
important policy implications. Table 2 illustrates the composition and magnitude of each
country’s social benefits response, while Table 3 addresses the timing and sustainability of
the response in each country.
Table 2: Observed & Expected Responses of Social Benefits to the Global Crisis, ECA
Source: Author. Data for Observed Responses from World Bank Social Benefits Observatory, SBO Plus Sample
(including Hungary and Lithuania from EC/SPC report (2010). Expected Response classification from
framework in Section II, covering severity of the crisis + pre‐crisis preparedness.
21
Table 3: Timing and Duration of Responses of Social Benefits to the Global Crisis, ECA
Source: Ibid.
Overall, social benefits in the ECA region responded as expected vis‐à‐vis the severity of
the crisis23. For the most part, social benefits in countries that faced more extreme
contractions in GDP and employment responded in a robust manner to the crisis (shown in
orange in the first section of Table 2), while social benefits in countries with limited crisis
impact did not radically change course from pre‐crisis trends (shown in green in the lower
section of Table 2). One exception was Ukraine, which was severely hit by the crisis, but
significantly limited social benefits by tightening program rules (discussed in more detail in
the next section). Other social protection instruments, such as pensions, were also
employed in crisis response. Box 1 provides some examples of how pensions were utilized
in crisis response.
23 While it would have been ideal to analyze the observed responses of social benefits in all countries in ECA, data availability issues revealed this to be unfeasible. However, expected response framework is developed and utilized for all countries in ECA to provide the broad context in which to evaluate the observed responses.
22
Box 1: Pensions in Crisis Response: Government Measures and Potential Behavioral
Responses
Other Social Protection measures were also deployed to mitigate the impact of the crisis on
household welfare. Most notable of these were the various pension policies adopted in certain
countries. The generosity of the (minimum) pensions was increased in some countries including
Armenia and Turkey. A minimum pension was introduced in Romania to protect the low income
pensioners during the crisis. Pensions increased in Bulgaria as part of the planned spending in
January and July 2009, despite significant decreases in overall revenues (World Bank, 2012a). The
indexation of pension benefits (to wage growth) was suspended in Serbia during the crisis
preventing the decrease that would have occurred due to the decreasing wages.24 The non‐
contributory social pensions were also deployed. For instance in Georgia the monthly old age social
pension was increased from 70 to 75 GEL in January 2009 and was further increased to 80 GEL in
November 2009. However, this may not necessarily be in response to the crisis as there is a history
of ad‐hoc adjustments to old‐age pension in Georgia keeping the benefit more or less steady as a
share of average wage.
Such pension policies impacted a large segment of the population given the high coverage of
pension benefits in the ECA region due to historical high labor force participation rates that resulted
in large percentages of elderly receiving pension benefits. Such extensive coverage indicates
potential positive impacts on poverty. Combined with the relative administrative ease of
implementation for pension increases and the political economy, the pension policies are likely to
be an attractive SP instrument during the crisis. However, the fiscal cost of such policies –both short
and long term‐ may be quite high. For instance in Romania, the increases in pensions (20% annual
average increase from 2008 to 2009) were likely the main driver for the decrease in poverty in 2009
(the headcount poverty declined from 5.7% in 2008 to 4.4% in 2009) (Azam and Isik‐Dikmelik, 2010).
Similarly, in Bulgaria, the expansionary social protection policy in 2009, with pensions registering the
largest increases, to some degree offset the severe fall in household incomes due to the crisis
(World Bank, 2012a). However, the rapid increase in pensions added even more pressure to a
fiscally unsustainable pensions system in Romania and came at a cost of further tightening of the
country’s fiscal situation. While pensions helped cushion the crisis impacts, they were also relatively
costly. The contributory pensions by design are not necessarily targeted to the poorer segments of
the population. Hence, if the objective were to provide protection to poor (or at risk of being poor)
similar welfare gains could have been achieved at lower fiscal cost through poverty focused social
assistance programs.
On the other hand, some evidence of certain behavioral responses was observed during the crisis.
For instance the number of workers that applied for early retirement benefits increased in Romania
24World Bank (2009).
23
during the crisis. People facing unemployment were probably looking to retire early or they were
rushing to retirement thinking that they might be losing the privileges in the current system with the
pension reform. Such retirees bring short‐term fiscal impact as well as medium to long‐term impact
as they are paid pensions for longer periods of time. Another program that experienced increased
number of beneficiaries was the disability program. Both in Latvia and Romania, the number of
disability pension beneficiaries increased following the crisis. In Romania, the main increase came
from the third category, where abuse can be expected to be most prevalent. Similar to early
retirement, the number of disability beneficiaries has probably increased, as people view the
disability benefit as an option for facing unemployment.
Pre‐crisis preparedness influenced the strength and timing of response of social benefits
in some countries. In most countries (for which we have actual response data) in the
“response needed and expected category,” benefits coverage increased almost immediately
or with a very short lag, as shown in Table 3. For example, the social benefits system in
Hungary was reasonably well‐prepared for the crisis with respect to the composition,
performance, and coverage of benefits. Similarly, the overall social benefits system in
Montenegro was reasonably well‐prepared to respond to the crisis in terms of its
composition and performance; however, fiscal pressures may have contributed to the
system’s actual limited response. In Latvia, which was among the least prepared, benefits
response lagged 6‐8 months from the onset of the crisis because reforms were needed to
strengthen social programs. However, once Latvia undertook the reforms, benefits
responded in a decisive and sustained manner (discussed in detail in the next section). For
other countries that lacked any targeted safety net program at the onset of the crisis, such
as Moldova and Tajikistan, response to the crisis was limited (and they are currently
developing such programs and registry tools now to be in a better position to assist the
poor in the next crisis and beyond).
In most cases, unemployment benefits represented the first line of response.25
Unemployment benefits responded in all crisis‐affected countries in the SBO Plus sample
that operated them (as shown in Table 2). The response was virtually immediate in all such
countries, however, the level of response varied. Hungary, Armenia, Romania, Lithuania,
Ukraine and Bulgaria (Annex 2) expanded coverage rapidly and decisively to mitigate the
labor market impacts of the crisis. Croatia, Montenegro, and Serbia moderately expanded
unemployment benefits (as observed in Table 3). It is important to note that overall
25Kuddo (2010a) analyzed labor market data from 26 countries and found that in 21 of those countries unemployment benefits expanded in 2009 compared to 2008. This expansion was immediate in most countries, beginning as early as the last quarter of 2008 in some countries.
24
coverage of unemployment benefits is rather low in ECA, due to the fact that such schemes
usually only apply to formal sector workers.26 Similarly, in most EU countries the
unemployment benefits were the first to respond to the crisis; however the coverage was
more robust due to high levels of formality. Box 1 presents a brief summary of the social
benefits response in old EU countries.
In many countries the expansion in unemployment benefits was greater than the increase
in registered unemployed. Armenia, Romania, Montenegro, and Bulgaria experienced
greater increases in unemployment benefits than the corresponding rise in the registered
unemployed. In Bulgaria, there were approximately 50 percent more registered
unemployed in 2009 compared with the pre‐crisis level in 2008; however, the coverage of
unemployment benefits more than doubled during the same period. This large expansion in
unemployment benefit recipients might have been due to the low coverage of
unemployment benefits among the existing unemployed (assuming these mostly consisted
of long‐term, youth or, informal unemployed workers), while coverage among newly
unemployed was probably much higher. The expansion in unemployment benefits appears
to have been sustained, which is not surprising in light of the continued high levels of
unemployment throughout 2010.
In several countries, the effects of the crisis extended beyond the duration of benefits,
and unemployment benefits started running out before the onset of recovery. This was
the case in Lithuania, Bulgaria, Romania, Hungary, and Ukraine (See Annex 2). The most
notable exception was Ukraine where unemployment benefits were immediately and
substantially expanded, but fiscal pressures subsequently led to tightening of the eligibility
criteria. In some cases, such as Lithuania and Hungary, social assistance benefits picked up
(or continued expanding), as people shifted from unemployment insurance to social
assistance benefits. The country context, especially the distributional impact of the crisis,
appeared to be the most important factor in the timing of the response. For instance in
Lithuania, the initial impact of the crisis was borne by unemployment benefits, which was a
sign that formal workers who were eligible for these benefits were the first to be negatively
affected by the changes in the labor market. This is not surprising, as almost 80 percent of
the workforce in Lithuania is employed in the formal sector.27 However, the labor market is
likely to take a long time to recover, and as unemployment benefits can only be claimed for
6 to 11 months, many of those recipients who could not find a job began to claim social
assistance. It is important to note that at the start of the recovery the only indication is an
overall improvement in the macro framework, which is not necessarily manifested in the
26 Kuddo (2009). 27 Ibid.
25
labor market or in household welfare immediately. The positive effects of the recovery are
likely to take longer to filter down to the household level, which means that governments
need to sustain their crisis responses well after the recovery begins.
Social assistance benefits also kicked in as crisis response mechanisms, though with
generally more moderate magnitude and with some lag in response. Coverage of last
resort social assistance (LRSA) programs expanded decisively in Hungary, Lithuania, and
Latvia (see Annex 2) – though with an eight‐month time lag in Latvia as previously
mentioned due to the need for reforms. In Hungary and Lithuania, LRSA programs
continued expanding in a sustained manner throughout the crisis, likely also compensating
for the expiration of unemployment benefits (as noted above). Social assistance programs
responded more moderately in Armenia, Croatia, Montenegro, Georgia, Romania, Bulgaria
and Serbia (Tables 2 and 3). In Armenia, the increase in overall coverage of the LRSA
program (Family Benefit) was modest, but the Government managed to increase coverage
of the poor, through a decisive effort to improve targeting accuracy (reducing leakage to the
non‐poor to make space for the eligible poor) and to reduce fraud and error.28 The increase
in coverage of the Targeted Social Assistance program in Georgia (see Annex 2) was
moderate but important, as the country does not operate an unemployment benefits
scheme. In contrast to the other countries in the SBO Plus sample, LRSA benefits in Ukraine
not only failed to respond to the crisis, they continued their pre‐crisis declining trend –
despite the fact that Ukraine was severely hit by the crisis. This exception is discussed in
more detail in the next section.
Tight eligibility thresholds seem to have moderated the response of LRSA benefits. Most
ECA countries operate very tightly targeted last‐resort social assistance benefits, commonly
in the form of “Guaranteed Minimum Income” (GMI) schemes (e.g., Ukraine, Croatia,
Montenegro, Hungary, Lithuania, Latvia, Romania, Bulgaria, and Serbia). GMI schemes
tailor benefits to each household according to the difference between their observed
incomes and some guaranteed minimum income level. In many cases, the eligibility
thresholds for these benefits are very tight, and have declined over time with inflation. This
has resulted in a “marginalization” of GMI programs in many ECA countries – programs that
become so narrowly targeted their coverage is tiny and political support low. GMI schemes
also face the challenge that, to increase coverage via an increase in the threshold, the
Government usually must also increase benefit amounts to all recipients (including those
already on the program). As such, they tend to be rather inflexible for crisis response – and
28 For instance Armenia employed measures to improve the targeting efficiency of its signature last‐resort and targeted social safety net called Family Benefit Program (FBP), which led to a clean‐up of the registries and allowed for more of the truly eligible to be covered by the program. (World Bank 2010).
26
generally lack the support of the Ministries of Finance due to this need for a simultaneous
increase in coverage and benefits.
Box 2: Response of Social Benefits to the Crisis in the EU
Social benefit systems in ECA countries were not alone in being challenged during the economic
crisis; countries in what is often dubbed the EU15 were also tested. A recent joint report by the
Social Protection Committee and the European Commission outlined the response of EU nations
to the recent economic crisis using monthly beneficiary data similar to that utilized in this paper.
In this box, we look in depth at the diversity of experience and the trends in program scale in
some of the old EU nations to provide context for the discussion of responses to the crisis in
ECA.
Different Challenges,
Different Channels
As in ECA, the EU15
has seen a variety of
different responses to
the crisis in terms of
social benefits. In
some countries, the
crisis seems to have
had a major impact
on social benefit
schemes primarily
through the channel
of unemployment
insurance and assistance. The United Kingdom is a prime example, where a crisis‐driven spike in
unemployment resulted in an immediate increase in claimants for the Job Seeker Allowance,
but not in other social assistance programs (see graph at right). Furthermore, a recurring trend
in old EU countries is the eventual drop‐off in claimants for unemployment benefits despite
continued high levels of unemployment; this is visible in the UK graph here, and also in the data
from Spain and the Netherlands.
500
1000
1500
2000
2500
3000
2006 2007 2008 2009 2010
Persons (thousands)
United Kingdom: Social Benefits (2006‐2010)
# Unemployed Job Seeker Allowance Claimants Incap/Inval Benefits
Disability Living Allowance Income Support Claimants
27
However, not all
nations shared this
experience. For
example, in Germany
unemployment
actually fell – but
short‐term work rose
dramatically. In other
words, the crisis hit
and people stayed
employed, but
worked more in
short‐term positions.
Taken together, these
two trends kept the number of individuals receiving unemployment benefits relatively stable
through the crisis years (see the above graph). Contrasting the German and British experiences
provides a valuable lesson for those who seek to assess the impact of a crisis on social benefits –
it is critical to consider all the potential ways in which a crisis can impact the economic behavior
of a given population.
While the UK and German social benefits systems responded to the crisis primarily through the
unemployment benefits channel (and not at all through increases in social assistance), in other
old EU countries the response was more uniform across schemes. That is, the crisis led both to
an expansion of unemployment benefits and social assistance programs (examples include
France and Portugal).
Fiscal Responses
The recent crisis naturally led to the weakening of fiscal positions in the old EU, through both
automatic stabilizers and stimulus packages. This has forced countries to look into medium and
long‐term fiscal consolidation policies, many of which will affect the provision of social benefits.
Some reforms will have direct impacts on social benefits. For example, Denmark has shortened
the benefit period for unemployed workers, while Ireland and the UK have reduced staff levels
in public service sectors of government. Other reforms will influence social benefit provision in
more indirect ways – these include the various fuel and VAT tax increases throughout the
region. Though some old EU countries (Ireland, the UK, Denmark, and Portugal, for example)
have conducted impact assessments to determine the distributional impact of these policies
changes, there is more work to be done to determine how fiscal consolidation will affect the
most vulnerable.
Source: Council Of European Union (2010)
0
1000
2000
3000
4000
5000
6000
7000
2006 2007 2008 2009 2010
Person
s (Tho
usands)
Germany: Social Benefits (2006‐2010)
# Unemployed Unemp Benefit Recipients Short‐Term Work SA Recipients
28
Somewhat unexpectedly, coverage of disability assistance benefits also increased during
the crisis. Most ECA countries operate non‐contributory disability benefits schemes as part
of their social safety net.29 From 2008 to 2009, the number of beneficiaries of disability
assistance increased in several countries, raising the question of whether the expansion
arose as a result of pressures from the global economic crisis or due to some other cause.
Although disability benefits in some of these countries are means‐tested and thus have the
potential to expand during times of crisis, means‐tested disability benefits only constitute a
small share in our sample (only four out of fourteen). Also, most of the non‐contributory
disability benefits are categorical, meaning eligibility is based solely on disability status, with
no additional income or living standards qualifications. Data from the Social Benefits
Observatory (SBO Plus sample) indicate that the monthly number of beneficiaries of
disability assistance increased by an average of 5 percent from 2008‐09 in countries
affected by the crisis that have disability benefit programs (Armenia, Serbia, and
Montenegro (Figure 8 below).30 This observation is puzzling as there is no discernible
reason why an economic crisis – as opposed to wars, earthquakes, or other natural disasters
– would increase disability rates. This phenomenon was also observed in the United States
(see Box 3), where people sought certification for disability status in order to receive
benefits as their unemployment benefits ran out. Lax disability certification criteria – and
generous benefits – were blamed for the crisis‐driven expansion in disability benefits in the
US. Similarly, coverage of disability benefits expanded significantly in FYR Macedonia as well
as in Albania over the crisis period, even though these two countries were not strongly
affected by the global recession.
29 Among the 14 countries that are covered by the Social Benefits Observatory, non‐contributory disability benefits exist in all of countries. In ECA, more than half of the countries have a non‐contributory disability benefit program as part of their safety net. 30 Coverage of disability benefits did not decline in any of the countries in our sample over that same period (despite fiscal pressures).
29
Figure 8: Disability Benefits Indexed Number of Beneficiaries
Source: SBO
While it is difficult to identify the exact reasons behind the expansion in the disability
programs during the crisis, we can speculate on some likely causes based on the common
characteristics of the programs. First, some people who met the disability criteria but had
not bothered to apply before the crisis may have needed the extra income after the crisis
began to negatively affect their welfare. Second, disability benefits are typically more
generous than other types of social assistance benefits, which is understandable given the
large costs involved in living with a disability, and this makes these benefits much more
attractive to those who qualify for them than, for example, LRSA benefits. Third, the
medical certification process in some of these programs may not be very rigorous, allowing
errors of inclusion to occur, particularly in the case of disabilities that can be hard to
disprove, such as depression. Finally, the lack of a recertification requirement or the
infrequency of recertification may also increase the likelihood of false applications because,
although the chances of approval may be low and the process arduous, the payoff is
generous and the benefits will then be received for years to come.
Overall, the administrative data collected in the Social Benefits Observatory (SBO Plus
Sample) revealed four important findings with respect to the response of social benefits
0
20
40
60
80
100
120
140
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11
2008 2009 2010
Bulgaria Disability Benefits Macedonia Disability/Permanent SA (MT)
Croatia Disability Allowance (MT) Montenegro Disability Benefit
Kyrgyz Rep MSB (disability only) Albania Disability Benefit (excl caregiver)
Armenia Disability benefit (Social Pension) Serbia Disability (caregiver's) allowance)
30
to the global crisis in ECA. First, pre‐crisis preparedness influenced the ability of social
benefits to respond to crisis in some countries. Second, unemployment benefits were
generally the first line of response in countries that operated them. Third, social assistance
responded by increasing coverage during the crisis (in addition to the protection they afford
to existing beneficiaries). Finally, disability benefits also expanded during the crisis period, a
somewhat unexpected response.
IV. LESSONS LEARNED: THE IMPORTANCE OF ACTION
The response of social benefits programs in ECA to the crisis provided insightful lessons on
how such programs can mitigate crises. Although, for the most part, social benefits appear
to have responded as expected, there were notable exceptions. In some countries, such as
Ukraine, the response was below what would have been expected – both in magnitude and
composition. In others, such as Latvia and Lithuania, the response far exceeded
expectations. While it is difficult to pinpoint the exact reasons for these differences, the
specifics of the country conditions and the programs reveal some common themes. These
include how structural reform, design, and implementation affected the success of social
benefits programs. We illustrate these lessons by comparing country experiences.
Structural Reform – The Importance of Timely Intervention
Latvia and Ukraine were both substantially affected by the crisis and faced similar fiscal
challenges. GDP contracted in Latvia and Ukraine by 17 percent and 15 percent
respectively, and both countries’ labor markets were hit hard. In addition, the fiscal
pressures were immense in both countries, with a fiscal deterioration between 2007 and
2009 of 7 percent and 4 percent of their respective GDPs.
However, Latvia’s social benefits system was much less prepared than Ukraine’s. Latvia’s
social benefits system appeared to be much less prepared for a crisis, given the composition
and the performance of its social benefits programs pre‐crisis (see Annex 1 for details on
the classification of “pre‐crisis preparedness”). In contrast, Ukraine operated a well‐
targeted LRSA program (but with low coverage), in addition to unemployment benefits and
a means‐tested child allowance program with large‐scale coverage. Our ex‐ante analysis
concluded that Ukraine’s social benefits system would be able to respond to the crisis more
readily, whereas Latvia’s system would only be able to respond after being reformed.
31
In practice, however, Latvia’s response exceeded expectations (albeit with a time lag),
while Ukraine’s response to the crisis was highly constrained. In Latvia, after an initial
time lag, unemployment and social assistance benefits responded robustly to the increased
number of claims, and this expansion has been sustained during the crisis. On the other
hand, in Ukraine, although unemployment benefits expanded initially, this was not
sustained as fiscal pressures led the government to tighten the eligibility criteria. Moreover,
Ukraine’s well‐targeted LRSA program continued its long‐term declining trend in coverage.
To improve its ability to respond to the crisis, Latvia undertook serious structural reforms
of its social benefits system31. The Latvian Government introduced fiscal adjustments at the
beginning of 2009 to stabilize and reorient expenditures, and to prepare the safety net to
meet the increasing needs of households. Prior to the crisis, Latvia spent only 0.8 percent of
its GDP on safety nets (excluding unemployment benefits, which constituted another
%0.35).32 Latvia also undertook other reforms to strengthen the functioning of the safety
net itself. These structural adjustments allowed the Government to increase the coverage
of the Guaranteed Minimum Income Program – albeit with the time lag ‐‐ and to introduce
a Public Works program to mitigate the negative effects of the crisis.
In contrast, Ukraine did not seize this opportunity to increase the efficiency of its social
benefit spending. Ukraine spent 2 percent of its GDP on safety nets, as of 2008, of which
1.3 percent was allocated to non‐means tested categorical benefits programs (Figure 20).
Although Ukraine faced immense fiscal pressures, there was room for improving the quality
and increasing the effectiveness of safety net spending by rationalizing expenditures or
consolidating benefits, which might have resulted in fiscal savings. Instead, Ukraine
maintained an increasing share of spending on non‐means tested programs, while the
coverage and spending for the LRSA program – which was oriented to the poor – continued
on a declining trend.33
31 Harrold, Sinnot, and Santos (forthcoming) presents a nice overview of the fiscal adjustment in the social sectors in Latvia during the global crisis. 32 ECA SP Database. 33In fact this trend is observed across the region. In most ECA countries, the overall spending on safety nets increased during times of prosperity in 2000s (both in real terms and as a share of GDP), with the composition of the spending shifting towards categorical programs (which are more widely supported by the median voter) in some countries whereas targeted anti‐poverty programs shrank over time. Public perceptions of poverty, social benefits, work, and work effort no doubt played a role in this shift to non‐targeted (categorical) programs before the crisis, which led to the marginalization of the kind of programs that could have been used in an effective crisis response.
32
Political economy factors likely contributed to the constrained response in Ukraine. The
availability and performance of certain social benefits during a crisis can be directly affected
by political economy factors. Public perceptions about the role of the state in promoting
equality and existing attitudes towards the poor and disadvantaged have implications for
the size and composition of the social benefits, as well as their design and implementation
features. This in turn can hinder the pre‐crisis preparedness of the social benefits system
and prevent the passage of the tough structural reforms that need to be undertaken during
crises.
Figure 9: Both Ukraine and Latvia had Room for Improving the Quality of Social Safety Net
Spending
Source: ECA SP Database
Design Matters – the Importance of Eligibility Criteria
The social benefits systems in Serbia and Bulgaria were similar in their level of pre‐crisis
preparedness, in how they were affected by the crisis, and in the expansion of their LRSA
programs in response. Bulgaria and Serbia both experienced relatively moderate GDP
contractions during the crisis (3.5 percent and 5.5 percent respectively). Moreover, their
pre‐crisis social benefits systems were among the less prepared in ECA. This is due to
extremely low coverage of their LRSA programs, and/or the relatively weak performance of
their child allowance programs (where they existed) compared to the better prepared
countries. In both countries, the LRSA programs expanded during the crisis, with the
33
Bulgarian GMI reversing its long‐term declining trend, while the Serbian Material Support
Program continued to increase in coverage over time.
However, the response appeared to have been much more flexible in Serbia because of
certain design elements for eligibility. In Bulgaria (like several other countries in the
region), the LRSA program had been shrinking over time. This was mostly due to the low
eligibility thresholds that were not adjusted to keep up with inflation, in conjunction with
the introduction of an “18‐month rule,” that limited the length of time an individual could
claim benefits (but which was abolished as of January 2011). These design factors resulted
in declining coverage and the marginalization of the program. On the other hand, the MOP
in Serbia had been expanding over time, largely due to the fact that the income eligibility
threshold was linked to the monthly consumer price index (CPI) by law. The indexation of
the eligibility threshold prevented the program from becoming marginalized, while keeping
the purchasing power of the benefit stable over time. While the GMI in Bulgaria expanded
only slightly and with some delay, the MOP program in Serbia appears to have responded
flexibly during the crisis (expanding more in 2009 than in 2008), partly due to built‐in design
features. Increasing the coverage of a well‐targeted program during a crisis is likely to take
some time and a concerted administrative effort. Design elements, such as low non‐indexed
eligibility thresholds, that lead to excessive marginalization of programs introduce
inflexibility in crisis response.
Implementation Matters – the Importance of Institutional Arrangements and Financing
Sources
Institutional and financing arrangements for social benefit programs can constrain their
ability to respond effectively to a crisis. While causality cannot be established,
observations in various countries suggests that centrally‐financed and ‐administered
programs (with eligibility criteria and selection determined centrally) have responded to the
crisis much better than the programs whose financing and administration are decentralized.
This was certainly the experience of the United States. The Supplemental Nutrition
Assistance Program (otherwise known as the Food Stamps program), a federal program
financed and administered centrally, has responded quite well to the crisis. In contrast, the
Temporary Assistance to Needy Families (TANF) program, which is co‐financed and
administered by the states, has not been quite so responsive (see Box 3 for more details
and graphs). While other factors may have contributed to this stark difference in response
(including workfare requirements for TANF), it appears that fiscal pressures at the state
level contributed to the lesser response by TANF – and encouraged local welfare offices to
shift welfare applicants to the federally‐funded food stamps entitlement program.
34
In Latvia, previously decentralized financing arrangements for the GMI were revised to
prevent benefits from being rationed at the local level during the crisis. Pre‐crisis, the GMI
program in Latvia was completely decentralized, with local governments financing 100
percent of the benefits. However, local governments are required to carry a balanced
budget (as opposed to the central government which is allowed to run a deficit), which puts
considerable pressure on their budgets, especially during times of crisis. This can lead local
governments to ration benefits at the local level. In Latvia, because of this balanced budget
constraint, local governments referred the LRSA applicants to public works programs that
were centrally financed.34 Recognizing this constraint, the central government then revised
the financing arrangements for the GMI and put in place a central government financing
guarantee (50 percent co‐financing) to supplement the municipalities’ spending on the GMI,
which eased the pressure on local governments. Without this change in the financing
arrangements (even with the same increase in the eligibility threshold), the expansion in the
program would probably have been much smaller.
34 World Bank (2011).
35
Box 3: The Response of Social Benefit Programs in the US during the Crisis
The impacts of the global crisis were felt deeply in the United States. The poverty rate across the
country increased from 13.3 to 14.3 percent)*. The labor market impacts were also severe and
continue to persist with the current unemployment rate hovering above 9% as compared to the pre‐
crisis rate of 5.3%.** As expected, social benefits expanded during the crisis. However, the degree of
response varied across different programs, highlighting the importance of implementation issues,
such as institutional and financing arrangements.
The first figure illustrates
this point clearly. The
Supplemental Nutrition
Assistance Program, a
federal program
financed and
administered centrally,
has responded well to
the crisis. The number of
SNAP recipients
expanded by more than
40% during the crisis
(October 2008 to
September 2010). In
contrast, the Temporary
Assistance to Needy Families (TANF) program, which is co‐financed and administered by the states,
has not been as flexible.
The coverage of TANF
increased by about 13%
during the same time
period). It is likely that
other factors contributed
to this stark difference in
response (including
workfare requirements
for TANF). However,
fiscal pressures at the
state level likely played a
role in the lesser
response by TANF – and
36
encouraged local welfare offices to shift welfare applicants to the federally‐funded food stamps
entitlement program.
Similar to several other countries analyzed in this study, disability benefits in the US also expanded
during the crisis (second figure). The number of beneficiaries increased by almost 10% by September
2010 (by almost 20% by April 2011) compared to pre‐crisis levels. This is likely a behavioral response
to the crisis by the affected workers. Those workers, who are on the border with respect to their
degree of disability, may apply for benefits when the downturn leaves them jobless and new jobs are
hard to find.
* “Poverty: 2008 and 2009”. American Community Survey Briefs September 2010, United States Census
Bureau.
** United States Bureau of Labor Statistics, http://www.bls.gov/
cps/prev_yrs.htm
V. CONCLUDING REMARKS AND POLICY RECOMMENDATIONS
This study traced the actual responses of social benefits during the recent global financial
crisis for 14 ECA countries. It also presented a conceptual framework to categorize
expected social benefits responses, taking into account the pre‐crisis preparedness of social
benefits systems, and the severity of the shocks.
Overall, the study found that social benefits responded largely as expected. The response
was typically more decisive in countries hit severely by the crisis, and much less so in those
that were less affected by the crisis. Pre‐crisis preparedness influenced the ability of social
benefits to respond to crisis in some countries. Not surprisingly, unemployment benefits
responded first. Unemployment benefits tended to expand rapidly with the spread of the
crisis, though their coverage is generally limited to formal‐sector workers, making them a
partial crisis‐response tool. Social assistance benefits also responded by increasing
coverage during the crisis (in addition to the protection they afford to existing
beneficiaries). The response of last resort social assistance benefits was more moderate
than unemployment benefits in many cases, perhaps due to the tight eligibility thresholds
for the common GMI programs (which have eroded over time with inflation). Interestingly,
the study also found that coverage of disability benefits increased in some countries,
representing a somewhat unexpected response since disability status would not
automatically be correlated with economic shocks.
37
Several lessons can be learned from the variations in social benefits responses across
countries. Immense fiscal pressures were an important constraint; but the ability of
Governments to undertake tough structural reforms made a difference in effective social
benefits response. While fiscal pressures faced cannot be underestimated, crisis presents an
opportunity to undertake difficult, but needed reforms, which in turn will help in
preparedness and the effective mitigation for the next potential crisis. Design matters for
flexible response of social benefits. For instance the response of LRSA in Serbia indicates
that indexation of eligibility thresholds for poverty‐targeted programs (as applicable) is
important to ensure flexible response. Finally, implementation matters in ensuring social
benefits can respond flexibly to a crisis. The experience of the US and other countries
indicate that the institutional and financing arrangements for social benefit programs can
constrain their ability to effectively respond to the crisis. Minimizing the local government’s
financing burden for social benefits will help to prevent rationing of these benefits during
crisis.
While many in ECA had social benefits programs ready to address the crisis, challenges
remain to ensure that all ECA countries are prepared in the future. In terms of outcomes,
poverty impacts of public spending on social assistance are limited in many ECA countries
due to low coverage of the poor35 and inadequate benefits levels. This low coverage is more
prominent among lower‐income countries, but coverage gaps also persist in several higher‐
income countries. Eligibility thresholds for poverty‐targeted programs are commonly very
rigid and too narrow, which result in large errors of exclusion and limit crisis responses.
Benefits are also fragmented, as most countries operate numerous programs with diverse
application procedures and eligibility rules. Navigating multiple schemes is costly for the
beneficiaries and parallel administration is costly for governments; gaps and duplications
abound. Benefits administration in many instances is also weak, with paper‐based records
and inadequate management information systems or oversight and controls.
With a renewed economic downturn, social safety nets and unemployment insurance will
once again be called upon to alleviate crisis impacts, which will likely be transmitted
through labor markets that have not fully recovered from the 2008‐10 crisis (persistent
unemployment). Fiscal constraints mean that many ECA governments will be under
increasing pressure to cut back on spending, indicating the need to improve effectiveness
and consolidate and target social (including universal) benefits, which could create tensions
35 Penny Williams, et al., “Social Safety Nets in Europe and Central Asia; Preparing for Crisis, Adapting to Demographic Change, and Promoting Employability,” ECA Knowledge Brief Vol.48 (Washington, DC: World Bank).
38
with the middle classes. Sustained structural investments in strengthening social safety nets
are thus warranted, even for higher‐capacity countries.
Governments should not wait for an economic crisis to build and/or strengthen safety
nets.36 The creation of safety nets requires time and political effort. It takes time to tailor
the design and institutional arrangements for an effective safety net system. In addition,
political economy issues play an important role in redistributive reforms. Building consensus
on the reforms is imperative, but time‐consuming. To assist in crisis‐preparedness,
governments can consolidate benefits, harmonize program rules and governance, and
invest in efficiency improvements in benefit administration systems.
First, consolidating benefits into fewer programs should be focused on clear objectives37,
social policy goals, and priorities, and design features should be simplified and harmonized
across programs. Assistance, including many universal benefits, should be focused on those
in need, while last‐resort programs should be expanded to cover a larger share of the poor
and vulnerable.
Second, governments can continue investing in systems38 and tools to improve
implementation efficiency, such as unified applications and intake processes, automated
integrated management information systems, unified payments systems, strengthened
oversight and controls, monitoring and evaluation, and clarified institutional rules, roles,
and financing with investments in capacity. Timely intervention is best for effective crisis
response but cannot be achieved without regular monitoring. In the absence of monitoring
data it is very difficult assess how the programs are progressing, whether they are reaching
their objectives, and whether there is any intervention needed during economic shocks.
Investments to strengthen the benefit administration are important in crisis response. If
monitoring systems are not already in place, it would be prudent to establish them, which in
turn will help with evidence based policy making.
Finally, governments can strengthen the links between benefits, employment, and social
services to reduce dependency and promote employability. Diagnostic tools and case
management can help identify and tailor activation approaches to target groups.
Governments should also seek to promote these linkages by investing in capacity,
36 Also emphasized in the Development Committee paper on Safety Nets for 2012 WB‐IMF Spring Meetings. (Please see World Bank 2012c). 37 Clear objectives are also important for World Bank’s support of safety nets as highlighted in IEG (2011). evaluation). 38 The recent IEG evaluation of World Bank’s support to Safety Nets also emphasizes the importance of a systems approach (please see IEG, 2011).
39
integrating service delivery, and optimizing institutional incentives for effective
employment support. Consolidation of benefits and investment in efficiency will continue to
strengthen social safety net systems across ECA and prepare countries to respond to future
crises.
40
REFERENCES
Alesina, A. and Glaeser, E., (2004). “Fighting Poverty in the US and Europe.” New York,
Oxford University Press
Atinc, T. M. (2003), “How the East Asian Crisis Changed the Social Situation,” in Katherine Marshall and Olivier Butzbach (eds.) New Social Policy Agendas for Europe and Asia: Challenges, Experience, and Lessons. Washington, D.C., World Bank
Azam, M. and Isik‐Dikmelik, A., (2010). “A Note on Recent Trends in Poverty in Romania.”
Washington D.C., World Bank
Council of the European Union, (2010). “2010 Update of the Joint Assessment by the SPC
and the European Commission of the Social Impact of the Economic Crisis and of Policy
Responses”, Brussels, 16905/10, Council of the European Union
Council of the European Union, (2012). “Third Report on the Social Impact of the Economic
Crisis and Ongoing Fiscal Consolidation 2011”, Brussels, 16905/10, Council of the European
Union
European Bank for Reconstruction and Development, (2011), “Life in Transition: After the
Crisis.” London, EBRD
Ferreira, F. and Schady N. (2008), “Aggregate Economic Shocks, Child Schooling and Child
Health.” Policy Research Working Paper No: 4701. Washington, DC: World Bank.
Figari, F., Salvatori, A., and Sutherland, H. (2010), “Economic Downturn and Stress Testing
European Welfare Systems.” ISER University of Essex
Fiszbein, A., Ringold, D. and Srinivasan, S. (2010), “Cash Transfers and the Crisis:
Opportunities and Limits.” Washington, D.C., World Bank
Frankenberg, E., Beegle, K., Sikoki B, and Thomas D. (1998), “Health, Family Planning, and Well‐being in Indonesia During an Economic Crisis.” DRU‐2013‐FGI/NICHD/UNFPA. Santa Monica, CA: RAND. Harrold, P., Sinnott, E., and Santos I. (forthcoming). “Fiscal Sustainability, Demographic
Change and Inequality: The Social Sectors from Crisis to Growth in Latvia” Washington, D.C.,
World Bank
41
IEG (Independent Evaluation Group). (2011). “Social Safety Nets: An Evaluation of World Bank Support, 2000–2010”. Washington, DC: Independent Evaluation Group, the World Bank Group. ILO/WORLD BANK. (2012). “Inventory of Policy Responses to the Financial and Economic Crisis”. International Monetary Fund (2011), “World Economic Outlook: Tensions from the Two‐Speed Recovery,” Washington, D.C.
Kuddo, A. (2009), “Informal sector in Europe and Central Asian Countries,” ECSHD Mimeo;
Washington, D.C., World Bank
Kuddo, A. (2010a), “Labor Market Monitoring in Eastern Europe and Central Asia Countries: Recent Trends (Round Three)”, Mimeo, Washington, D.C., World Bank
Leventi, C., Levy, H., Matsaganis, M., Paulus, A. and Sutherland, H. (2010), “Modelling the
Distributional Effects of Austerity Measures.” Social Europe Research Note, European
Commission, Brussels
Mitra, P., Selowsky, M., and Zalduendo, J. (2010) “Turmoil at Twenty: Recession, Recovery, and Reform in Central and Eastern Europe and the Former Soviet Union.” Washington, D.C., World Bank
Narayan, A., and Sanchez‐Paramo, C. (eds) (2012), “Knowing When You Do Not Know: Simulating the Poverty and Distributional Impacts of an Economic Crisis.” Washington, D.C., World Bank
Organization for Economic Cooperation and Development, (2011). “Economic Crisis and
Beyond: Social Policies for the Recovery.” Paris, OECD Ministerial Meeting on Social Policy
Ozdemir, E., Sanoussi, F., and Ward, T. (2010), “The Potential effects of the recession on the
risk of poverty”, Social Europe Research Note, European Commission, Brussels
Paxson, C., and Schady, N. (2005), “Child Health and Economic Crisis in Peru.” The World
Bank Economic Review Vol. 19(2): 203‐224.
Sundaram,R, and Isik‐Dikmelik A. (2011). “How ready are ECA’s safety nets to respond to
the renewed food crisis?” World Bank, Washington, D.C., mimeo.
Sundaram,R, and Isik‐Dikmelik A. (2010). “Technical Note on Proposed Reforms to Income
Tested Child Allowances in Romania.” Washington D.C., World Bank
42
Sundaram R., and Strokova V. forthcoming, “Overview of Social Assistance Programs in
ECA.” Washington D.C., World Bank
Thomas, D., Beegle, K., Frankenberg, E., Sikoki, B., Strauss, J., and Teruel, G. (2004), “Education in a Crisis.” Journal of Development Economics Vol. 74(1):53‐85
Tiongson, E.R., Sugawara, N., Sulla, V., Taylor A., Gueorguieva A., Levin V., and Subbarao K. (2010), “The Crisis Hits Home: Stress‐Testing Households in Europe and Central Asia.” Washington, D.C., World Bank
World Bank (2009a), “Pensions in Crisis: Europe and Central Asia Regional Policy Note,”
Washington, D.C., World Bank
World Bank (2009b), “Serbia Programmatic Public Expenditure Development Policy Loan”,
Program Document. Washington D.C. World Bank.
World Bank (2010), “Armenia: The 2008–09 Global Economic Crisis, Policy Responses, and
Household Coping Strategies.” Washington, D.C., World Bank
World Bank (2011a), “The Jobs Crisis: Household and Government Responses to the Great
Recession in Eastern Europe and Central Asia.” Washington, D.C., World Bank
World Bank (2011b), “Georgia: Poverty Dynamics since the Rose Revolution” Washington
D.C., World Bank
World Bank (2011c), “Romania Social Assistance Modernization Project: Project Appraisal
Document”. Washington D.C., World Bank
World Bank (2012d), “Country Assessments of Readiness of SSN Systems”, Brief.
Washington D.C., World Bank
World Bank (2012a), “Bulgaria: Household Welfare During the 2010 Recession and
Recovery”, Washington D.C., World Bank.
World Bank (2012b), “Resilience, Equity, and Opportunity: The World Bank’s Social
Protection & Labor Strategy, 2012‐2022.” Washington D.C., World Bank.
World Bank (2012c), “Safety Nets Work: During Crisis and Prosperity”, for Development
Committee Meeting. Washington D.C., World Bank.
43
Annex 1 ‐ Methodology on Pre‐Crisis Preparedness of SB Rankings39
In order to assess the pre‐crisis (as of 2008) preparedness of the Social Benefits systems in Europe
and Central Asia countries, we developed a composite score based on several dimensions. These
dimensions relate to three broad areas: the composition of social benefits (i.e. the existence of the
types of programs that could be leveraged in crisis response), the performance of social benefits
(with respect to overall performance indicators such as targeting accuracy, coverage etc.), and the
information capacity as an indicator of the efficacy of the administration of benefit programs.40
While these indicators cover a good portion of the relevant information, the resulting index is not a
precise measure. There are many other factors (such as quality of implementation, administrative
capacity to handle increase in applications, design issues such as ease of application etc.) that are
difficult to measure and therefore cannot be included. Moreover, composite indices can be
constructed in many different ways (simple average, weighted average etc.). However, the concept
was to sort countries based on the available social benefits information, and the index used in this
study reflects our best judgment.
Specifically, a composite score for each country was calculated using household survey data, as well
as administrative information. The composite score was comprised of nine dimensions covering: (a)
existence and performance of last resort social assistance (LRSA) programs (coverage, targeting
accuracy, generosity – all from household survey data, latest pre‐crisis year); (b) existence and
coverage of unemployment benefits (again, pre‐crisis); (c) existence and performance of child /
family allowance programs (coverage, targeting accuracy, generosity); and (c) information capacity
(degree to which administrative data on social benefits are available, since these are needed for
monitoring and expansion of social benefits). Details on each aspect of the composite score follow.
Existence and Performance of Last Resort Social Assistance Program
Whether the country has a last resort social assistance program before the crisis is considered first,
for two reasons: (a) such programs could be well‐positioned to scale up (in coverage and/or benefit
levels) in times of crisis; and (b) if they are well‐targeted, they are a more cost‐effective way of
channeling assistance to the poorest given that fiscal constraints could prevent expansion of more
universal programs in crisis times. However, the existence of such a program is not enough for
39 This methodology builds on the methodology developed for the ranking of ECA cash transfer programs with respect to their ability to respond to a future food crisis. As such, it draws heavily on the background note on this subject by Sundaram and Isik‐Dikmelik (2011) 40 The existence and coverage of unemployment benefits and a poverty‐targeted social assistance program would be one way of categorizing countries with respect to their pre‐crisis preparedness. However there is not enough variation across ECA countries to result in meaningful categories for pre‐crisis preparedness. Hence, more detailed indicators were used to construct the composite index to rank countries with respect to their pre‐crisis preparedness of social benefits.
44
effective response. Hence, the pre‐crisis performance of the program is estimated using household
surveys along three dimensions: (i) coverage defined as the percentage of the poorest quintile
covered by the program; (ii) targeting accuracy defined as the percentage of benefits that went to
the poorest quintile; and (iii) generosity estimated from household surveys is also considered. If the
program has low coverage then it would have limited impact, since increasing coverage takes time,
administrative capacity (to register and screen new applicants), political will, and fiscal space—with
fiscal space in particular being very tight in a crisis situation. Targeting accuracy indicates the
effectiveness with which the program transfers money to the poorest quintile. Again, improving
targeting accuracy requires a long‐term investment, and cannot be achieved very quickly in a crisis
situation. Finally, the generosity of transfers is considered – namely how effective are these
transfers in protecting beneficiary households from indigence? If the transfer amount is very low, it
will again limit the capacity of the LRSA program in protecting recipients during a crisis. Figures A1.1‐
A1.3 presents the performance of the LRSA programs in ECA across the above‐mentioned
dimensions.
Figure A.1.1. Coverage of LRSA programs
Source: ECA SP Database (Sundaram & Strokova, forthcoming)
05
1015202530354045
LRSACoverage of the Poorest Quintile (%)
45
Figure A.1.2. Targeting Accuracy of LRSA Programs
Source: ECA SP Database (Sundaram & Strokova, forthcoming)
Figure A.1.3. Generosity of LRSA Programs
Source: ECA SP Database (Sundaram & Strokova, forthcoming)
0102030405060708090
100
LRSAPercent of Total Benefits Received by the Poorest Quintile (%)
0
10
20
30
40
50
60
70
LRSABenefits as Percent of Post‐Transfer Consumption,
Beneficiary Household, Poorest Quintile
46
Existence and Coverage of Unemployment Benefits
Given the potential role of unemployment benefits as “automatic stabilizers” during a crisis
(especially a crisis with significant labor market impacts), the existence of unemployment benefits is
also an important dimension to consider in crisis‐preparedness. Most countries in ECA have some
sort of unemployment scheme in place. However, simple existence of the benefit is not enough,
whether all workers have access to the benefit when unemployed or not is also an important
dimension to consider. Clearly, the higher the degree of formality in a country the higher the
potential coverage of unemployment benefits would be. In order to approximate the coverage of
unemployment benefits in the labor market, we use the proportion of formal sectors workers
among the employed based on formality numbers estimated by Kuddo (2009). We acknowledge
that this is not a perfect measure41, but use it in the absence of standardized unemployment benefit
coverage data42. Table A1.1 presents the information on the existence of unemployment benefits
and the degree of formality in each country.
Table A1.1. Unemployment Benefits and Degree of Formality
Countries Unemployment Benefits (potential coverage)
UB exists Percentage formal
Armenia Yes 64
Kosovo No Program
Albania Yes 49
Latvia Yes 92
Moldova No Program
Russia Yes NA
FYR Macedonia Yes 46
Kyrgyz Republic Yes 42
Belarus Yes NA
Kazakhstan Yes 65
41 No matter how high the degree of formality in the labor market, the unemployment benefits will not cover all, such as new entrants to the labor market and the long term unemployed who will also be vulnerable during a crisis. 42 A similar performance indicator can be the ratio of people receiving Unemployment benefits among those who are unemployed. However, the ECA SP database currently doesn’t estimate this indicator as it is not a standard one. But this is one of the potential new indicators to add to ECA SP database.
47
Bosnia & Herzegovina Yes 36
Azerbaijan Yes 37
Tajikistan No Program
Georgia No Program
Montenegro Yes NA
Hungary Yes 95
Croatia Yes 75
Estonia Yes 95
Ukraine Yes 68
Turkey Yes 55
Poland Yes 85
Serbia Yes 46
Bulgaria Yes NA
Lithuania Yes 80
Romania Yes 53
Existence and Performance of Child/Family Allowance Programs
While a well‐targeted cash transfer that provides coverage for all the poor and/or unemployment
benefits that have high coverage for unemployed are first best options in terms of social benefits’
response, a well functioning child allowance program can also be used in crisis response. Hence, we
consider whether the country has at least one child allowance program since a well functioning child
allowance program can help at least to prevent or minimize the longer term and often irreversible
effects due to early childhood malnutrition, due to withdrawal of children from school, and so on. If
the country does have a child allowance program, we estimate the performance of this program
along the same three dimensions as LRSA: namely: coverage, targeting accuracy43, and generosity
of this program. Figures A1.4‐A1.6 presents the performance of the LRSA programs in ECA across the
above‐mentioned dimensions.
43 Note that looking at the ‘targeting accuracy’ of a universal child allowance program is not uncontroversial – universal child allowances could very well be a part of the social contract with people being willing to pay high taxes to support such a socially inclusive policy.
48
Figure A.1.1. Coverage of Child Allowance programs
Figure A.1.2. Targeting Accuracy of Child Allowance Programs
Source: ECA SP Database (Sundaram & Strokova, forthcoming)
020406080100
Means‐tested Non‐means‐tested
Main Family and Child BenefitsCoverage of the Poorest Quintile (%)
0102030405060708090
Means‐tested Non‐Means‐tested
Main Family and Child BenefitsPercent of Total Benefits Received by the Poorest Quintile (%)
49
Figure A.1.3. Generosity of Child Allowance Programs
Source: ECA SP Database (Sundaram & Strokova, forthcoming)
Information Capacity
The information capacity is also considered as an indicator of the efficacy of the administration of
benefit programs – especially since information monitoring systems would be needed to monitor
the crisis and process an increase in applications during a crisis. This indicator is admittedly a
somewhat subjective evaluation. It takes into account the degree to which information is readily
available on numbers of applicants, beneficiaries, and spending on SB programs, which can be an
indicator of efficacy of program management (benefits administration). This can also signal the need
for structural improvements in the medium term. In addition, if the authorities do not have the
information it would be harder to monitor and respond/intervene as needed. After all, the ability to
respond is very dependent on having information.
Composite Score
We develop the composite score as a simple average of all these different dimensions (9
dimensions in total). This approach indicates that, when there are missing values in any given
dimension, these are assumed to be as good or as bad as the dimensions for which there is
information. While this is not a perfect method and there are many different ways of obtaining a
composite score based on these dimensions, the resulting categories do not appear to change. In
addition, for the most part, the resulting categories seem in‐line with the overall information about
the countries’ Social benefits’ systems.
05101520253035404550
Means‐tested Non‐means‐tested
Main Family and Child BenefitsBenefits as Percent of Post‐Transfer Consumption,
Beneficiary Household, Poorest Quintile
50
Annex 2‐ Individual country briefs
Armenia
Summary of the data:
Registered unemployed Definition Number of persons looking for work who are registered in the regional and local
employment offices Unit Persons Comment Source State Employment Service Agency Unemployment benefit Definition Monthly number of persons receiving unemployment benefit Unit Persons Comment The unemployment benefit is received by persons who have been insured for a least a
year and are registered with the State Employment Service Agency. The amount of the benefit is set at 60% of the minimum monthly salary, as set in Armenian legislation. The duration of the unemployment benefit payment is determined by the number of covered years, with each three years of insured coverage adding an extra month of unemployment benefit. However, the minimum duration is 6 months and the maximum duration is 12 months. After expiry of unemployment benefits, a person can again obtain unemployment benefit after another year of covered insurance and
51
registration as unemployed.
Source State Employment Service Agency Social monthly assistance (Family Benefit Program, FBP) Definition Monthly number of individuals or families receiving social monthly assistance. Unit Persons Comment Family Benefit Program is a household‐level last‐resort targeted social assistance
program. It is hybrid means tested (HMT), with eligibility determined by a score‐based formula, which includes a set of coefficients, some of which are scores (e.g. vulnerable group categories, income, housing conditions, geographic area), while others are filters (e.g. car ownership, real estate acquisition, excessive electricity / long‐distance phone consumption). The amount of the benefit is a minimum basic benefit, with supplements for each child (which vary based on the targeting score), and top‐ups for families with four or more children and those residing in mountainous regions or near borders. There is no limit to the duration of the FBP, but the family has to re‐certify for eligibility annually.
Source Ministry of Labor and Social Issues Child benefit Definition Unit Comment This is a universal benefit provided as child care allowance for children under 2 years
of age. The amount of the monthly benefit is 18,000 AMD, and it is paid until the child turns 2. There is also a childbirth benefit, which is a lump sum of 50,000 AMD, paid for the first child, 50,000 AMD for the second child, and 430,000 AMD for each subsequent child. The sum is paid during the 6 months after childbirth or adoption.
Source Social Disability benefit Definition Unit Comment Disability classification (Group I: total incapacity for work, requiring constant
attendance; Group II: total incapacity for work, not requiring constant attendance; and Group III: partial incapacity for work) is assessed by a specialized medical committee. The amount depends on the basic pension and years of covered employment. There is no maximum for disability pension.
Source
52
Bulgaria
Registered unemployed Definition Number of persons looking for work who are entered on the registers of public
employment services at the end of each month. Unit Number of persons Comment Source Public Employment Services Unemployment benefit Definition Monthly number of persons receiving unemployment benefit. Unit Number of recipients Comment The unemployment benefit is granted to those who have been insured for at least 9
months during the last 15 months and who are registered with the PES. The benefit is not means‐tested. Daily unemployment benefits amount to 60% of the average contributory income for the last 9 months and cannot be less than the minimum which is defined annually by the law (currently BGN 6/€ 3 a day). Duration of the benefit depends on the length of contribution period and is not to exceed 12 months.
Source Public Employment Services Social monthly assistance (GMI) Definition Monthly number of individuals or families receiving social monthly assistance. Unit Number of recipients
53
Comment Monthly social assistance is provided to persons or families whose income is lower than a minimum income threshold and meets specific conditions. The benefits are means‐tested and financed by the state budget. The amount of the benefit is equal to the difference between the differentiated minimum income and the income of the persons or the families from the preceding month. The guaranteed minimum income is currently BGN 65/€ 33 per month. Until 01/01/2011, the benefit was subject to a duration limit of continuous 12 months (renewable), which has now been abolished.
Source Social Assistance Agency Child benefit Definition Monthly number of children receiving monthly child assistance. Unit Number of recipients Comment Monthly child assistance is a flat rate benefit granted to families with children until
their completion of secondary education, but not after the age of 20. Currently, the benefit is BGN 35 (€ 18) per month for each child. The benefit is means‐tested and financed by the state budget. The average monthly gross income per family member must be lower or equal to BGN 350 (€ 179).
Source Social Assistance Agency Disability benefit Definition Monthly number of persons receiving monthly disability benefit Unit Number of recipients Comment People with permanent disabilities are entitled to a monthly allowance for social
integration according to their individual needs according to the degree of reduced capacity or the type and degree of disability. Cash add‐ons to supplement income are intended to cover additional costs for transport services; information and telecommunication services; training; medical treatment and rehabilitation services; accessible information; municipal rental housing, diet and medicines. The benefit is not means‐tested and is financed by the state budget.
Source Social Assistance Agency
54
Croatia
Registered unemployed Definition Number of persons aged from 15 to 65, fully or partly capable of working, without a
job, who are active job seekers and available for work, who are entered on the registers of the Croatian Employment Service at the end of each month.
Unit Number of persons Comment Source Croatian Employment Service Unemployment benefit Definition Monthly number of persons receiving unemployment benefit. Unit Number of recipients Comment The unemployment benefit is granted to those who have been insured for at least 9
months during the last 24 months and who are registered with the CES. The benefit is not means‐tested. Unemployment benefits amount to 70% of the average contributory income for the last 3 months during the first 90 days and 35% of the average contributory income for the remaining period. However, there is a cap of 70% of the average monthly wage in Croatia (HRK 3,740 or EUR 505) for the first 90 days and 35% for the remaining period (HRK 1,870 or EUR 252), as well as the minimum which
55
cannot be less than 50% of the minimum monthly wage (currently HRK 1,407 or EUR 190). Duration of the benefit depends on the length of contribution period and ranges from 90 days if the person’s contribution period is below 2 years to 450 days if the person’s contribution period exceeds 25 years.
Source Croatian Employment Service Social monthly assistance (GMI) Definition Monthly number of individuals receiving monthly social support allowance Unit Number of recipients Comment Monthly social support allowance is provided to persons or families whose income is
lower than a minimum income threshold and meets specific conditions. The benefits are means‐tested and financed by the state budget. The amount of the allowance is set at HRK 500 (EUR 67). It is adjusted according to family size and set against the actual income of the household. The allowance is additionally increased for up to xx percent for some specific categories such as a person incapable of work living alone or in the family, children of single parents, etc.
Source Ministry of Health and Social Welfare Child Allowance Definition Monthly number of beneficiaries of child allowance (Or Monthly number of children
receiving monthly child allowance) Unit Number of recipient families Comment Monthly child allowance is an income‐tested benefit financed by the stated budget,
granted to families with income per family member equal or below HRK 1,663 (EUR 225) per month. Some categories of beneficiaries are not subject to income testing such as children of war veterans or children with serious health impairment. The benefit is received until the child’s completion of secondary education, but not after the age of 21 (with the exception of children with severe health impairment who receive the allowance until the age of 27). Currently, the benefit per month for each child ranges from HRK 200 (EUR 27) to HRK 300 (EUR 40), depending on the level of monthly income per family member. The allowance per child may be additionally increased by up to 25% for specific categories of children such as children without one or both parents, children with health impairment)
Source Croatian Institute for Pension Insurance Disability Allowance Definition Monthly number of persons receiving monthly disability benefit Unit Number of recipients Comment People with severe disability or serious mental illness are entitled to a monthly
allowance for the purpose of fulfilling their needs and including them in the everyday life in the community on equal terms with others. The benefit is not means‐tested and is financed by the state budget. The amount of the benefit is HRK 1,250 (EUR 169)
Source Ministry of Health and Social Welfare
56
Georgia
Targeted Social assistance (TSA) Definition Number of approved beneficiaries Unit Number of Households Comment The cash allowance is awarded to families which rating score is less than 57 001. The
score for HH is estimated by means test. The amount of assistance is 6+24n where n is family size.
Source Social Service Agency Family Allowance Definition Number of approved beneficiaries Unit Number of Households Comment In this program no new beneficiaries are approved. The permanent decrease is caused
by this fact. "Family Allowance" is provided to families who have switched to the program before January 1st, 2007. Beneficiaries of different categories receiving the "Family Allowance" and amount of social allowance: Single, unemployed pensioner (single pensioner ‐ age pensioner or disabled) For single unemployed pensioner ‐ 22 GEL;
57
For family of two unemployed pensioners ‐ 35 GEL; Orphan child, besides guardian's disability status ‐ 22 GEL per person; Unemployed disabled, the first group disabled (blind) ‐ 22 GEL per person; Underage disabled ‐ 22 GEL per person; Large family with 7 or more underage children ‐ 35 GEL.
Source Social Service Agency Disability Benefit Definition Number of approved beneficiaries Unit Number of persons Comment Georgian citizen who has the I or the II group disability (except of the person who has
a disability status from the childhood or the veteran of war who has the III group of disability) has to apply to the regional branch of the Social Service Agency to fix the pension. Amount of the pension: For the I group disabled people ‐ 80 GEL, For the II and the III group disabled people ‐ 70 GEL.
Source Social Service Agency
58
Hungary
Unemployed ILO Definition Unemployment according to ILO definition Unit Persons – seasonally adjusted Comment See: http://nui.epp.eurostat.ec.europa.eu/nui/show.do?dataset=lmhu_m&lang=en Source Eurostat Jobseekers Allowance and Assistance (Unemployment Benefit) Definition Unemployment benefit = Jobseekers’ allowance recipients and recipients of jobseekers’
assistance Unit Recipients Comment Source Employment and Social Office (www.afsz.hu) Regular Social Assistance, Availability Support, or Public Work (Social Assistance
Benefit) Definition Recipients of either (1) regular social assistance; (2) availability support; or (3) public
work Unit Recipients Comment Regular social assistance is an income supplement provision in the form of cash,
provided by the local government of the settlement. Its aim to guarantee a minimal
59
standard of living for those who have no income. From the 1 July 2006 the conditions of the provision and the way of calculation of the amount of support changed. Before that the local government awarded regular social assistance to a person who was over 18 years of age, was of active age, and had lost at least 67 per cent of his or her working ability or received blind persons’ benefit, or to a person who was of active age but not in employment, in the case that their subsistence was not provided by other means. By the new terms for the support is entitled only one person in a family. The assessing of the entitlement and the amount of the assistance based on the income projected to the consumer unit instead of the previous income per capita. The consumer unit is the rate which shows the structure of consumption within a family. The first major member of the family and the disabled child’s rate is 1.0 while the ratio of the companion (spouse) and a child is lower (0.9‐0.7). The amount of support is variable and supplements the family’s effective total income to the limit of the entitlement. The regular social assistance from 1 January 2009 was changed to benefit for active aged which consist of the regular social assistance and the „support for to be ready to work”. The change in the benefit system was built up completely until 31 March 2009.
Source Ministry of Local Government Disability Subsidy Definition Disability subsidy recipients Unit Recipients Comment Financial support for severely disabled persons over the age of 18, who are unable to
care for themselves or need permanent assistance from others. The regulation of regular social assistance is altered from 1 January 2009. Due to change of regulation the number of recipients was decreased.
Source Hungarian Treasury (www.allamkincstar.gov.hu)
60
Latvia
Registered unemployed Definition Number of persons looking for work who are entered on the registers of public
employment services at the end of each month Unit Number of persons Comment Source Public Employment Services Unemployment benefit Definition Monthly number of persons receiving unemployment benefit Unit Number of recipients Comment The unemployment benefit is granted to those who have been socially insured for at
least 1 year and who have paid at least 9 months of contributions in 12 months before registering as unemployed. The benefit is earnings‐related and not means‐tested. The amount of unemployment benefit is determined in proportion to the insurance period and the income on the basis of which unemployment contributions are paid and decreases over time. The benefit is paid up to 9 months and during the last three months the benefit is flat (LVL 45 (€ 63) per month) for those, who contributed for less than 20 years. Temporarily, between 1 January 2010 and 31 December 2012, the part of the benefit exceeding LVL 11.51 (€ 16) per day is paid at half rate.
61
Source Public Employment Services Guaranteed minimum income benefit (GMI) Definition Persons in household receiving municipal GMI benefit Unit Number of recipients Comment The Guaranteed minimum income (GMI) benefit is provided to persons or families
whose income level is lower than the level of income set by the Cabinet of Ministers. The GMI level as well as eligibility conditions, formalities, calculation and payment procedures for the GMI benefit are set by central government, but until 2009 the benefit was paid out by local municipalities from the municipal budget. The benefit is now co‐financed by the central government. The benefit is calculated as the difference between the GMI level set by the Cabinet of Ministers (LVL 40 (€ 56) for adults and LVL 45 (€ 63) for children) and the person's or the household’s average monthly income over the last three months. The municipality can establish a higher GMI level (but not higher than the needy level (or 50% of the minimum wage) for persons receiving old‐age and disability pensions. The benefit is granted for a period of 3 months or 6 months and is renewable. The benefit is income‐ and assets (property) tested and can be granted in cash or in kind.
Source Ministry of Welfare Housing Benefits Definition Persons in household receiving municipal housing benefit Unit Number of recipients Comment The housing benefit is a separate municipal benefit. The amount of this benefit varies
from one municipality to another depending on resources available. If the person is granted the status of a needy person and she/he has expressed a wish to be a tenant of a social flat (housing), the person can rent a flat as social housing where reduced rent and utility payments are charged. Since 2009, the benefit is co‐financed by the central government.
Source Ministry of Welfare
62
Lithuania
Unemployed Definition Unemployment according to ILO definition – Total Unit Persons – seasonally adjusted Comment See: http://nui.epp.eurostat.ec.europa.eu/nui/show.do?dataset=lmhu_m&lang=en Source Eurostat Unemployment Benefit Definition Unemployment benefit recipients Unit Recipients Comment The unemployed have possibility to receive unemployment benefit if he has a minimum
period of insurance: 18 months within 3 years preceding unemployment (there are exceptions for certain groups of unemployed people who contributed but have not acquired the necessary social insurance record due to important reasons). The duration of payment of Unemployment Insurance Benefit depends on the length of the insurance record: Service years Duration less than 25 years 6 months; 25 ‐ 30 years ‐7 months, 30 ‐ 35 years‐ 8 months; 35 years and over‐9 months. The duration of payment is prolonged for additional 2 months for elderly persons within 5 years till pension age.
Source National Labour Exchange Social Benefit Definition Number of recipients of social benefit Unit Recipients
63
Comment Families and single residents are entitled to Social Benefit if either single resident or one spouse works or does not work because they are full‐time students or pensioners, or individuals above retirement age, or disabled, or nursing a disabled or sick family member, or registered at the local office as unemployed and receiving Unemployment Benefit or are long‐term unemployed (more than 6 months), or taking care of a child under the age of 3 years or under the age of 8 years, or family is raising three or more children, etc.
Source Ministry of Social Security and Labour, The Social Assistance Information System Unemployed Definition Unemployment according to ILO definition – Total Unit Persons – seasonally adjusted Comment See: http://nui.epp.eurostat.ec.europa.eu/nui/show.do?dataset=lmhu_m&lang=en Source Eurostat Unemployment Benefit Definition Unemployment benefit recipients Unit Recipients Comment The unemployed have possibility to receive unemployment benefit if he has a minimum
period of insurance: 18 months within 3 years preceding unemployment (there are exceptions for certain groups of unemployed people who contributed but have not acquired the necessary social insurance record due to important reasons). The duration of payment of Unemployment Insurance Benefit depends on the length of the insurance record: Service years Duration less than 25 years 6 months; 25 ‐ 30 years ‐7 months, 30 ‐ 35 years‐ 8 months; 35 years and over‐9 months. The duration of payment is prolonged for additional 2 months for elderly persons within 5 years till pension age.
Source National Labour Exchange Social Benefit Definition Number of recipients of social benefit Unit Recipients Comment Families and single residents are entitled to Social Benefit if either single resident or one
spouse works or does not work because they are full‐time students or pensioners, or individuals above retirement age, or disabled, or nursing a disabled or sick family member, or registered at the local office as unemployed and receiving Unemployment Benefit or are long‐term unemployed (more than 6 months), or taking care of a child under the age of 3 years or under the age of 8 years, or family is raising three or more children, etc.
Source Ministry of Social Security and Labour, The Social Assistance Information System
64
Montenegro
Registered unemployed Definition Unit Comment Source Unemployment benefit Definition The unemployment compensation is a contributory benefit due to registered
unemployed who meet a set of criteria as minimum duration of contractual work and contribution to unemployment insurance record, terms of employment termination and deadline for submission of the claim.
Unit Registered unemployed, individuals Comment The unemployment compensation is paid by the Pension and Disability Insurance
Fund and delivered through the Employment Office upon meeting certain eligibility criteria: contribution for unemployment insurance for at least 9 months continuously of for at least 12 months with interruptions in the last 18 months; submission of the claim within 42 days after the employment was terminated, employment was not terminated voluntarily or the loss of job was due to the employee’s fault. The duration of benefit receipt is between 3 months (minimum) and 12 months (maximum) and depends on the length of insurance record. The unemployment benefit amount equals
65
to 65% of the minimum pay standard in Montenegro for the last two years. In addition to the “regular” unemployment benefit, since 2009, through changes in the Law on Employment, a new unemployment benefit has been introduced that amounts to the minimum pension. Eligible are elderly unemployed (male over 60 years of age and women over 55 years of age) with minimum 10 years of contribution record; and long‐term contributors (30 years of contribution record and age of 50 or above). The duration of this benefit is up to 24 months. The unemployment insurance contribution amounts to 1% of gross wage out of which 0.5% is paid by the employee and 0.5% ‐ by the employer.
Source Law on Employment, Ministry of Labor and Social Welfare Monthly financial social assistance: Family Material Support(FMS/MOP) Definition Monthly last‐resort social assistance for families in material need, defined with an
income and asset test Unit Individuals (single‐member families) and multi‐member families Comment Monthly benefit for family, ranging from €60.5 to €115, depending on the number of
recipient family members. The family income should be lower than a pre‐set standard that varies with family size, and is equal to the maximum benefit. Certain assets (land and dwelling ownership, productive assets, vehicle) are considered along with income. The actual benefit that is received by a family of certain size is defined as the difference between the pre‐set income standard and the actual income of the family. There is no time limit for receiving the benefit. Re‐certification is due once in 12 months or more often if the material circumstances change. The benefit is designed by the Ministry of Labor and Social Welfare and delivered by the Centers for Social Work. The benefit is conditional to regular school attendance of children from recipient families.
Source Law on Social and Child Protection, Ministry of Labor and Social Welfare Child Allowance Definition Means‐tested monthly benefit for families with children who are FMS/MOP
beneficiaries Unit Individuals Comment The child allowance is a benefit for children whose family is FMS/MOP beneficiary, for
which are eligible the first three children in the family. The child allowance is extended without a means test to families with children with physical, mental and sensor disorders and children without parents. The amount of the child allowance is 18.2 € for children in FMS/MOP beneficiary families; 24.2€ for children with physical, mental and sensor disorder which can be enabled for normal life and work; 30.2€ for children with physical, mental and sensor disorder which cannot be enabled for normal life and work; and 30.2 € for children without parents. The benefit is extended to biological, adopted and fostered children up to 18 years of age.
Source Law on Social and Child Protection, Ministry of Labor and Social Welfare Personal Disability Benefit and Caregivers’ Allowance Definition Non‐contributory categorical disability benefit for those whose disability has occurred
before becoming 18 years old Unit Individuals holding medical certification of disability Comment Eligible for personal disability benefit are persons whose inability for independent life
occurred before they became 18 years old. The amount of this benefit is €75 per
66
month since 2009, €50 before that. The benefit is not means tested. No recertification is envisaged. The caregivers’ allowance is extended only to those with severe disability (irrespective of whether it has been acquired from birth or later in life) who do not have disability insurance, and also meet the income and asset test for FMS/MOP. The amount of the caregivers’ allowance is €60 since 2008, up from €55 before that.
Source Social Protection Law, Ministry of Labor and Social Welfare
67
Romania
Registered unemployed Definition Number of persons looking for work who are entered on the registers of public
employment services at the end of each month. Unit Number of persons Comment Source Public Employment Services Unemployment benefit Definition Number of unemployment indemnity recipients Unit Number of persons Comment The unemployment benefit is granted to those who have been insured for 12 months
during the 24 months preceding registration. The benefit is earnings‐related and not means‐tested. The amount of unemployment benefit depends on the minimum gross wage, level of earnings, and length of contribution period. The amount of benefit was reduced by 15% during the period between 3 July 2010 and 31 December 2010. Duration of payment varies according to contribution period from 6 month (graduates and those who contributed for 1‐5 years) to 12 months (those who contributed for at least 10 years).
Source Ministry of Labour, Family and Social Protection Social Aid (GMI)
68
Definition Number of Social aid recipients Unit Number of persons Comment Minimum guarantee income is provided in the form of an income tested monthly
benefit (Social Aid) by the local authorities. The amount of Social Aid is a difference between Guaranteed Minimum Income, which is set by the Government at the national level, and individual’s or family’s net income. Minimum Social Aid amount is RON 10 (€ 2.30) per month. The work‐able recipients of this benefit are required by law to provide, on a monthly basis, a maximum of 72 hours of community work, at the request of the mayor. The Social Aid is paid until the beneficiary ceases to meet the conditions.
Source Ministry of Labour, Family and Social Protection Family Allowances (CFA and SFA) Definition Total number of complementary family allowance and single‐parent family allowance
beneficiaries Unit Number of persons Comment Complementary family allowance (CFA) is an income‐tested benefit granted to families
with children whose net income per family member is less than RON 470 (€108). Eligible child must live with family and be of school age attending a form of education until the age of 18 years. The monthly benefit amount differs according to the number of children within the family: RON 50 (€12) for a family with 1 child to RON 70 (€16) for a family with 4 or more children. For a family entitled to Social Aid (GMI benefit), CFA is increased by 25%. Single‐parent family allowance is an income‐tested benefit to children in single‐parent families. The monthly benefit amount differs according to the number of children: RON 70 (€16) for a single‐parent family with 1 child to RON 90 (€21) for a single‐parent family with 4 or more children.
Source Ministry of Labour, Family and Social Protection
69
Serbia
Registered unemployed Definition Unit Comment Source Unemployment benefit Definition Contribution‐based benefit paid by the Public Employment Service to registered
unemployed who meet certain conditions. Unit Individuals, registered unemployed Comment Contribution‐based benefit paid by the Public Employment Service to registered
unemployed who meet certain conditions: length of work and contribution record of 12 consecutive or 18 non‐consecutive months, terms of separation, commitment to active job search. The unemployment benefit amounts to 50% of the last reference earning; cannot be lower that 80% of the minimum wage for the month before the benefit is awarded and higher than 160% of the minimum wage for the same reference month. The duration of the benefit depends on the length of contribution record. The benefit is not taxable. While receiving the unemployment benefit, the recipient is not allowed to engage in gainful employment. Unemployment benefit can be paid on a one‐time basis upon the request of unemployed person. Unemployed person who receives unemployment benefit for at least 3 months since the recognition
70
of his rights, and has became employed for indefinite period of time, is entitled to one‐off incentive for employment. Incentive equals 30% of total unemployment benefit amount (net of taxes and contributions), that would have been paid to him/her until expiry of his/her right to unemployment benefit
Source Law on Employment and Insurance Against Unemployment, 2009. Ministry of Labor and Social Policy
Social monthly assistance: Cash Social Assistance (‘New MOP’) Definition Guaranteed minimum income program, non‐contributory cash benefit granted to
families whose income is below a pre‐set standards defined by the Ministry of Labor and Social Policy, and who also meet certain asset criteria
Unit Number of recipient families or households, number of individuals (monthly data) Comment Monthly social assistance is provided to persons or families whose income is lower
than a minimum income threshold (defined as base income threshold and scaled up with equivalence scales to reflect the size and composition of the unit of assistance), passes an asset test and meets other conditions. The benefits are means‐tested and financed by the state budget. The amount of the actual benefit paid is equal to the difference between the differentiated income threshold and the actual income of the person or the families from the preceding month. The cash social assistance basic amount is currently 60 Euro. The basic amount is increased by 50% for each subsequent adult and by 30% for each child in the unit of assistance. There is a limit of six recipients in one unit of assistance. There is no time limit for the benefit receipt. Eligibility is recertified each 12 months. Able‐bodied working‐age beneficiaries receive the benefit for 9 months in one calendar year; those who are considered not able to work receive it for 12 months in one calendar year. The cahs social assistance is delivered by the Centers for Social Work. The basic benefit amount is updated with the cost of living index twice a year.
Source Law on Social Assistance, 2011; Ministry of Labor and Social Policy Child Allowance Definition The monthly child allowance is a non‐contributory benefit provided to families with
children whose income per capita is below a threshold defined by the law in 2002 and updated twice a year with the respective cost of living index.
Unit Number of recipient children Comment The monthly child allowance is income tested. The income threshold is similar in size to
the one for cash social assistance but they are defined independently. The child benefit is administered and delivered by the municipal social departments. The basic benefit is defined by the Law in 2002 at Euro 27.5 and applies for the first child. Up to 4 children in a family receive the benefit. The second, third and fourth child receive a benefit which is higher than the base by Euro 9. The benefit amount is indexed with the cost of living index twice a year.
Source Law on Financial Support for Families with Children, Ministry of Labor and Social Policy Disability (caregiver’s) allowance Definition Benefit paid to persons with high level of disability who need the help / care of a third
person in performing everyday vital activities. Unit Individual with medical certification of disability Comment The benefit could be contributory (paid by the Pension Fund) or non‐contributory (paid
by the Ministry of Labor and Social Policy. The basic amount is stipulated in the Law in
71
2004 at Euro 40, and is subject to adjustment each month by the cost of living index. Those with 100% disability receive increased by 70%) caregivers’ allowance. The allowance is awarded and discontinued based on medical assessment. Granted for indefinite time in case of permanent disability. The allowance is not taxable.
Source Law on Social Protection and Provision of Social Safety to Citizens, Ministry of Labor and Social Policy
72
Ukraine
Registered unemployed Definition Number of persons without employment who are looking for work and are entered on
the registers of public employment services at the end of each month. Unit Number of persons Comment Source Public Employment Service Unemployment benefit Definition Monthly number of persons receiving unemployment benefit. Unit Number of recipients Comment The unemployment benefit is granted to those who are registered with the PES and
who have been employed for at least 26 weeks over the 12 months period preceding the unemployment and paid insurance contributions. Additional requirement is that a person should be ready to participate in public works, if such work appears suitable. The benefit starts eight days after registration. The benefit is not means‐tested and its size depends on the claimant`s insurance history. In some instances, however, benefit may be granted regardless of insurance period. Duration of the benefit depends on the length of contribution period and cannot exceed 360 days cumulatively over two years period. Monthly amount of unemployment benefit ranges from 50% to 70% of the average contributory income for the last 6 months. It cannot be less than the
73
subsistence minimum set annually by the law. It cannot exceed an average wage reported in a claimant`s residence region for the previous month. The benefit can also be paid in a lump sum.
Source Public Employment Service Social Assistance to Poor Families (GMI) Definition Monthly number of individuals or families receiving assistance. Unit Number of recipients Comment Monthly social assistance is provided to persons or families whose income is lower
than a minimum income threshold (subsistence minimum) and meets specific conditions. The benefits are means‐tested, they are financed by the state budget and administered by local welfare offices. The amount of the benefit is equal to the difference between the differentiated minimum income and the average monthly income of a family for 6 months preceding application. The benefit is granted for 6 months period and is renewable. Over this period the benefits amount is revised with changes in income threshold and/or if the family claims change in income status.
Source Ministry of Social Policy Child Care benefit Definition Monthly number of individuals receiving child care benefit for children under three
years Unit Number of recipients Comment Monthly child care benefit is provided to a person who takes immediate care of a
child/children below three years old. The amount of the benefit per each child is equal to the difference between the differentiated minimum income (subsistence minimum) and the average monthly income of a family for 6 months preceding application. The benefits are financed by the state budget and administered by local welfare offices. The benefits are income tested, however, there is a minimum flat benefits that is granted per child regardless of income. The benefit is granted for the period until a child reaches the age of three years old but recertification is to take place every six months. If recertification does not take place or, as a result of recertification, the amount of the benefit appears lower than a minimum flat benefit, a person will be paid a minimum flat benefit.
Source Ministry of Social Policy Single Parent benefit Definition Monthly number of individuals receiving social benefits to single parents Unit Number of recipients Comment Monthly single parent benefit is provided to a parent who is single (unmarried,
divorced, widowed) and takes care of a child/children below 18 years old. The amount of the benefit per each child is equal to the difference between the 50% of differentiated minimum income (subsistence minimum) and the average monthly income of a family for 6 months preceding application. The benefits are financed by the state budget and administered by local welfare offices. The benefits are income tested but cannot be lower than 30% of subsistence minimum which is granted regardless of income. There is also a maximum limit to the benefit size which is differentiated for children below and above six years old. The benefit is granted for the period until a child reaches the age of 18 years old (in some instances 23 years old) and is automatically recalculated at 6 years. For income‐tested benefit, the
74
recertification is to take place every six months. If recertification does not take place or, as a result of recertification, the benefit amount appears lower than a minimum flat benefit, a person will be paid a minimum flat benefit.
Source Ministry of Social Policy
Social Protection & Labor Discussion Paper Series Titles 2011‐2012
No. Title 1219 Do Social Benefits Respond to Crises? Evidence from Europe & Central Asia
During the Global Crisis by Aylin Isik‐Dikmelik, November 2012 1218 Building Results Frameworks for Safety Nets Projects by Gloria M. Rubio, October 2012 1217 Pension Coverage in Latin America: Trends and Determinants by Rafael Rofman and Maria Laura Oliveri, June 2012 1216 Cash for Works in Sierra Leone: A Case Study on the Design and
Implementation of a Safety Net in Response to a Crisis by Colin Andrews, Mirey Ovadiya, Christophe Ribes Ros and Quentin Wodon,
June 2012 1215 Public Employment Services, and Activation Policies by Arvo Kuddo, May 2012 1214 Private Pension Systems: Cross‐Country Investment Performance by Alberto R. Musalem and Ricardo Pasquini, May 2012 1213 Global Pension Systems and Their Reform: Worldwide Drivers, Trends, and
Challenges by Robert Holzmann, May 2012 1212 Towards Smarter Worker Protection Systems: Improving Labor Regulations
and Social Insurance Systems while Creating (Good) Jobs by David A. Robalino, Michael Weber, Arvo Kuddo, Friederike Rother, Aleksandra Posarac and Kwabena Otoo
1211 International Patterns of Pension Provision II: A Worldwide Overview of Facts
and Figures by Montserrat Pallares‐Miralles, Carolina Romero and Edward Whitehouse,
June 2012 1210 Climate‐Responsive Social Protection by Anne T. Kuriakose, Rasmus Heltberg, William Wiseman, Cecilia Costella,
Rachel Cipryk and Sabine Cornelius, March 2012 1209 Social Protection in Low Income Countries and Fragile Situations: Challenges
and Future Directions by Colin Andrews, Maitreyi Das, John Elder, Mirey Ovadiya and Giuseppe
Zampaglione, March 2012
1208 World Bank Support for Pensions and Social Security by Mark Dorfman and Robert Palacios, March 2012 1207 Labor Markets in Middle and Low Income Countries: Trends and Implications for
Social Protection and Labor Policies by Yoonyoung Cho, David Margolis, David Newhouse and David Robalino,
March 2012 1206 Rules, Roles and Controls: Governance in Social Protection with an Application
to Social Assistance by Lucy Bassett, Sara Giannozzi, Lucian Pop and Dena Ringold, March 2012 1205 Crisis Response in Social Protection by Federica Marzo and Hideki Mori, March 2012 1204 Improving Access to Jobs and Earnings Opportunities: The Role of Activation
and Graduation Policies in Developing Countries by Rita Almeida, Juliana Arbelaez, Maddalena Honorati, Arvo Kuddo, Tanja
Lohmann, Mirey Ovadiya, Lucian Pop, Maria Laura Sanchez Puerta and Michael Weber, March 2012
1203 Productive Role of Safety Nets by Harold Alderman and Ruslan Yemtsov, March 2012 1202 Building Social Protection and Labor Systems: Concepts and Operational
Implications by David A. Robalino, Laura Rawlings and Ian Walker, March 2012 1201 MicroDeterminants of Informal Employment in the Middle East and North Africa Region by Diego F. Angel‐Urdinola and Kimie Tanabe, January 2012 1120 Employment Generation in Rural Africa: Mid‐term Results from an Experimental Evaluation of the Youth Opportunities Program in Northern Uganda by Christopher Blattman, Nathan Fiala and Sebastian Martinez, December 2011 1119 Measuring Governance and Service Delivery in Safety Net Programs by Gloria M. Rubio, September 2011 1118 Assessing Safety Net Readiness in Response to Food Price Volatility by Margaret Grosh, Colin Andrews, Rodrigo Quintana, Claudia Rodriguez‐Alas, September 2011
1117 Social Safety Nets in Fragile States: A Community‐Based School Feeding Program in Togo, August 2011 (also available in French) by Colin Andrews, Elena Galliano, Carolyn Turk and Giuseppe Zampaglione,
August 2011 1116 Strengthening Governance of Social Safety Nets in East Asia by Sara Giannozzi and Asmeen Khan, August 2011 1115 International Portability of Health‐Cost Coverage: Concepts and Experience by Martin Werding and Stuart McLennan, July 2011 1114 Liberia’s Cash For Work Temporary Employment Project: Responding to Crisis in Low Income, Fragile Countries
by Colin Andrews, Prospère Backiny‐Yetna, Emily Garin, Emily Weedon, Quentin Wodon and Giuseppe Zampaglione, July 2011 1113 Employability and Productivity among Older Workers: A Policy Framework and Evidence from Latin America by Edmundo Murrugarra, July 2011 1112 Cash Transfers, Children and the Crisis: Protecting Current and Future Investments by Ariel Fiszbein, Dena Ringold, Santhosh Srinivasan, June 2011 1111 Severance Pay Programs around the World: History, Rationale, Status, and Reforms by Robert Holzmann, Yann Pouget, Milan Vodopivec and Michael Weber, May 2011 1110 Portability of Pension, Health, and other Social Benefits: Facts, Concepts, Issues by Robert Holzmann and Johannes Koettl, May 2011 1109 Disability and Poverty in Developing Countries: A Snapshot from the World Health Survey by Sophie Mitra, Aleksandra Posarac and Brandon Vick, April 2011 1108 Advancing Adult Learning in Eastern Europe and Central Asia by Christian Bodewig and Sarojini Hirshleifer, April 2011 1107 Results Readiness in Social Protection & Labor Operations by Laura Rawlings, Maddalena Honorati, Gloria Rubio and Julie Van Domelen, February 2011 1106 Results Readiness in Social Protection & Labor Operations: Technical Guidance Notes for Social Service Delivery Projects by Julie Van Domelen, February 2011
1105 Results Readiness in Social Protection & Labor Operations: Technical Guidance Notes for Social Safety Nets Task Teams by Gloria Rubio, February 2011 1104 Results Readiness in Social Protection & Labor Operations: Technical Guidance Notes for Social Funds Task Teams by Julie Van Domelen, February 2011 1103 Results Readiness in Social Protection & Labor Operations: Technical Guidance Notes for Labor Markets Task Teams by Maddalena Honorati, February 2011 1102 Natural Disasters: What is the Role for Social Safety Nets? by Larissa Pelham, Edward Clay and Tim Braunholz, February 2011 1101 North‐South Knowledge Sharing on Incentive‐based Conditional Cash Transfer Programs by Lawrence Aber and Laura B. Rawlings, January 2011
To view Social Protection Discussion papers published prior to 2011, please visit www.worldbank.org/sp
Printed on recycled paper
About this series...
Social Protection & Labor Discussion Papers are published to communicate the results of The World Bank’s work to the development community with the least possible delay. The typescript manuscript of this paper therefore has not been prepared in accordance with the procedures appropriate to formally edited texts.
The findings, interpretations, and conclusions expressed herein are those of the author(s), and do not necessarily reflect the views of the International Bank for Reconstruction and Development / The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The author(s) attest(s) that the paper represents original work. It fully references and describes all relevant prior work on the same subject.
For more information, please contact the Social Protection Advisory Service, The World Bank, 1818 H Street, N.W., Room G7-803, Washington, DC 20433 USA. Telephone: (202) 458-5267, Fax: (202) 614-0471, E-mail: [email protected] or visit us on-line at www.worldbank.org/sp.
N o v e m b e r 2 0 1 2
Abstract
Social benefits can potentially play an important role in protecting the poor and minimizing the impacts of an economic crisis. While many studies estimate the impacts of a crisis, there is little evidence of the actual response of social safety nets to systematic shocks. This study traces the response of social benefits during the 2008–10 global crisis for 14 countries in ECA. The study first sets out a framework for defining the “expected” response of social benefits covering an assessment of pre-crisis preparedness of social benefits and the severity of the crisis for all countries in the ECA region to provide the context; and then develops a typology of all countries categorized by expected response. Using this typology the study analyzes the monthly administrative data on the observed patterns within social benefit programs. Main findings indicate that actual responses were largely in line with expectations. Pre-crisis preparedness clearly influenced the ability of social benefits to respond to the crisis. Unemployment benefits were generally the first line of response in countries that have them, while social assistance programs also expanded coverage during the crisis. Lessons learned from the 2008-2010 global crisis (such as the importance of structural reform, design, and implementation which affect the success of social benefits programs in crisis response) are also presented. The study concludes with some policy recommendations to help ECA countries prepare for future crises.
Do Social Benefits Respond to Crises?
Evidence from Europe & Central Asia During the Global Crisis
Aylin Isik-Dikmelik
D I S C U S S I O N P A P E R NO. 1219