Report No. 32433-SK

Slovak Republic
The Quest for Equitable Growth
In The Slovak Republic

A World Bank Living Standards Assessment

September 22, 2005

Poverty Reduction and Economic Management Unit
Europe and Central Asia Region




Document of the World Bank


CONTENTS

ACKNOWLEDGMENTS               ......................................................................................               vii
EXECUTIVE SUMMARY              ......................................................................................                ix
 EmploymentOpportunitiesWere Slow to Reactto Growth                                 ...............................................   x
 LaborMarketSegmentationHas BeenHistoricallyHigh.................................................                                    xi
 What I s the Government's Strategy for TacklingSocialExclusion?...............................                                    xiii
                                                                                                                                     ...
 Has the Strategy WorkedTo Date?..................................................................................                 xiv
 Who Are the Winners andWhoAre the Losers?.............................................................                            xiv
 Canthe Effectiveness ofthe StrategyBe Increased?                        ......................................................    xvi
 Canthe Living Standards ofthe MostVulnerable ReceiveMore Support?...................xvii

CHAPTER 1 SETTING THE SCENE
             -                                    ..............................................................................      1
 A. I s the ReformProcessBringingthe SRCloser to Lisbon?............................................                                  1
 B. The Objectivesand ScopeofThisTask.........................................................................                        3
 C. The Structureofthe Report...........................................................................................              3

CHAPTER2 THE "NEW" SLOVAKREPUBLIC
             -
AND THE LISBONTARGETS              ............................................................................................       7
 A. The 2002 ElectionsLedto aWave ofFar-reachingReforms........................................                                       7
 B. The LisbonTargets for Employment RemainDistant.................................................                                  10
 C. The Post-2002ReformsWere Designedto BridgeThisGap......................................                                          12
 D.I sLisbonGettingAny Closer?....................................................................................                   14
 E. MonitoringCapacity I s CurrentlyVery Limited.........................................................                            16

CHAPTER 3 ATWO-EDGED SWORD: COMPETITIVENESS
             -
AND LABORMARKETDUALITY INTHE "NEW" SLOVAK REPUBLIC                                                                       .............19
 A. A GrowingLaborForcewith RelativelyLow Skills                              ...................................................    19
 B. EmploymentHas GrownRelativelySlowly andUnevenly........................................                                          22
 C UnemploymentHas RemainedUnchanged
  .                                                           ................................................................. 26
 D. Butthe UnemployedAre IncreasinglyExcluded........................................................                                27
 E. And EmploymentMobility RemainsLimited.............................................................                               28
 F. But Job DynamicsAre RelativelyHigh......................................................................                         30
 G. AndWage InequalitiesRemainLow ButAre Growing..............................................                                       36
 H. I sthe ReformBeginningTo BearFruit?.....................................................................                         40
 I. EmergingIssues...........................................................................................................        42

CHAPTER4 HAVE RECENTREFORMSWORSENED
             -
THE ODDS FORTHE MOSTVULNERABLE?                                    .............................................................     47
 A. Why Reformthe IncomeTax andthe Benefits System, andHow?.............................                                             48
 B. The Milestonesofthe ReformandItsMainFeatures.................................................                                    5 1

iv                                                      The Questfor Equitable Growth in the Slovak Republic


   C. Have the ReformsReducedthe Incomeofthe Poor?..................................................                                      59
   D. What HasHappenedto PovertyandIncomeDistribution?.........................................                                           63
CHAPTER5 .WHY ITREMAINSIMPOSSIBLETOADEQUATELY
MONITOR POVERTYAND INEQUALITY INTHE SLOVAKREPUBLIC                                                                            .............71
   A. DataLimitations PreventanAccurate Profile ofPoverty............................................                                     71
   B. Assessingthe Current Capacityto Monitor Trends inLivingStandards....................                                                72
   C . Areas ofPotentialImprovement..................................................................................                     73
   D. Conclusions.................................................................................................................        78


BREAKTHE CIRCLE OFPOVERTYAND DEPENDENCE?.....................................
CHAPTER 6 .CANCOMPETIITIVENESSAND INCENTIVES
                                                                                                                                          81
   A. Hasthe StrategyWorkedto Date andfor Whom?.......................................................                                    82
   B. Priority Actions Are Neededto Curb Poverty.............................................................                             85
   C. Improving the Capacity for Monitoring
       the Impactof RecentandFutureChanges I s Paramount                                  .............................................    92
REFERENCES..................................................................................................................... 93


ListofTables
Table 2.1:MainMacroeconomicIndicators. 1994-2004.........................................................                                   8
Table 2.2: RegionalVariations inUnemploymentRates inthe EU                                      .......................................    16
Table 3.1:Number of EmployedPersons inthe SlovakRepublic,2000-04 .........................                                                 31
Table 3.2: Job Flow Ratesby Regions. average over 2000-04 (%).......................................                                       32
Table 3.3: Job Flowsby Industries. Average over 2000-04 (in%)........................................                                      33
Table 3.4: Differences inJob Flows by Share of SkilledWorkers.
Average over 2000-04 (in%).................................................................................................                35
Table 3.5: The Share ofExcessJob ReallocationDueto Employment
Shifts within Groups (in%)................................................................................................... 36
Table 3.6: Gini Coefficientsby OwnershipType and Gender, 1999and2003......................                                                 37
Table 3.7: OLS Estimation Results-All Types ofOwnership
as a Dummy, 1999 and2003          ..................................................................................................       38
Table 3.8: Decompositionofthe Wage Differential According to OwnershipType,
1999and2003........................................................................................................................        40
Table 4.1: PersonalIncomeTax in2004 and2002................................................................                                52
Table 4.2: The Flat Tax inCentral Europe.............................................................................                      53
Table 4.3: SocialAssistanceBenefitsBefore andAfter the Reform, 2004 and2002............54
Table 4.4: Focusof Active Labor MarketPolicies in2004 and2002....................................                                          57
Table 4.5: Assumptions of the Benefit Impact Model                        ............................................................     60
Table 4.6: SimulatedPersonalIncomeTax by Income Deciles (yearly SIX),
2002 and2004........................................................................................................................ 66

Contents                                                                                                                                V



Table 4.7: ProgressivityandRedistributive Impactof the Tax System. 2002 and 2004 .......67
Table 4.8: Benefitsby IncomeDeciles(yearly SKK). 2002 and 2004                                      .................................  68
Table 4.9: Impact of Benefits on IncomeInequality. 2002 and2004 (percent).....................                                         68
Table 4.10: Impact of Reformson PovertyIndicators(?!increase)......................................                                    69
Table 4.11:Poverty Estimateswith Behavioral Changes. 2004 ............................................                                 70
Table 6.1:WhichActive Market PoliciesAre Most Likelyto Work?...................................                                        89


ListofFigures
Figure 2.1: InternationalComparisoninGDPTrends. 2000.04 ..............................................                                  9
Figure 2.2: Growth andEmployment. 1998-2004.................................................................                           10
Figure 2.3:International ComparisonsofActivity andParticipation Rates                                       .......................... 11
Figure 2.4: InternationalComparisonsof Employment
andUnemploymentby EducationLevel, 2004......................................................................                           12
Figure 2.5:InequalitiesinUnemploymentby EducationalLevel: SR and EU, 2004..........-15
Figure 2.6: UnemploymentRate andRegionalVariance.......................................................                                15
Figure 3.1:Trends inthe WorkingAge PopulationandIts Skill Composition,
SR, 1998-2004.......................................................................................................................   20
Figure 3.2: Trends inthe SkillMix ofthe Population,
Labor Force andEmployment, 1998-2004............................................................................                       21
Figure 3.3: SkillComposition of the Labor Forceand Sources ofIts Change, 1991-2004...21
Figure 3.4: Trends inEconomicActivity by Gender, the SR andthe EU, 15<age<64 .........22
Figure 3.5: EconomicActivity by Education, the SR andthe EU, 2004, 25<age<64 ...........22
Figure 3.6: Growth andEmployment, 1990-2004.................................................................                           23
Figure 3.7: International Comparisons of Responsiveness
of Employmentto GDP Growth ............................................................................................                24
Figure 3.8: Trends inthe Employment Rate, 1998-2004.......................................................                             24
Figure 3.9: Decompositionof Trends inthe EmploymentRate
andthe EmploymentLevel, 1999-2004................................................................................. 25
Figure 3.10:Trends inUnemploymentRates, 1998-2004.....................................................                                 26
Figure 3.11: Decompositionof Unemployment Rate Changes, 1999-2004.........................                                             -27
Figure 3.12: STU andLTU Cycles, 1998-2004.....................................................................                         28
Figure 3.13: UnemploymentRates, District Level, 2004 versus 1998..................................                                      29
Figure 3.14: Job Creation, Job Destruction and Employment Shares
by OwnershipCategoriesin2000-04 ....................................................................................                    34
Figure 3.15: Decomposition ofGenderWage Gap in 1999 and 2003 ...................................                                        39
Figure 3.16: SeasonallyAdjustedTrends inEmploymentby Type.......................................                                        41
Figure 3.17: Sources ofGrowth inthe Active Population....................................................                              $42
Figure 3.18: Negative Correlation Between Employment Rates
and Skills by Districts, 2004.............................................................................................              43

vi                                                      The Questfor Equitable Growth in the SlovakRepublic



Figure4.1:The "Poverty Trap" for DifferentFamilyTypologies
underthe 2002 BenefitsSystems......................................................................................................         50
Figure4.2: SocialExpendituresinthe SlovakRepublic,2001-04 .................................................                                 53
Figure4.3: Active LaborMarketPolicyin2004: Beneficiaries andCosts ....................................                                      58
Figure4.4(a): ExamplesofResultsofSimulations, 2004versus 2002:
InMaterialNeedfor SubjectiveReasons                ........................................................................................ 62
Figure 4.4@): Examplesof Resultsof Simulations, 2004 versus 2002:
InMaterialNeedfor ObjectiveReasons                  ................................................................................        63
Figure4.5: EffectivePersonalIncomeTaxRatesin2004 and2002
for a SingleWage-earner ................................................................................................................... 66


ListofBoxes
Box 1.1:The LisbonStrategy andEmploymentTargets...................................................................                           2
Box 3.1:Are RegionalInequalitiesinthe EmploymentSituation
a Matterof Skill Composition?     ......................................................................................................... 45
Box4.1:The Milestonesofthe Reform:Action PlansandNew Legislation.................................                                           51
Box4.2:Average SocialAssistance Benefits andNumberofRecipients.......................................                                       55
Box4.3:ReductionintheNumber ofRecipients
ofthe SocialAssistance BenefitsbetweenJanuary 2001 andDecember2004..............................                                            56
Box 5.1:ExistingDataCapacity for Monitoring Povertyinthe SR...............................................                                  73

                         ACKNOWLEDGMENTS

This participatory report is the result of a team effort involving staff of the Slovak Ministry
o f Labor, Social Affairs and Family (MoLSAF), Slovak and international experts, and a
multi-disciplinary World Bank team. The Bank team i s grateful to all government and non-
government officials who cooperatedwith the work ofthe team. The Slovak experts included
Matus Senaj (MoLSAF), Zuzana Siebertova (Comenius University,Bratislava) and Katarina
Kotovova (Comenius University,Bratislava).

TheinternationalconsultantsinvolvedwereRobertBarnes,ValerieEvans, TizianaTamborrini,
Manuela Coromaldi (University o f Torvergata, Rome, Italy) and Andrzej Tomaszewski
(University of Warsaw). The World Bank team was composed of Pierella Paci (Task Team
Leader and author of Volume I),      Anton Marcincin, Diane Steele, Marcin J. Sasin, Helen
Shariari, Marcel0 Bisogno, Bilal Erdogan, Helena Makarenko, Elena Markova and Emily
Evershed.

The process leading up to the report was highly participatory. The scope o f the work was
developed in consultation with Government and Slovak academics. Early findings o f the
background papers were presented in technical workshops held in Bratislava in which a
number of selected experts from the MoLSAF, the Ministry of Finance and the Statistical
Office (SO) participated. The findings were also extensively discussed with Mr. Miroslav
Beblavy, State Secretary o f the MoLSAF, and Mr. Peter Mach, President o f the SO.

Thisreportwasundertakenunder theguidance ofMr.RogerGrawe andMs.DanielaGressani,
Country Directors, Ms.Cheryl Gray, Sector Director, and Mr.Asad Alam, Sector Manager.
ThePeerReviewers forthereportwereMr.MiroslavBeblavy(State SecretaryoftheMoLSAF,
Government ofthe Slovak Republic), Ana Revenga (EASHD), Stefan0 Paternostro (PRMPR)
and Carlos Cavalcanti (AFTP4). In addition, the report has benefitedfrom comments and
suggestions from BernardFunck, Thomas Laursen, Mary Canning, DenaRingold, Anis Dani
and IngridBrockova.

The team i s grateful to the entire Bratislava office for their support and advice at various
stages o f the project. It i s also particularly grateful to the SO of the Slovak Republic for
permissionto use the Microcensus 2003 and the Labor Force Survey and to the MoLSAF for
permissionto use the InformationSystem onAverage Earnings (ISAE) database. Inaddition,
the team is gratehl to the Office o f the Plenipotentiary o f the Government of the Slovak
Republic for Roma Communities for its advice and support inthe design and implementation
o f the HouseholdSurvey o fVulnerable Communities, to the UNDP for its collaboration with
this survey, and to MichalVasecka and other staff ofthe Institute for Public Affairs for their
participationinthe Advisory Group on the survey and for general support and advice.


                           EXECUTIVESUMMARY

1.      The Slovak Republic (SR) was something ofalatecomer to the economic transformation
that swept through Central and EasternEurope inthe 1990s. However, since 1998 the country
has implementedan extensive and far-reaching packageof social and structural reforms which
affects areasranging frompublic finance to employment, education, healthand social benefits.
The very demanding political reforms implemented after the 2002 election, in particular,
have been praised by many for the depth and speed o f the changes and the sharp increase in
competitiveness and economic dynamism that they have brought about.' Indeed, there is wide
agreement that the radical and comprehensive package of reforms implementedby the 2002
government has positionedthe SR well-not         only inrelation to other EU8 countries but also
inrelationtotheEU15-in         terms o fincreasedcompetitiveness and employment opportunities,
two o fthe mainaims o fthe E ULisbon Strategyfor Employment and SocialInclusion.

2.      However, concerns have also been voiced about the potential negative effect of the
reforms on the other dimension of the Lisbon strategy: the drive for social inclusion. The
argument i s that in view of the pre-existing inequalities in access to employment across
regions and individuals, the increasedemphasison work as an individual's rightand duty, and
as the main route out o f poverty, may amplify the economic disadvantages and increase the
social exclusion o f those who face strong and complex barriers to employment.

3.      Given these contrasting views and the government's determination to curb social
exclusion and reduce poverty, it i s somewhat surprising that little systematic effort has
been made to monitor the impact of the reform process begun in 1998-and               the growth
it generated-on the incidence, depth and profile of poverty in the SR. Data limitation i s
the main reason why such a systematic evaluation has not been attempted. While this i s a
significant limitation, it i s extremely importantfor the sustainability o fthe reformprocessthat
its effects to date be quantified, even if inan imperfect manner.

4.      Withinthis context, this report provides the following:


        0  Abaseline analysis ofthe current performance ofthe Slovakeconomyinthe areasof
           employment, labor market segmentation and social exclusion, against which future
           development can be assessed and evaluated.
        0  Anassessment (to theextentpossible, giventhedataconstraints) ofthe impactofthe
           far-reaching structuralreformsimplementedafter the 2002 elections on(i) living
                                                                                        the
           standards o f the most vulnerable; (ii) distribution o f income and welfare; (iii)
                                                   the
          job creation andemployment opportunities; and (iv) the levels o f and inequalitiesin
           work income.
        0  Anefforttoadvise onany actions/policies that mayfurtherincreasethe effectiveness
           ofthe recent reforms as a strategy for promoting social inclusion.


I   Among other sources, the World Bank's Doing Business in 2005 Report rated the SR as (i)the
    world's top reformer in improving its investment climate over the past year, and (ii) 17
                                                                                        number
    amongtheworld economiesinterms ofease ofdoingbusiness,just beforeLithuaniaandwell ahead
    o f some o f the other EUmember states.

X                                              The Questfor Equitable Growth in the Slovak Republic



        0  An effort to advise on setting up the institutional structure for the regular and
           comprehensive monitoring of the dynamics of growth, job creation, employment,
           living standards and social exclusioninthe SR.

                EMPLOYMENT     OPPORTUNITIES     WERESLOWTO REACT GROWTH   TO

5.      The government that emerged from the September 2002 elections, as a modification
o f the previous coalition, inherited a country deeply dividedbetween a relatively prosperous
majority well equipped to benefit from the growing competitiveness of the Slovak economy
and the economic growth that came with it and a relatively small group that inthe aftermath
of economic transformationhad become increasingly trapped ina vicious circle of low skill-
low employment opportunities-state dependency-low intensity ofjob search. A crucial link in
this chain was the disincentive to work created by the generous social assistance system of
the 199Os, which led to some household typologies (typically, those with large numbers of
children and potentially low-earning adults) beingfinancially better off when out of work.

6.      Inanattempttobreakthisviciouscircle, thegovernmentundertookasetofpolitically
demanding and far-reaching structural reforms whose mainobjective was to foster growth
and increase average living standards by strengtheningthe flexibility and competitiveness
o f the economy and increasing the incentives for individuals to work and acquire human
capital. The focus was on employment as the main pillar of a comprehensive growth and
poverty reduction strategy. The economy reacted well, with economic growth rising to
well above 4 percent in2003 and reaching an impressive5.5 percent in 2004-double                 the
EU15 average and well in excess o f the average o f the Visegrad 4+ countries, although
somewhat slower than that of the Baltic States. In addition, the double-digit inflation of
the 1999-2000 period began to decline and a clear long-term prospect for good growth
performance emerged, conditional on the reform process not being halted or reversed.
Moreover, the combination o f growing GDP and declining inflation led to an increase in
average private consumption per capita (up nearly 3.5 percent in 2004) and a renewal o f
domestic demand, which suggests a positive trend in the average living standard o f the
Slovak p~pulation.~

7.      However, a concern sometimes expressed is that the reforms may not have been
equally successful inreducing poverty and tackling social exclusion and that the emphasis
on the individual's work potential as the route out o f poverty may have further increased
the social exclusion o f those who for whatever reason may not be able to find or seek
employment. In addition, in the flexible and competitive economy that has emerged over
the last year, the inability o f these groups to benefit from the country's high growth rate,
the overall improvement inliving conditions and the increase intargeted measures such as


    The fisegrad 4+ countries are the Czech Republic, Hungary, Poland and Slovenia. The Baltic
    countries are Estonia, Latvia and Lithuania. Figure 2.1 in Chapter 2 separates the EU8 into these
    two groups of countries because the Baltic States have growth trends that differed considerably
    from the other four countries, partly as areflectionof their differentinitialconditions.
    The GDPgrowthrate, which fell to as low as 1.5 percent in 1999, was 4.6 percent in2002 and4.5
    percentin2003 androseto 5.5 percentin2004. Thus, the post-2002reformshaveclearly beenvery
    successful inpromotinggrowthand in improving the averageliving standard.

Executive Summary                                                                             xi


activation programs and other labor market policies (see below) may have increased their
economic vulnerability. This concerni s reinforcedbythe findingthat labor market outcomes
have lagged somewhat behindGDP growth and that, as employment opportunities for the
average population have improved, unemployment has become increasingly concentrated
among the unskilledand in specific regions.

8.     Indeed, the high economic growth o f recent years has translated into a growth o f
employment at rates that, with the exception of 2003, are considerably lower than those of
the EU15 countries although somewhat higher than the average for the EU8. The slowdown
of 1999-2000 resulted ina drop inemployment between 1998 and 2000 o f 100,000 jobs (or
nearly 5 percent of total employment). After 2000, the increase inGDP growth resultedina
mildbutmuchsmaller increaseinemployment. In2003 the increaseinemploymentwas close
to 2 percent (less than halfofthe GDP growth rate for that year) and employment actually fell
over the first half o f 2004 only to pick up sufficiently inthe second half to obtain an overall
increase of 30,000 jobs, despite a 5.5 percent growth rate.

9.     Duringthe same period the working age population continued to increase-slightly
until2003 andthenby as muchas2 percent in2004-and            the activity rate ofthe working age
population increased slightly, leading to a growth inthe labor supply o f 128,000 people ('just
over 5 percent) over the period 1999-2004. Thus, the small increase inemployment was just
enough to absorb the increaseinthe labor supply, and unemploymentremainedunchanged.

10.    However, what is particularly striking for the external observer i s not so much the
recent trends in labor market indicators but the absolute level o f some of these indicators.
At around 70 percent, the participationrate of the working age population inthe SR i s very
similar to the average for the EU15 countries and considerably higher than the EU8 average
o fjust over 66 percent. Nevertheless, the following considerations are worth noting:

           At less than 58 percent, the employmentrate inthe SR is the thirdlowest inEurope
           (with only Hungary andPolandhaving lower rates) and is far lower thanthe Lisbon
           target o f 70 percent.
           As a result of the average participation rates and the very low employment rates,
           the SRhasthe secondhighestunemploymentrateinthe EUafter Poland, morethan
           twice the EU15 average. Inaddition, the unemployment rate among workers with
           only a basic education i s nearly five times higher than inthe EU15 and more than
           double the EU8 average.

                LABOR    MARKET   SEGMENTATION HASHISTORICALLY HIGH   BEEN
11.    O fparticular relevance for policymakers i s the fact that muchof the difference inlabor
market indicators between the SR and the rest o f Europe i s due to the extreme poor prospect
of particular groups o fworkers, with the most strikingdisparities beingacross regions and in
skill levels. Thus, for example, in2004:


       0   Over half of the workers with only a basic education were unemployed, an increase
           of over 70 percent since 1998.
       0   The unskilled were 2.5 times more likely to be unemployed than the average

xii                                        The Questfor Equitable Growth in the Slovak Republic



          population, by far the second highest differential among the EU25 after the Czech
          Republic.
       0  The SRwas the EUcountry with thehighestproportionoflong-termunemployment
          (LTU) (65 percent o f the unemployed) and the lowest proportion o f short-term
          unemployment (only 10 percent of total unemployment), and a clear link between
          low skills and LTUwas emerging.
       0  Regional differences inthe incidence of unemployment are also very large. At the
          NUTS3 level, the SRhasthe secondlargestvariance inregionalunemployment rate
          intheEU25afterItaly. Italsohasthe secondhighestregionalunemploymentratein
          the EU25 (at NUTS2 level), after Italy.

12.    Similar conclusions are derived from an analysis of labor market dynamics observed
inthe Slovak enterprise surveys. These conclusions support the hypothesis ofincreasedlabor
market duality, with a dynamic "segment" growing in accordance with the growth o f the
economy and the other segments stagnating or even contracting over the same period. The
dynamic part i s driven predominantly by foreign companies, which have positive net job
creation (i.e., they are creating more jobs than they are destroying), have a growing lead in
terms of the productivity-related characteristics of the labor force (especially education), and
are paying increasingly higher wages. These jobs are concentrated mainly in the Bratislava
region and in other regions with easy access to markets, and they attract the better qualified
workers.

13.    The considerable differentials within regions and types o f workers are not a new
phenomenon inthe SR. On the contrary, it has been a latent feature of the Slovak economy
since socialist times, although one that was hidden to some degree by the emphasis on full
employment o f the pre-transition area. The economic liberalization that followed the 1998
election brought this disadvantageto the surface as the shift infocus from full employment to
increasedproductivity resulted ina sharp reduction inthe employment opportunities for low-
skilled workers. The incidence ofunemploymentfor these workers grew sharply and became
increasingly of a long term nature. However, untilthe 2004 reforms, the living standards of
these workers were, for the most part, maintainedby very generous social benefits which for
some household typologies exceededtheir potential labor income.

14.    However, these generous benefits substantially reduced the incentives for potentially
low-paid unemployed workers to actively seek employment, and many left the labor force
entirely. Thus, while successful inpreventing households from falling into financial poverty,
theywere amajor cause of'socialexclusion for these workers andtheir families. This is what
the past 2002 reforms sought to address.

Executive Summary                                                                           ...
                                                                                           Xlll




          WHATI s    THE GOVERNMENT'S   STRATEGY  FOR  TACKLING  SOCIAL EXCLUSION?
15.    With the 2003 "Strategy o f Growth in Employment based on the Labor Market
Reform" the newly elected government set out a comprehensive strategy to reduce the
historically highlevel o f labor market disadvantage faced by some groups. The cornerstone
of this strategy was the reforms o f January 2004, which were designed to break the vicious
circle o flacko fskill-poverty-dependence that ledto the wideninggapbetweenthe economic
opportunities o f the skilled and the unskilled. The strategy was multi-pronged and was
basedon the principles o f (i)  "making work pay"; (ii) increasingemployment opportunities
for everyone but targeting the most vulnerable groups; and (iii)  addressing the issue of skill
disadvantage by increasing the educational attainments o f the most vulnerable groups in
general and o f their children inparticular.

16.    For the point ofview ofthis report, the mainelements ofthe reform canbesummarized
inthe following list:


       0   The implementation of a new Labor Code, designed to foster labor market
           flexibility, by:

           o Introducing the option of a number ofnon-standard labor contracts
           o Increasing the employer's flexibility to terminate a contract.


       0   The reform of the welfare system, designed to (i)     eliminate the poverty trap by
           increasing incentives to work and (ii)
                                                improve the targeting of social welfare benefits
           by:

           o Replacingthepre-existingtwocategoriesofbeneficiarieswithasingleentitlement
              structure basedon current, as opposedto past, individual labor market status and
              search
           o Reducingtheminimumbenefitfor asingleadulttoanamountbelowthesubsistence
              minimumandweakening the linkbetweenfamily size andbenefit level
           o Introducing anumber ofnew benefits for the workingpoor, available to families
              in which the adults are either working or participating in active labor market
              programs (ALMPs)and expanding the entitlement to socialbenefits to the first 6
              months o f employment
           o Providing subsidies to municipalities (and other stakeholders) for organizing
              training and targeted employment opportunities for the LTU ("activation
              programs")
           o Replacing the previous income-related family benefits with a flat (not income-
              related) benefit per each child coupled with tax credits for minor children
           o Introducing basic and high-school scholarship schemes inaddition to meal and
              schooling subsidies.


       0   The labor market reform, designedto improve the employment opportunitiesand
           potential incomes of disadvantaged groups and the long-term unemployed through
           the following means:

xiv                                          The Questfor Equitable Growth in the Slovak Republic



          o Individualandtargetedworkwiththeunemployed(in-depthinterviews, individual
              action plans for job-seekers, testing and assessment of personal characteristics,
              etc.)
          o The mandatory character oftheALM programs andthe reimbursement o ftravel
              andchild-care coststojob-seekers actively seekingemployment andparticipating
              inALMprograms
          o Subsidies to employers hiring LTUand to LTUjob-seekers planning to start a
              business
          o Targetedinterventions to regions with the highest incidence o funemployment
          o An increaseinthe minimumwage of around 5 percent inreal terms since 2003.


      0   Thereformoftheadministrationofwelfare systems,designedto mergethe social
          agenda with the labor market policies inthe regional offices.

17.   The government's strategy for tackling social exclusionis thus comprehensive, radical
and imaginative. All o f the mechanisms seem to be in place to address poverty and social
exclusion inthe long term by interrupting the vicious circle that has prevented the poor from
breaking out o fpoverty for generations. However, four questions emerge:


      0   Has the strategy worked to date?
      0   Who are the gainers and losers inthe reform?
      0   Can the effectiveness o f the strategy be increased?
      0   How can the living standards of the most vulnerable be supported without reducing
          the incentive to participate inthe drive to "earnyour way out ofpoverty"?

                              HAS STRATEGYWORKED DATE?
                                   THE                     To
18.   A first assessment o f its impact on a number o fbasic indicators suggests that, overall,
the "Strategy o f Growth in Employment" has been reasonably effective in achieving the
followingtargets:


      0   Eliminating the disincentive to work provided under the old system by the high
          replacement ratio faced by individuals with a low earningpotential
      0   Increasing the disposable income o f the majority o f households that are entitled to
          benefits
      0   Providing better employment opportunities and access to ALM programs for the
          average worker and for the most vulnerable groups.

19.   However, questions remain regarding the magnitude of these effects and their
sustainability over time. Inaddition, it is important to note that, giventhe short time-frame of
the analysis, itis difficult to establishwhether the recentpositivetrends insomeindicatorsare
the result of the reforms or the effect o f short-termtrends.

                      WHOARETHE WINNERSAND WHOARETHE LOSERS?
20.   Owing to data limitations, the impact of the reform on income and income poverty
cannotbeevaluatedbut canonly besimulated. Thisreportuses two different approaches. First,
it computes pre-reform andpost-reformincome levels for different household typologies and

Executive Summary                                                                             xv


calculates the net impact o fthe reformby comparing these two levels. As a secondapproach it
uses aTax-Benefit Modelto compare for all samplehouseholds their income levels before the
reform with the simulated post-reform levels so as to identifythe gainers and losers as well as
the distributional and poverty changes.

21.    The first approach reachesthe following conclusions:


       0  The combined effect of the reform o f the tax and benefits regime has increased the
          disposable income o f the majority o f householdsthat are entitledto benefits
       0  The only households that may have lost out are those with three or more children
          where all o f the adults are unemployed, and that were entitled before the reform to
          a higher benefits schedule
       0  Even for the most affected households the maximum potential loss in disposable
          income is limited.

22.    These conclusions are confirmed by the findings of the Tax-Benefit Model, which
concludes that the reform has only a small potential first-round effect on the incidence of
poverty, with the poverty headcount reduced by 2 to 3 percent according to simulations
based on the 2003 Microcensus. This i s the case despite an overall reduction o f around
16 percent in the average transfer paid and thanks to the improved targeting o f the social
benefits.

23.    A primary factor behind this result is the reduction o f the tax burden on the lower
income groups which has beenassociatedwith the move to a "flat rate" personal income tax
(PIT) combined with a substantial increase in deductions and credits. This i s an important
result, since it shows that concerns about the potentially negative distributional effect o f the
tax reform may be misplaced. Inthe Slovak version, the flat rate tax has not been merely an
efficiency measure. Its implementation i s likely to have increased the progressivity of the
income tax system while replacing five tax brackets (and a fairly steep scale of tax brackets)
by a single one. Indeed, in2004 the PIT burdenwas reduced for everyone, but particularly
for households with lower incomes. This was due to the effect o f the increase in deductions
and credits which has more than compensated for the reduction in progressivity associated
with the move to a single tax rate. However, while the incidenceof poverty has declined, its
intensity and the related poverty gap have increased considerably-by       81 and 62 percent,
respectively-    which suggests that the income levels o f those in poverty have declined as
result of the change inthe tax-benefit regime.

24.    Much o f the ultimate impact o f the reforms, however, depends on their second-
round impact on labor market outcomes. The use o f quarterly data suggests that the reform
package implemented after 2002 may have been associated with a relatively high increase
inemployment with a year to year seasonally adjustedgrowth of 1.7 percent in2003, driven
primarily, but not entirely, by a growth in self-employment, and a subsequent slowdown to
0.3 percent in2004. This interpretation is, however, biased to the base effects. An inspection
of quarter to quarter changes reveals that after a sharp drop inthe second half o f 2003 (-0.3
percent and -0.4 percent in the third and fourth quarters, respectively) overall employment
began to increase in 2004 with a total increase of 46,200 jobs-and      a 0.4 percent increase

xvi                                          The Questfor Equitable Growth in the Slovak Republic



inthe employment rate-in       the period from the second quarter of 2004 to the first quarter of
2005 (all seasonally adjusted).

25.    It is too early to determine to what extent this is the beginning of a trend and whether
this is attributable to the success of the 2004 reforms, but the signs are encouraging.The fact
that this growth is drivenexclusively by the growth inself-employment-in           the context of the
decliningnumber ofemployees-is          alsoworth noting, as itpointsto economic liberalizationas
the maindriver andhas important implications for policymaking.

26.    Incontrast, the reformsof2004 do not seem to havehadany significantimpactonthe
participation rates o f the Slovak population. The steady growth o f the active population (at
around 1percentin2004)wasexclusivelytheresultofthegrowthintheworking agepopulation.
Combinedwith the trends inemployment, this resultedinthe rate o funemployment peaking
during the first quarter of 2004 at 19.5 percent (15-64 age brackettthe second highest
figure in the records (after the corresponding period o f 2001). Subsequently, this rate fell
rapidly throughout 2004 and 2005.

27.    Ofparticular interest istheassessmentofthe effect o ftheALMPsalready implemented.
The available data show that the "activation contribution" i s by far the most popular policy
introduced to date in terms o f number o f beneficiaries and i s a very successful measure,
despitethe lackofemploymentopportunities insomeareas. However, thequestion ofwhether
participation in this scheme will permanently increase the employment opportunities of the
participants remains open. The internationalevidence regardingthe success of these types of
interventions i s contradictory and country-contingent,

28.    However, it i s also important to realize that, in addition to their potential effect on
futureemploymentprospects,ALMPscanplay asignificant roleas astrategy to reduce social
exclusion. This is particularly important ina dual labor market such as the SR, where a small
pocket o fthe labor force i s at riskofbecoming increasinglymarginalizedand "discouraged".
Withinthis context, theprograms canbeseenasaway ofkeepingtheseworkers engagedinthe
labor market and the economy rather than having them rely exclusively on social assistance.
Thus, these policies would act more as an alternative form o f social assistance, designed
to break the cycle of dependence and social exclusion, rather than as a route to permanent
employment.

                    CANTHE EFFECTIVENESS STRATEGYBEINCREASED?
                                             OF THE

29.    Themainmessagesthat emergefromthisreportinrelationtoincreasingtheeffectiveness
of the reform strategy are outlined below:


       0   Itisstilltoo earlytopassfinaljudgmentastowhether thepolitically demandingstrategy
           adopted by the Slovak government after the 2002 election for reducing poverty and
           social exclusion hasworked. This ispartlydue to data limitations andpartlyto the fact
           that the full impact o f the new policies i s not yet fully reflected inthe available data.
           Some ofthese policies will also need some time to show their full effects.
       0   However, signsarepositive, withmanyindicatorso faccessto economic opportunities
           showingclear improvement since the middle o f2004. Whether these positive trends

Executive Summary                                                                        xvii



          are the results ofthe reforms rather than ofinternationalmacroeconomic trends, EU
          accession or other factors remains to be ascertained.
       0  Inthe meantime, itis essential(i) set upanadequatemonitoringsystemthat can
                                             to
          provide a reliable baseline against which future developments can be evaluated; (ii)
          to conduct detailed evaluations of the impact of the reform on a regular basis; and
          (iii)institutionalizeasystemthroughwhichthefindingsoftheevaluationarefed
               to
          back into policymaking. This would add an essential element o f evidence-based
          policymakingthat would increasethe effectiveness o fthe reform process as well as
          enhancing its long-mn sustainability.

30.    Intheclimateofreformandchangethathasprevailedoverrecentyears,thegovernment
needsaccurate, up-to-date andrelevantinformationinorder tomakesoundeconomic andsocial
policy decisions. I t needs to measure and monitor poverty, employment and unemployment,
social inclusion and other dimensions of living standards at regular and short intervals. In
addition, in the context of increased accountability that characterizes the new SR, analysts
need data with which to model economic behavior and to provide answers to questions about
whether instituted policies have been successful, and also to see how new policies might
affect the population. However, the lack o f systematic collection of good quality data on the
main indicators of welfare, as well as the limited analytical capacity, makes it impossible to
satisfy the government's objective o fevidence-basedpolicymaking. Unfortunately, evidence-
based policymaking in the SR i s presently hampered by the lack of reliable and regularly
produceddata.

31.    Inshort, ifthe economic transformationofthe SRisto becomplete, it is essentialthat:
(i) statisticalsystemisupdatedtotheleveloftheotherEU25countriesandtotheminimum
   its
level required for adequate monitoring and reporting to the EU on priority indicators; (ii)
adequate capacity i s built within the government for evidence-based policymaking; and (iii)
closer links are establishedbetweenthe StatisticalOffice (SOSR) andthe Slovak government,
especially theMinistryo fLabor, SocialAffairs andFamily (MoLSAF) inthecaseo fmonitoring
living standards. Both the SOSR and the government agree with these conclusions and are
working towards improving capacity.

      CANTHE LIVING     STANDARDSOF    THE MOST   VULNERABLE   RECEIVE  MORE  SUPPORT?
32.    Itisalso important to recognizethat evenunderthe bestpossible scenario, inwhichthe
reformprovesextremelysuccessfulinpromotingemploymentopportunities and"making work
pay", pockets o funemploymentand social exclusionwill remain, especially ina country with
a highly segmentedlabor market, like the SR. Inorder to improve the living standards o f the
most vulnerable, a comprehensive and effective system o f social assistance i s needed. While
the 2004 changes inthe tax and benefits structure have potentially increased the disposable
income slightly for most low-income households, they seem to have had a negative effect on
the income level o f poor households with unemployed adults and large numbers of children.
These households are highly concentrated at the bottom end o f the income distribution (both
pre- and post-2004) and thus it i s not surprising to find that the 2004 reforms may have
left the poverty headcount basically unchanged but increased the intensity o f poverty, as the
disposable incomes o f the poorest households were the only ones negatively affected by the
reform.

xviii                                       The Questfor Equitable Growth in the Slovak Republic



33.    To some extent this was the unavoidable consequenceof the shift to work incentives
and the reduction in the basic benefits which was required to reduce the fiscal pressures
generatedby the formerly overgenerousbenefits system. However, inview of the barriers to
employment that remain (at least inthe short run) for unskilledworkers and for those in the
lagging areas, the government may wish to target these groups and communities with short-
term measures designedto help them overcome the loss of livelihood.

34.    In the context of the emphasis on providing the right incentives, any individual/
household level intervention o f this type would needto be designed as a conditional transfer
program(i.e., as one inwhich entitlement to thebenefit is conditionalon someform ofpositive
behavior). As the focus of this intervention i s likely to be households with large numbers of
children, it may be advisable to make the condition related to children. Programs of this type
that have proved to be successful insimilar contexts have provided transfers conditional on
school attendance and regular visits to health clinic^.^ Such programs directly address the
vicious circle of the intergenerational transmission ofpoverty that i s so evident inthe SR.

35.    The 2004 reform has already introduced a number o f conditional transfers such as
the "activation allowance" and the secondary and higher education scholarships. I t has also
introducedsimilarlytargetedinterventions atthecommunitylevelviathetrainingandeducation
grants and other interventions. These are steps inthe right directionthat needto be reinforced
inorder to haveamoreperceivable effect andto ensure their long-termsustainability.

36.    Inprinciple, fundingfor these additional types ofinterventions shouldnot be difficult
to obtain. The simulations in this report suggest that in its current format the new benefits
system maintains a relatively highdegree o f inefficiency, with benefits beingpaidnot only to
the poor but also to households inthe top income deciles. This i s partly due to the shift away
from income contingency o f all family-related benefits. While child-benefits tend not to be
income related across the EU countries, it is arguable whether this i s a choice in a country
with limitedresources andvery deep pockets of poverty. Inaddition, the removal of income
conditionality for the parental allowance i s more difficult to justify inthe context of limited
resources for social assistance. Reverting to a system of income-related family benefits may
generate enough savings to finance the increased emphasis on conditional transfers.




   Programsofthis type that havereceivedparticularattentionare, for example,Progressa inMexico
   andBorsa Escolar inBrazil.

                                    CHAPTER 1              -
                           SETTINGTHE SCENE

1.1   The Slovak Republic (SR) i s one of the eight Central European (CE) countries which
joinedthe EuropeanUnion(EU) inMay2004 (EU8) andthe onewithbyfar the mostprofound
disparities, rooted in a long history of geographical and ethnic inequalities. Somewhat of a
latecomer to the process o f economic transformation that swept through Central and Eastern
Europe inthe 1990s, the SR has since 1998become a strong and innovative reformer.


    A. I s   THE   REFORM     PROCESS      BRINGING    THE  SR CLOSER LISBON? TO


 1.2   Slovak and international observershave increasingly come to praise the depth andspeed
of the changes and the sharp increase in competitiveness and economic dynamism that have
occurred inthe SR.Among other sources, the Doing Business in 2005 Report5rated the SR as
the world's top reformer in improving its investment climate over the past year, and number
 17 among the world economies interms of ease of doing business, just before Lithuania and
well ahead of some of the other EU member states. Thus, in the eyes o f many, the reforms
have provided the SR with the possibility of taking full advantage of the new opportunities
presentedby its recent entry into the EU. Inparticular, there is wide agreementthat the radical
and comprehensivepackage o f reforms implementedby the 2002 government have positioned
the SR well-not    only relative to other EU8 countries but also relative to the EU15-in terms
of two of the main aims o f the EU Lisbon Strategyfor Employment and Social Inclusion (as
described in Box 1.1): increased employment and competitiveness. Indeed, the spirit of the
2004 reforms was very much in line with the principles underlying the Lisbon Treaty and, in
particular, with the objectives of "making work pay" and increasing employment opportunities
for the overallpopulationby increasing the competitiveness o fthe economy.

 1.3   However, concerns have also been voiced about the potential negative effect o f the
reforms on the other dimension o f the Lisbon strategy: the drive for social inclusion. The
argument i s that-given    the extent of the pre-existinginequalities inaccess to employment
across regions and individuals-the     increased emphasis on work as an individual's right and

    WorldBank (2004), Doing Business in 2005.

2                                                The Questfor Equitable Growth in the Slovak Republic



dutyandasthemainrouteouto fpovertymayamplifypre-existingeconomic disadvantages and
increase the social exclusion o f those who face strong and complex barriers to employment.

1.4     Given these contrasting views and the government's determination to curb social
exclusion and reduce poverty, it comes as something o f a surprise that little systematic effort
has been made to monitor the impact o f the reform process begun in 1998-and                the growth it
generated-on the incidence, depth and profile of poverty inthe SR. Data limitation is behind
the failure to produce regular poverty profiles that are reliable and comparable over time. Until
the release of the Microcensus data last summer, the latest quantitative indicators referred to
1996, well before the beginningofthe reformprocess.6


                         Box 1.1: The LisbonStrategyand EmploymentTargets
  The 2002 Councilmeeting inLisbonwas set upto makethe EUthe most competitive and dynamic
  knowledge-based economy inthe world, capable o f sustainable economic growth with more and
  better jobs and greater social cohesion (EC[2003a]). The strategy that followed had three main
  dimensions-employment targets, competitiveness and social inclusion-and              was designed to
  enable the EUto regain the condition o f full employment and to strengthen cohesion by 2010.
  Inorder to anchor thepolicies, theEUsetambitioustargets for allEUcountriesatahorizonof2010.
  Inthe caseofemployment, thetarget isanemploymentrateof70percentonaverageand60percent
   for women. New targets have been added inStockholm: namely, to have an employment rate o f 50
  percent for older workers by the same time, and inBarcelona: to delay by five years the exit age
   from the labor force. In 2004 and 2005 the European Commission concluded a mid-termreview
   of the Strategy andnotedthat progress hadbeen slow to date: while there was some improvement
   inemployment rates, particularly for women, withnetjob creationwaning recently, the 70percent
   target will be "difficult to reach" by 2010 (EC [2003b]). Progress was also disappointing on many
   other fronts. These failures were attributed to a too broadscope o f the strategy and, most important,
   to a lack o f political commitment.
   Asaresult,theCommissionistryingtoreinvigorate the Lisbonprocess.Inparticular, itintendstomake
   itmore selective andfocused ongrowth,jobs, competitiveness andknowledge economy. Interms of
   labor marketpolicies, this wouldmean implementing the recommendations o fthe EC's Employment
   Taskforce. In its report, the Taskforce has identitied four key requirements on which success in
  job creation will depend. These are: (0 increasing the adaptability of workers and enterprises; (ii)
   attracting morepeople to the labor market; (iii) investing moreandmore effectively in human capital;
   and (iv) ensuring the effective implementation of reforms through better governance.

   To address the commitment problem, the Commission intends to strengthen ownership through
   puttingmore explicit responsibilities on countries. Tasks and accountabilities are to be clearly
   assigned between the EU and the national level. According to the proposal, governments will
   be expected to prepare and discuss with their parliaments a single NationalAction Program for
   growth andjobs as well as taking ownership o fthis implementation.
   However, for the moment, the 2010 deadline has not been lifted.


 1.5     Inarapidly changingenvironment, the expectationisthatthepovertyprofile islikelyto
have changed considerably since 1996. However, owingto the absence o f regularly produced



    World Bank (2002), Slovak Republic: Living Standards, Employment, andLabor Market Study

Chapter 1                                                                                           3


nationally representative dataonincomeandconsumption, the 2003 Microcensusprovidesthe
only potential source o f information on poverty and inequality inrecent years. However, this
datasethas severallimitations as asourceofinformationfor bothastatic poverty profile andan
adequateevaluation ofthe impact ofthe reformprocess. Iti s unsuitable for deriving areliable
poverty profile for 2002 because: (i)   owing to the way the sample is selected and to the very
low responserate, the data are likely to be strongly biasedtowards anunder-representation of
the poorest groups inthe Slovak society; (ii) does not contain information on consumption
                                                 it
or other wealth indicators; and (iii)   the income information is highly unreliable and differs
considerably inthe individual and the household files.7

1.6    Inaddition, the data cannot be usedto evaluate the impact of the reform process on
poverty and social exclusion because they refer to 2002, and thus they precede the most
radical reformswhich were implementedinJanuary 2004. Inaddition, the sample o fthe 2004
Microcensus i s derived from the 2001 Census. While this makes the data inprinciple more
representative ofthepopulationo fthe SRthanthedatainthe 1996Microcensus-with                asample
drawn from the obsolete 1991 Census-it         also makes the two samples highly incomparable.
Therefore, changes in poverty indicators over the two datasets do not necessarily reflect the
impact o f the reform.

1.7    Nevertheless, despite the data limitation, giventhe speed of change and growth inthe
SR over recent years and the government's focus oncurbing poverty andsocial exclusionvia
improved labor market opportunities, it is paramount to attempt at least a partial evaluation
of the impact of the reforms with the available data, however limited.The sustainability o f
the reformprocess and its success rely heavily onthe ability to carry out this evaluation on a
regular basis and to feed the findingsinto policymaking.


                   B. THEOBJECTIVES SCOPEOF THISTASK
                                                AND


1.8    Withinthiscontext, theaimofthisreportistoprovideaninitialevaluationofthe impact
o f the reform process on those areas which it was most designed to address: labor market
outcomes and social exclusion. The focus of the labor market analysis is on the two distinct
phases of the reform as highlightedabove: (i)       the 1998-2003 period, during which changes
were relatively small andthe focus was primarily onmacroeconomic andfiscal sustainability;
and (ii) post-January 2004period, when radicalchangesto the Labor Code were combined
         the
withfar-reaching reforms o fthe tax andsocialbenefits systems inanattempt to boostboththe
demand for and supply o f labor. However, given the short interval since the implementation
of the 2004 reforms, the following limitations shouldbe taken into account:



    The difference between the income values in the individual and the household files may be the
    source of the wide differences in poverty indicators for the SR reported by different sources. For
    example, within the space o f a couple o f months EUROSTAT and the European Commission
    came up with two very different poverty rates (for the same poverty line and both based on the
    Microcensus 2003). The range was from 5 percent (which made the SR the country with the lowest
    relative poverty inthe EU) to 21 percent-the highest poverty incidence inthe EU.

4                                           The Questfor Equitable Growth in the Slovak Republic



       e  The report can, at best, pick up only the short-term impact, which-given          the
          complexity of the reform packageand its focus on creating the right incentives-is
          likely to be anunderestimation ofthe cumulative long-term impact
       e Any conclusion i s necessarily tentative, as it may reflect cyclical variations rather
          than medium-termtrends and thus it is difficult to derive firm conclusions on the
          basis of such a short time span.

1.9    Data limitation prevents a reliable analysis of the effect o f the reforms on living
standards and social exclusion, as data on these aspects of individual and social well-
being are only available at a point in time (namely, 2002) and even then they suffer from
importantlimitations. The approach followed inthis report i s to usethe 2002 datato construct
counterfactual pre-2004 and post- 2004 income levels usingsimulationtechniques.

1.10 Withinthese limitations, the main objectives of this task are the following:

       e  To provide a baseline analysis of the current performance o fthe Slovak economy-
          inthe areas of living standards, employment and social exclusion-against       which
          future development can be assessedand evaluated
       e  To advise on any actions/policies that may further increase the effectiveness o f the
          recent reforms as a strategy for promoting social inclusion
       0  To set up the institutional structure for the regular and comprehensive monitoring
          o f the dynamics of growth, job creation, employment, living standards and social
          exclusion inthe SR.

1.11 The scope of this task was decided in close consultation with the government of the
SR, civil societies, Slovakthink-tanks andacademicinstitutions, andits focus reflects closely
the interests and requirements of the government. The choices were to focus on the limited
number of issues highlighted above with particular reference to the linkbetween promoting
competitiveness and enablingreforms onthe one hand, and improving labormarketoutcomes
and the living conditions of the most vulnerable groups on the other.

1.12 The report fits into a broader program designedto provide technical assistance to the
Ministryof Labor, SocialAffairs andFamily (MoLSAF) and other relevant ministriesinthe
following ways:

          Evaluating the impact o f the SR's recent economic success and the impact of its
          comprehensive and cross-cutting package ofreforms on the economic opportunities
          and living standards of the most vulnerable groups
          Enhancing the current capacity for (i)     monitoring poverty and social exclusion,
          (ii) evaluating the impact o f economic reforms, and (iii)delivering evidence-based
          policymaking.

1.13 Thus, this task goes beyonda traditional report andhas the combinedcharacteristics of
Analytic andAdvisory Activities (AAA) and technicalassistance.Adopting this non-standard
and highlyparticipatory approach has allowed the team and the task to be more responsive to
the specific needs of a highmiddleincome country inthe midstofprofoundtransformation.

Chapter I                                                                                     5


However, compared to atraditional LivingStandardsAssessment (LSA), itdoes notattempt to
quantifystandard indicatorsofpoverty, to derive apovertyprofile or to analyze its dynamics.
This decisionis motivatedby the persistent lack ofreliable, nationallyrepresentative data on
income and consumptionand the total absenceof information onother dimensions ofpoverty
and vulnerability. These limitations have been highlightedrepeatedly in the past but never
fully addressed. This LSAhas made a deliberate attempt to move forward by increasing the
quality o f the data and creating the necessary institutional capacity to make full use o f the
improved data inpolicymaking.


                          C. THESTRUCTURE OF            THE  REPORT


1.14 The format of this reporti s also somewhatunusual. I tconsistsofthree separateeconomic
notes presented intwo volumes. Volume Iis a BriefingNote that presents the mainfindings
and their policy implications. The remaining volume contains a set of Policy Notes designed
to address the priority policy issues selectedjointly by the government andthe team.

1.15 The choice o f this format responds to the need (i)    to react more quickly and provide
more timelypolicy advice to acountry experiencing rapid anddeeptransformation, and (ii)       to
set the timingo fthe differentpieces so as to meet the requirement o fthe reformprocessrather
than being conditioned by the availability of all of the material. Indeed, for the most part our
Slovak partners have had early access to the Policy Notes, and early discussions have taken
placebetweenthe team andthe government on eacho fthese components. Inaddition, follow-
upactivities have already beenputinplacebyboththe WorldBankteam andthe government
o f the SR. This i s particularly true of the Technical Note Building Capacityfor Systematic
Monitoring of Poverty and Social Exclusion in the Slovak Republic, which led to follow-up
activities with the StatisticalOffice of the SR (SOSR) - via the StatisticalCapacity Building
Trust Fund- and within the MoLSAF and the Ministryo f Education (MoE) -via a technical
assistanceloan.

1.16 After consultationwith the government of the SR and the World Bank Country Team,
three self-standing Policy Notes of highpolicy relevance were included inthis report:

           An analysis of trends and dynamics inthe labor marketwith special reference to a
          profile ofemployment andjob creationandananalysis o fthe barriers to better labor
           market indicators at the national, regional and individual levels
           A Technical Note on data limitations and suggestions for capacity building in the
           area of poverty monitoring and impact evaluation
           A Poverty and SocialImpactAnalysis (PSIA), lookingat the distributive effects of
           the recent reforms ofthe personal income tax andbenefits systems.

 1.17 An additional component o f his programmatic task has been postponed until the end
of 2005 owing to delays inobtaining the necessary data. This final Policy Note i s intended
to respond to the lack o f nationally representative data on different indicators of welfare by
compiling a multidimensionalprofile o f poverty invulnerable communities. The profile i s to

6                                          The Questfor Equitable Growth in the Slovak Republic



be based on a statistically representative survey of selectedvulnerable communities, jointly
financed bytheWorldBankandthe UNDP.Themultidimensionalhouseholdsurvey, currently
inits final stage, focuses onthe communities coveredby the Socioeconomic Mapping of the
Roma Communities.

1.18 Volume Ibegins by describing the rationale for the report, its scope and objectives. It
then moves on to give an overview of the macroeconomic climate and its changes over the
last decade, with particularreference to the periodof economic liberalization that began with
the 1998 election. Chapter 3 analyzes in some depth the main trends inthe labor market in
an attempt to explain why the rapid economic growth experienced over recent years has not
resulted in equally positive trends in employment. Chapter 4 assesses the impact o f recent
reforms to the tax and benefit systems on poverty and income inequality, while Chapter 5
describes the limitations of the data currently available and the system currently inplace for
monitoring poverty and social exclusion in the SR. Finally, Chapter 6 makes a number of
suggestions onhow the current strategy for employment and social inclusioncanberendered
more effective.

                                   CHAPTER2 -

     THE "NEW" SLOVAKREPUBLICAND THE
                              LISBONTARGETS

2.1   Muchhas takenplace inthe SRbothpolitically and economically over the last decade,
and much o f the changes occurred after the 1998 elections, which brought a wave of key
liberalization policies, growing inflows o f foreign direct investment (FDI) and large-scale
industrial restructuring, triggering a significant productivity growth. The introduction o f
measures designed to reduce the high current account deficit inthe balance o f payments led
to a period of economic slowdown during 1998-2000, in sharp contrast to the impressive
but unsustainable growth rates o f the mid-1990s. Although by 2001 the economy was on a
recovery pathwith a 3.8 percent realGDPgrowth, the inequalities ineconomic opportunities,
which had grown larger over the 1990s, showed no signs of declining.


  A. THE2002 ELECTIONS             LED WAVEOFFAR-REACHING
                                         TO A                                  REFORMS

2.2   Thus, the government that emerged from the September 2002 elections, as a
modification o f the previous coalition, inherited a country profoundly divided. For the
most part the population was relatively prosperous and well equipped to benefit from the
growing competitiveness o fthe Slovak economy andthe economic growththat came with it.
However, inthe aftermath o feconomic transformation, arelatively small group hadbecome
increasingly trapped in a vicious circle o f low skill, low employment opportunities, state
dependency, low intensity of job search. A crucial link in this chain was the disincentive
to work created by the generous social assistance system o f the 1990s, which led to some
household typologies (typically those with large numbers o f children and potentially low-
earning adults) beingfinancially better off when out o f work.

2.3   Inan attempt to break this vicious circle, the government undertook a set of politically
demanding and far-reaching structural reforms with the objective of fostering growth and
increasing average living standards by strengthening the flexibility and competitiveness of the
economy and increasing the incentives for individuals to work and acquire human capital. The
focus was on employment as the main pillar of a comprehensivegrowth and poverty reduction

 8                                                    The Quest for Equitable Growth in the Slovak Republic



 strategy. The economy reactedwell, with (i)            economic growth rising to well above 4 percent in
 2003 and reaching an impressive 5.5 percent in2004; and (ii) double-digit inflation of the
                                                                            the
 1999-2000 periods declininga8

 2.4     In2004the growthperformanceofthe SRcloselyapproachedthat ofthehigh-growthyears
 of the mid-1990s (see Table 2.1) and was more than double the average for the EU15 andEU25
 countries (growthratesof2.3 and2.4, respectively).Indeed, after asharpturnaroundin2001, over
 recentyears the growthrateofthe Slovak economyhasbeenconsistentlydoublethat ofthe EU15
 andwell inexcessofthatofthe Esegraa'$+ coun~es,'Oalthoughsomewhatslowerthanthatofthe
 Baltic States(see Figure2.1).

 Table 2.1: Main Macroeconomic Indicators, 1994-2004
                                          1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
 GDP real growth                           6.2     5.8   6.1    4.6    4.2   1.5   2.0 3.8   4.6  4.5  5.5
 Private consumption real growth           1.0     5.0    8.2    5.8    6.0  2.7  -0.9 4.9   5.5 -0.8  3.5
. Gross fixedinvestmentrealgrowth         -2.5     0.6  29.1    15.0  11.0 -19.6  -7.2 13.9 -0.6 -1.5  2.5
 Unemploymentrate, registered             14.4 13.7     12.6    12.9  13.8  17.5  18.2 18.2 17.8 15.2  14.3
 Unemploymentrate, LFS                    13.7    13.1  11.3    11.8  12.5  16.2  18.6 19.2 18.5 17.4  18.1
  GG expenditures, % o fGDP
  (ESA95)                                 57.8    54.1  61.5   65.0 60.8    56.9  59.9 51.5 50.9 39.2 38.5

  GG budget deficit, % o fGDP
  (ESA95)                                 -6.1    -0.9  -7.4    -6.2 . -3.8 -7.1 -12.3 -6.0 -5.7 -3.7  -3.3

  Real wage growth                         3.2     4.0    7.1    6.6    1.6 -3.0  -4.9  1.0  5.8 -2.0   2.5
  Consumer price inflation, yIy           13.4     9.9    5.8    6.1    6.7 10.6  12.0  7.1  3.3  8.5   7.5
 N B S refinancingrate (end-year)         12.0     9.8    8.8    8.8    8.8  8.8   8.8  8.8  6.5   6.0  4.0
  Current account balance, % ofGDP         4.9     2.6   -9.4   -8.6   -8.9 -4.8  -3.5 -8.4 -8.0 -0.9  -3.5
  Foreigndirect investment, % of
  GDP                                      1.8     1.7    1.8    1.7    2.6  2.1   8.7  5.7  7.2   2.8  1.8

  External debt, % of GDP                 23.5    21.1  30.6 33.1     34.0 47.2 49.9   48.7 43.3 42.6 43.6
  SKWUSD (average)                        32.0 29.7     30.7    33.6  35.2 41.4 46.2 48.3   45.3 36.8 32.3
 Source: Statistical Office o f SR, Labor Office, Ministry of Finance, NBS, and EUROSTAT.




     It is worth noting that the significant decline inheadline inflation after 2000 was largely a result
     of a cautious approach and ultimately a freeze inregulatedprices inthe pre-election period, hence
    the pick up in 2003. Thus the reduction attributable to earlier structural reforms is smaller than
     generally thought.
     Only the Baltic States hadhigher growth rates, with Irelandand Polandcoming very close with 5.4
     and5.3, respectively.
 lo  'The Esegrad 4+ countries are the Czech Republic, Hungary, Poland and Slovenia. The Baltic
    countries are Estonia, Latvia and Lithuania. Figure 2.1 separates the EU8 into these two groups of
    countries becausethe Baltic States have growth trends that differedconsiderably from the other 4
    countries, partly as a reflection ofthe different initial conditions.

Chapter 2                                                                                          9


              Figure2.1:InternationalComparisoninGDPTrends, 2000-04

                                +Slovakia                 +Visegrad  4+
                                -A-  Baltic countries    +EU15




                              2000        2001        2002       2003   2004


                                          Source:EUROSTAT.


2.5    Inaddition, ifthe reform process is not haltedor reversed, the growth rates ofthe SR
seem set to remain high in the years to come. This conclusion i s based on four important
differences between the "new" Slovak economy and its transition counterpart, namely: (i)
after 2000, structural reforms replaced over-investment as the main drivers o f growth; (ii)
following an overall reduction in government expenditure, greenfield investment" replaced
fiscal expansion as the main source o fgrowth indomestic demand; (iii)       the country continued
to integrate intothe globalproductionnetworksofEuropeanmultinationalcorporations andits
patternoftrade specialization shiftedaway from unskilled labor-intensive products into more
advanced productions; and (iv) privatization receipts, rather than international borrowing,
financed the country's widening current account deficit. All these factors suggest that (i)       the
economic climate i s now inplace for sustainable long-run growth, but (ii) the most radical
                                                                                as
reforms were only implementedin January 2004, it i s important to give the new economic
system enough time to realize its full potential and achieve a new high productivity-high
employment equilibrium.

2.6    The combination of growing GDP and declining inflation ledto an increase inaverage
private consumption- up nearly 3.5 percent in 2004 - and a renewal o f domestic demand,
which suggests a positive trend in the living standards o f the Slovak population. Thus, the
post-2002 reforms have clearly been very successful inpromoting growth and in improving
the average living standard of the population.

2.7    However, the concern i s that they may not have been equally successful in reducing
poverty and tackling social exclusion. On the contrary, the argument is that the emphasis on
the individual's work potential as the route out o f poverty may have Mher increased the
social exclusiono fthosewho for whatever reasonmaynotbe able to findor seek employment.
Inaddition, inthe highlyflexible andcompetitive economy that has emerged inthe SRover
the last year, the inability o f these groups to benefit from the high growth rate, the overall
improvementinliving conditions andthe increasednumber o f special targeted measuressuch
as activation programs and other labor market policies, may have increased their economic


    A firm enters a foreign market via foreign direct investment(FDI) either by setting up an entirely
    new plant(greenfield investment)or acquiring anexistingindigenousfirm.

10                                         The Questfor Equitable Growth in the Slovak Republic



vulnerability. This concerni s reinforcedby the findings that, the growth inGDPhas not been
hllyreflectedinmorefavorable labormarketoutcomesandthatunemployment isincreasingly
concentrated inparticular groups of the populationand inspecific regions.

2.8    Indeed, the high economic growth of recent years has not fully translated into
employment growth. As i s evident from Figure 2.2, the slowdown of 1999-2000 reduced
employment by 100,000 jobs between 1998 and 2000 (a drop of nearly 5 percent) despite
what, byEuropeanstandards, is aperfectly respectableGDPgrowth ofaround 2 percent.After
2000, the acceleration ofgrowth resulted ina mildbut much smaller increaseinemployment,
which only in 2003 was close to 2 percent (less than half of the GDP growth rate for that
year). Despite a5.5 percent GDP growth rate in2004, the annual increaseinemploymentwas
minimal.

                     Figure2.2: GrowthandEmployment, 1998-2004




                 2250                                                      6 0

                 2200                                                      4.0

                 2150                                                      2.0

                 2100                                                      0.0

                 2050                                                      -2.0

                 2000                                                      -4.0
                        '98     '99    '00     '01       '02     '03  '04


                                   Source:Authors' calculations.


2.9    Unfortunately, the mildincrease inemployment experienced after 2000 was more than
compensatedby the increase in the active population. This resulted from the combination o f
the following:

       0  A steady increase inthe working age population which was slight until 2003 but
          then reached 1.5percent in2004
          A slight increaseinthe activity rateo fthe working agepopulationin2003 which led
          to a growth inthe labor supply over the last five years.

Thus, the increase in employment experienced to date has been just enough to absorb the
increase inthe labor supply, and unemployment has remainedgrossly unchanged,


        B. THELISBON        TARGETS EMPLOYMENT
                                         FOR                        REMAIN  DISTANT

2.10 However, what i s really striking for the external observer i s not so much the recent
trends inlabor market indicatorsbutthe absolute leveltakenby some ofthese trends. Figure
2.3 shows the following:

Chapter 2                                                                                                  11


         At around 70percent, the participationrate ofthe working agepopulationinthe SR
         i s very similar to the average for the EU15 countries and considerably higher than
         the EU8 average o fjust over 66 percent.
     0   However, at less than 58 percent, the employment rate inthe SR i s the third lowest
         inEurope(after Hungary andPoland) andis far lower than the Lisbonobjective of
         70 percent.
     0   Much of the difference inemployment ratesbetweenthe SR and the rest of Europe
         i s linked to the very low employment rates o f Slovak workers with low skills,
         especially males. Working age males with only a basic education in the SR are
         employed at a rate that is less thanhalfthat o fthe EU15 (Figure 2.4).
     0   As a result of the average participation rates and the very low employment rates,
         the SRhas the secondhighest unemployment rate inthe EUafter Poland, more than
         twice the EU15 average. Inaddition, the unemployment rate among workers with
         only basic education i s nearly five times higher than in the EU15 and more than
         double the EU8 average.

   Figure2.3: InternationalComparisonsofActivity andParticipationRates,2004

     Participationrates,2004                 Emolovmentrates.2004            Uneploymentrates, 2004

                                                                                               I




                             lionrate,                    loymentrate,                            Ite, 20
                                                                                                    ~




                            ofworking-                                                               force
                      -age population
                                          T L I ,   I    I    I      I                            -u
   50 55 60 65 70 75 80 65                 50  55 60    65   70     75 80   0 2 4 6 8 10 12 14 16 16 20

                                             Source: EUROSTAT.

12                                           The Questfor Equitable Growth in the Slovak Republic


     Figure2.4: InternationalComparisons of EmploymentandUnemploymentby
                                       EducationLevel, 2004

     EmploymentRate by Education                            Unemployment Rate by Education

    on
    I"


                                                                               -
    80

    70                                                      45                 --
    60                   -25<age<64-                        40                         N15--
                                                                                       N 8 -
    50                                                      35                 -
                                                            30                         SK  -
    40                                                      25
    30               `"9-EU15,   female                     20
                     eSK,     female                        15
                     ***e. EU15, male                       10
     10              d.lSK,male                               5
                                                              0                            1
          BAS        SEC       TERT                                BAS      SEC      TERT

                                         Source: EUROSTAT.


    c. THEPOST-2002REFORMS                  WERE    DESIGNED BRIDGE THIS GAP
                                                                  TO


2.11 The considerable disadvantageo funskilledworkers inthe SR i snot anewphenomenon.
Onthe contrary, ithasbeenalatentfeatureofthe Slovakeconomy sincesocialist times, although
one that was somewhat hiddenby the emphasis on full employment of the pre-transition area.
The economic liberalization that followed the 1998 election brought this disadvantage to the
surface as the shift infocus from full employment to increasedproductivity resulted ina sharp
reduction inthe employment opportunities for low-skilled workers. Unemployment for these
workers increasedsharply andbecameincreasingly longterm. However, untilthe 2004reforms
the living standards of these workers were, for the most part, maintained by very generous
social benefits which for some household typologies exceededtheir potentiallabor income.

2.12 While successful in preventing households from falling into poverty, these generous
benefits substantially reduced the incentives for potentially low-paid unemployed workers
to actively seek employment, and many left the labor force altogether. The government that
emerged from the 2002 election set itself to addressing this issue by reversingthe incentive
structure and creating a substantial financial incentive to seek and find employment. I t did
this with an extensive, far-reaching, but also internally consistent, reform package that
focuses on fostering growth and improving the average living standard o f the population
by fostering job creation and making work pay. Thus, the reforms addressedthe need to (i)
introduce more flexibility into the labor market; (ii) the existing disincentive to work
                                                       reduce
andprovideanincentive for the long-termunemployedtojoin the labor market; (iii)         make the
taxation andregulatory frameworks more supportive o fnew enterprise creationandbusiness
sector development; and (iv) make the public education system more efficient, effective and
responsive to the needs o f the twenty-first century environment.

Chapter2                                                                                     13


2.13 The mainelements of the reform can be summarized as follows:


      0  The implementation o f a new Labor Code, designed to foster labor market
         flexibility by:

         o Introducing the option of a number ofnon-standard labor contracts
         o Increasing the employer's flexibility to terminate a contract
         o Redefiningthe rules for workers' representation.


      0  The reform of the welfare system, designed to increase incentives to work and
         improve the targeting o f social welfare benefits by:

         o Reducing the minimum benefit for a single adult to an amount below the
            subsistenceminimumand introducingaceiling for the maximumbenefitpayable
            to a family with more than 4 children
         o Introducinganumber ofnewbenefits for the working poor, available to families
            where the adults are either working or participating in active labor market
            programs (ALMPs) and expanding the entitlement to social benefits to the first 6
            months o f employment
         o Providing subsidies to municipalities (and other stakeholders) for organizing
            "activation programs"
         o Replacing the previous income-related family benefits with a flat (not income-
            related) benefit per each childcoupled with tax credits for minor children
         o Introducing basic and high-school scholarship schemes inaddition to meal and
            schooling subsidies.


      0  The labor market reform, designed to improve employment opportunities for
         disadvantaged groups andthe long-term unemployedandtheir potentialincome via
         the followingmeans:

         o Individual and targeted work with the unemployed-i.e.,           in-depth interviews,
            individual action plans for job-seekers, testing and assessment o f personal
            characteristics, etc.
         o The mandatory character ofthe ALM programs and the reimbursement of travel
            andchild-care coststojob-seekers activelyseeking employmentandparticipating
            inALM programs
         o Subsidies to employers hiring LTU and to LTUjob-seekers planning to start a
            business
         o Targeted interventionsto regions with the highest incidence ofunemployment
         o An increase inthe minimumwage o faround 5 percent inreal terms since 2003.


      0  Thereformofthe administrationofwelfare systems,designedto mergethe social
         agendawith the labor market policies inthe regional offices.

2.14 The rationale for the reforms i s clear and i s fully consistent withthe overallobjective of
increasing employmentopportunities and strengthening the incentivesto take up ajob for the

14                                         The Questfor Equitable Growth in the Slovak Republic



overall population. However, the concern of their critics i s that, by emphasizing individuals'
work potentials, the reformedwelfare systemhaspenalizedthe most vulnerable groups inthe
Slovak society who, for whatever reason, may not be able to seek andor find employment.
This is the case despite the fact that the reforms provide for special targeted measures for
those who may find it difficult to find ajob (i.e,, "activation o fjob-seekers" and other labor
market policies).


                        D. I s LISBON    GETTING      ANYCLOSER?

2.15 Figure 2.4, above, clearly emphasizes one o f the growing concerns emerging in
what i s otherwise a very brightpicture of development inthe SR: the fact that, despite the
attempts by the 2002 government to include them more actively inlabor market activity,
parts o f the Slovak society are still struggling to join in the drive for economic growth
and higher living standards. In 2004 over half o f the workers with only basic education
were unemployed (see Figure 2.5), an increase o f over 70 percent since 1998. The second
panel o f Figure2.5 also shows that in2004 theunskilledwere 2.5 times more likely to be
unemployedthan the average population, by far the second highest differentialamong the
EU25 after the Czech Republic.

2.16 Additional warning signs regardingthe welfare ofthe most vulnerable are given by the
following observations:


       0   Over the same period, long-term unemployment (LTU) had nearly doubled, and in
           2004 the SR was the EUcountry with the highest proportion of LTUand the smallest
           proportion of short-termunemployment.Inaddition, a clear linkbetweenlow skills and
           the occurrence ofLTUis emerging.Iz




l2 Unemployment is defined as long term if its duration exceeds 12 months and short term if its
   durationis shorter than 12months.

Chapter2                                                                                                  15


 Figure 2.5: Inequalitiesin Unemployment by EducationalLevel: SR and EU,2004

           Unemployment rates by                                          Ratioof low sklll -to-
                  education                                            total unemployment rate
                                                                  cz
                                                                  SK
        40                                                        EEa
                                                                  AT
        35&                                                       BLV
                                                                    E
                                                                   FI
                                                                  ES
                                                                  CIT
                                                                    Y
                                                                  HU
                                                                  SE
        20                                                        P15L
        15                                                        DK
                                                                  LT
                                                                   IE
                                                                  OR
                                                                  UK
                                                                  FR
                                                                  N L
                                                                  D E
         0     ,    ,    ,    ,   ,    ,    ,
           1998 1999 2000 20012002 2003 2004                      PSI
                                                                    T
                                                                      0.5    1.0   1.5      2.0    2.5



                                               Source:EUROSTAT.

                   Figure 2.6: UnemploymentRate and RegionalVariance

                                                                    I Rate
                                                                        Unemployment
                                                                    +Uarianca   Among Regions
                                                                                            l--~-
                 25




                                                                                              50.00
                 20


                                                                                              40.00

                  15


                                                                                              30.00


                  10

                                                                                              20.00



                  5
                                                                                              (0.00




                  0                                                                           0.00




                                               Source: EUROSTAT.


     0  Regional disparities are very large (see Figure 2.6 and Table 2.2). Table 2.2 shows
        that, usingtheNUTS3 level ofdisaggregation, the SRhasthe secondlargestvariance
        inregionalunemploymentrateintheEU25after Italy. Italsohasthe secondhighest
        regional unemployment rate at NUTS2 level in the EU25, after Italy, but it ranks
        only number 7 among the EU25 interms of the percentage difference between the
        region with the highest and the region with the lowest unemployment rate at the
        NUTS2 level.

16                                           TheQuestfor Equitable Growth in the Slovak Republic




                         Highest Regional  Lowest Regional                                 Across
                             UR (NUTSZ)        UR(NUTSZ)         Difference ("/.IVariance
                                                                                  Regions(NUTS3)




2.17 However, it is important to note that the fact that the Czech Republic is the only EU
country with similar levels ofdisadvantagefor the unskilledreinforces the conclusionthat this
is aproblemthat the SRhasinheritedfromitspast anddefinitely notthe resulto frecentreforms.
Indeed, to some extent the large disparities that emerge across regions and across skill levels
are likely to overlap with the existinginequality across ethnic groups. Unfortunately, owing
to data limitations, it i s currently impossible to disentangle these effects and to statistically
quantifythe extent to which ethnicity interacts with skill level and geographical location in
increasing the vulnerability of some individuals. The issue than becomes one o f whether or
not the reforms have succeededinreducing this disadvantage by increasing competitiveness
and incentives to work. Unfortunately, it i s still too early to arrive at firm conclusions on this
issue, but the first signs are encouraging, as described inthe next chapters.


           E. MONITORING I s CURRENTLYVERYLIMITED
                                   CAPACITY

2.18 The referenceto data limitations bringsthis introductory chapter to its last point. Within
the context of extensive and rapid change and deep socioeconomic inequalities, it comes as
something of a surprise that little systematic effort has been made to monitor the effects of

Chavter 2                                                                                     17


recent growth and structural reforms on the incidence and depth o f poverty in the SR. Data
limitations are behind the failure to produce regular poverty profiles that are reliable and
comparable over time, andthat canbeusedto evaluatethe impact ofthe reformeffort andthe
growth that it has generated.

2.19 In the absence o f regularly produced nationally representative data on income and
consumption, the 2003 Microcensus-with        datafor 2002-provides    the mostrecent sourceof
information on income and the only one since 1996. This prevents any analysis o f the trends
inthese variables over the lasttwo yearsandanadequate evaluation ofthe impact ofthe2003
and 2004 reforms. This i s inaddition to the fact that the 2003 data are not representative of
the Slovak population (see Chapter 5 for more details).

2.20 The government o f the SR, the Statistical Office o f the SR (SOSR), and the academic
community are fully aware o fthis and arebeginningto work together to addressthese issues.
Withinthis context, anduponthe requesto fboththe government andthe SOSR, the Technical
Note(Volume 11)describesthe limitationsofthe dataandsuggests ways inwhich the quality of
social statistics can be improved and their use for policy designand monitoring enhanced.13




 l3 More details are provided in the Technical Note, Building Capacityfor Systematic Monitoring of
    Poverty and Social Exclusion in the SlovakRepublic inVolume I1o fthis report.


                                   CHAPTER 3 -

 A TWO-EDGED SWORD: COMPETITIVENESS
       AND LABORMARKETDUALITYINTHE
                    "NEW" SLOVAK REPUBLIC


3.1   In the absence of national representative information on living standards which can
be compared over time, recent developments in labor market outcomes can be taken as an
indication o f the extent to which the "new" SR has been successful inadopting the Lisbon
agenda. Iti s important to point out that the SRenteredthe EUwith a substantial disadvantage
interms ofemployment rates compared to boththe EU15 andthe other EU8 countries, with'
the exceptionofPoland. This was the combinedresult of the following:


      0  Two periods o flarge declines inemployment: (i) recession o fthe early transition
                                                           the
         years of 1990-93; and (ii)the second round of restructuring that followed the
         macroeconomic crisis o f 1998
      0  The relatively slow response o f employment to recent highgrowth (see Figure 2.2
         inChapter 2)
      0  The growing labor force.

3.2   The SRwas also at adisadvantageinterms ofhavingarelatively unskilledlabor force, a
comparatively highincidence o funemployment among the unskilledandvery highdifferences
inunemployment ratesacross regions.All these indicators suggestaveryhighdegreeoflabor
market duality. This chapter analyzes these issues further anddiscusses recent trends.


        A. A GROWING         LABOR     FORCE    WITH RELATIVELY L O W SKILLS


3.3   Incomparison withother EUcountries, the Slovak labor force is relatively young and
unskilled.The averageage o flabor marketparticipantsi s around 37 years, compared to about
39.5 years inthe EU15 andabout 38 years inthe EU8.Moreover, only 13 percent o fthe labor
force has a tertiary education, compared to 20 percent for the EU15 and 16 percent for the
EU8.

20                                                        The Ouest for Eguitable Growth in the Slovak Republic


The WorkingAge PopulationI s Growing
3.4    The working age population (aged 15-64) has been growing consistently for the
past several years at an average annual rate of around 1 percent (see Figure 3.1, left panel),
reflecting a wave o fyounger workers enteringthe labor market. This growth has a two-sided
effect. On one hand it has put considerable pressure on the unemployment rate, but on the
other hand it has kept adding new skills to the labor force, as newcomers to the labor market
had, on average, better education than their elder counterparts.

Education Levels Remain Relatively Low
3.5    However, comparedto other EUcountries the educationleveloftheworking population
remains relatively low, with the second lowest share of university graduates inthe EU (10.5
percent o f the population and 13 percent of the labor force). Perhaps surprisingly,however,
theproportionofthepopulation(or the laborforce) with only aprimary education isrelatively
small by EUstandards (see Figure3.1, right panel), with a relatively larger proportion o f the
population having attended-but                    oftennot completed-secondary       school.    l4


      Figure 3.1: Trends in the Working Age Population and Its Skill Composition,
                                                        SR, 1998-2004
                                             --
         Growth In working age population                                   Population, skill compositlon 2004
         -Working     age populallon,mn,left
         &Growthofworking     age population.%yiy,right
     3.80 I__--._--                                 3.0%


     3.75                                           2S0A

     3.70                                           2.0%

     3.65                                           1.5%

     3.60                                           1.0%

     3.55                                           0.5%


     3.50                                           0.0%
           98      99 00    01    02   03    04
                                                                         0%     20%     40%     60%     80%   K10%


                Source: EUROSTAT and Slovak Labor Force Survey, authors' calculations.

3.6    Fortunately, however, the trends are encouraging, with (i) share o f individuals with
                                                                                  the
only aprimary education falling from 14 percent to 8.5 percent between 1998 and 2004, and
(ii) proportionofthosehavingcompletedtertiaryeducationincreasingfrom11percentto
    the
13percent (as presentedinFigure 3.2). As is evident fromFigure 3.3, the mainsource ofthese
changes was the general upgrading o f skills in the population with an increase inthe share
of Slovaks with a tertiary education, which more than offset the falling participation rates o f
those with a higher education.



l4 Itis, however, important to point put that around half o f those classified as having an education
   higherthanbasic (ISCED levelbelow 3) haveonly so-called "incomplete secondary" qualifications
   andor vocational education.

Chapter 3                                                                                                                  21


             Figure 3.2: Trends in the Skill Mix of the Population, Labor Force
                                                 and Employment, 1998-2004

       Share of basic education in:                                          Share of tertiary educationin:
     30-                   l C I P o p u l a t i o n  -
                           -Labor            fo rce
                            -0 -Employment            -




                                                                                                 -Po       pulatio n
                                                                                                 -Lab0       r fo rce
                                                                                                 -&-~Ernploymenl


      o                                                h
          '98  '99  '00   '01     '02      '03      '04                      '98    '99    '00     '01   '02    '03   '04



              Source:EUROSTATand Slovak Labor Force Survey, authors' calculations.

Figure 3.3: Skill Composition of the Labor Force and Sourcesof I t s Change, 1991-2004

                                                                        -
       Skill composition of labor force, 2004                          Decompositionof changes in high-
                                                                           skill share of the labor force

                                                                       0dropoutofotherskills,%p
                                                                        -participation       growth of high-skilled, %p
                                                                                skill upgrade of population, %p
                                                                        101growthinhigh-skillshareinlaborforce,%p
                                                                  1.2%
                                                                  1.0%
                                                                  0.8%
                                                                  0.6%
                                                                  0.4%

                                                                  0.2%
                                                                  0.0%
                                                                 -0.2%

                                                                 -0.4%
                                                                 -0.6%


     0%      20%     40%         60%             80%     100%


                  Source: EUROSTATandSlovak LaborForceSurvey, authors' calculations.

Activity Rates Are Relatively High
3.7    The economic activity of the Slovak population compares favorably with that o f its
European peers. In2004 the total participation rate for the working age population was 70
percent, higher than the EU8 average of 68 percent and exactly at the average EU15 mark.
Moreover, owing to the relatively highenrollment o f young Slovaks insecondary education,
the relative advantage of the SR becomes more evident when only the 25-64 age bracket is
considered. Then the total participation rate inthe SR rises to 78.9 percent, higher than the
EU15 average of 75.4 percent and that o f the EU8 (76.9 percent).

3.8    The main factor behind the relatively high activity rates i s the exceptionally high
economic activity o f Slovak women (Figure 3.4). For the 25-64 age bracket, the total female

22                                                   The Questfor Equitable Growth in the Slovak Republic


  ::c==
participationrate standsat 71percent compared to 66 percent inthe EU15, andparticipation
rates are particularly high in the ages between 30 and 54, after which economic activity
plummets,partly becauseo f their lower retirement age (see Figure 3.4).

   Figure3.4: TrendsinEconomicActivity by Gender, the SR andthe EU,15cagec64

  7 2 T T o t a l  7                     67 T--Women
                                         65

  70
  68
       ~




                      --     EU25
                                         59                  -   - E U %
  62                                     57                  -EU8

  60     '98  '99 '00 '01 '02 '03 '04    55                                     7 20
                                             '98  '99 '00 '01 '02 '03  '04           '98  '99'00 '01'02 '03 '04


                      Source: EUROSTAT and Slovak Labor Force Survey, authors' calculations.

But a Large Proportionof the UnskilledAre Excluded
3.9        However, theunskilled,especially men,donot seemtopartakeinthishighparticipation.
As Figure3.5 shows, onaverage the participationrate o fpersonswith secondary and tertiary
educations i s slightly higher than the EU average. However, the picture for the unskilled i s
totally different, with only a little more than 50 percent participatinginthe labor market. The
situation i s particularly striking for unskilled men, nearly 40 percent o f whom are neither
employednor looking for ajob, compared tojust over 20 percent inthe EU15.

     Figure3.5: EconomicActivityby Education, the SR andthe EU,2004,25<age<64


                                         x)O         Women                    100
  x)o   l - - - - - - T o t a l I    i   90                                   90

                                         80                                   80

                                         70                                   70

                                         60                                                     --
                                                                              60                       EU15
                                                                                                -EU8
                                         50                                   50

  40                                     40                                   40
            B A S      SEC      TERT            B A S     SEC       TERT              B A S     SEC     TERT


                      Source:EUROSTAT and Slovak Labor Force Survey, authors' calculations.


      B. EMPLOYMENTGROWN            HAS                RELATIVELY         SLOWLYAND UNEVENLY

3.10 Thegrowthinthelaborforce overrecentyearshasnotbeenaccompaniedbyanequivalent
increase inemployment. Indeed, as shown inFigure3.6, the two rounds o f restructuring that
occurred at the beginning of transition and then after 1998 resulted in a sharp reduction in

Chapter 3                                                                              23



employment. Despite a continuouslypositive growthrate inGDP, total employmentbottomed
out in2000 at closeto the 1994level ('just under2.1 million) andnearly 15percentlower than
in 1990.Since 2000, employmenthas begunto grow but far more slowly than GDP growth
rate.

                    Figure3.6: Growth and Employment, 1990-2004



                            I          eGDPgrowth,growth,%
                                             Employmentlevel,'000,axislef?
                                                                   nghtaxis
                                             Employment %y/y, left y/y,   axis
                                                                              6.D

                        2200                                                   4'0

                        2150                                                   1.o

                        2100                                                   0,o

                        2050                                                   -2,0

                        MOO                                                    -4$0
                   I         90 '93 94'95 '96'97 'B199'00"01 '02"03'04




~




 l5 Only Polandand Hungary have lower employment rates.

24                                                     The Quest.for Equitable Growth in the Slovak Republic



         Figure3.7: InternationalComparisonsofResponsivenessof Employment
                                               to GDP Growth

        EmploymentGrowth/CDP Growth

   1 00

   0.80 4
   0.60

   0.40

   0.20

   0.00

  6.20

  -0.40

  6.60

  -0.80

  -1.00

                                     I-sR         *ELI8     -~-EalticCauntries   *EU15    1
                                                -
                                 Source: Authors' calculations based on EUROSTAT.


                      Figure3.8: Trends inthe EmploymentRate, 1998-2004


         Bnployment rates, 2004              MO -Trends In employment-                    `do   Trends In employment-
                                             95 - ratebysge                   -            95 irate byeducatlon 2 5 < a w 6 4
                                                a
                                             90 __                    -1s-64               90
                                             85 --25.64 -0-15-24      -0-25-49             85
                                                                                           80

                                             75                                            75
                                                                                           70

                                             65
                                             70                                            65
                                                                                           60

                                             55                                            55
                                                                                           50
                                             50                                            45
                                                                                           40
                                                                                           35
                                                                                           30
                                                                                           25
                                                                                           20
                                             2 0 ,    ,   .    .    .   ,       ,              1998 19992000 2001200220032004
       50 55 60 65 70          75                                            ,
                                   80            1998 19992000 2OOt2OO220032004



                    Source:EUROSTAT and Slovak Labor Force Survey, authors' calculations.

3.12 This i s due to the increase inthe working age population, described above. Figure 3.9
decomposesannual changes inthe employment rate into two main components: the change in
employment and the change inpopulation. Over the whole period, the growing working age
population continued to put pressure on the employment rate. In 1998 and 1999 a fall intotal
employment reinforcedthis effect, resultinginanoverallreductioninthe employment rate.By
way of contrast, in2000 and 2001 the growth in employment was barely sufficient to offset
the effect ofthe populationgrowth. Inthe exceptional year of 2003, strongjob creation inthe
economy boosted the employment rate. The trend in2004 was for the most part drivenby the
increase inthe working age populationwith aconsequentreduction inthe employment rate.

Chapter 3                                                                                        25


Large Inequalities ExistinEmployment Opportunities
3.13 The average trends in employment hide significant disparities in the magnitude and
direction o f the changes experienced by different groups of workers. The post-1999 period
has seen a significant increase in the employment of older workers at the expense o f the
youth. While this trend i s in accordance with the Lisbon targets, it i s a concern in a country
with arelatively young labor force (see Volume I1for more details) andrelatively highyouth
unemployment.

3.14 Similarly, o f concern is the sharp decline in the employment o f unskilled workers
-especially males.Thedecomposition ofemployment growthintohigh/medium/low skill
components, presented inthe left panelof Figure 3.9, shows this trend clearly. The shedding
o f low skilled workers explains most of the decline in employment experienced over the
post-recession years to 2000 and employment opportunities for the low skilled continued to
decline even insubsequentyears despite the overall increase inemployment. Incontrast, the
employment o f university educatedpeople grew consistently after 1999.

               Figure3.9: DecompositionofTrends inthe EmploymentRate
                               andthe EmploymentLevel, 1999-2004

               Decomposition of employment                    Decomposition of employment
                rate growth % points, (15<y<64)
         1.0

         0.5

         0.o               -
        -0.5                                   R C

        -1.0

        -1.5


                        0populationgrowtheffect
                        -employment   growth effect -
                       &employment    rate increase                   *TOTAL     em pl.growth
        -3.0                                            -80
              1999  2000   2001  2002  2003     2004        1999 2000 2001 2002 2003 2004


                  Source:EUROSTAT and Slovak Labor Force Survey, authors' calculations.



3.15 As a result o f the substantial shift in employment opportunities for different skill
groups over the 1998-2004 period, average employment rates for unskilled workers
declined by over 10 percentage points or nearly 30 percent. By 2004 the employment rate
for Slovak male workers with only a basic education was only 30 percent, less that half o f
the EU15 average. Unskilledfemale workers scored slightly better comparedto their EU15
counterparts. At around 25 percent, their employment rate was well over half o f that o f
EU15 women.

26                                              The Questfor Equitable Growth in the SlovakRepublic


                 C. UNEMPLOYMENT               HAS          REMAINEDUNCHANGED

3.16 Inthe SR significant discrepancies exist betweenthe registered unemploymentfigures
and those reported by the LFS. Chapter 3 inPolicy Note IinVolume I1discusses this issue
to some extent. This Policy Note focuses on the unemployment figures that emerge from
the LFS.In2004 the LFS gave an unemployment rate of 18.3 percent, more than double the
EU15 rate (Figure 3.10, left-handpanel).

                 Figure3.10: Trends inUnemploymentRates, 1998-2004

        UneDlovment rates. 2004    50     Unemploymnt rate by age

                                                             -15-64
                                   45  --25-64~
                                           w - 0 ~ 1 5 . 2 4 -25-49     -
                                           -50-64
                                   40


                                   35  Q"                                       35
                                             I
                                   30  -7                                       30

                                   25                                           25

                                   2150                                         20

                                                                                15

                                   10                                           10  1
                                    5                                            5


      0 2 4 6 8 101214161820        O     L   .    .       .   .   .   ,   *     0-
                                        1998 1999 2000 20012002 2003 2004            1998 1999 2000 20012002 2003 2004




                Source:EUROSTAT and Slovak Labor Force Survey, authors' calculations.


3.17 After a sharp increase inthe early 1990s, the unemployment rate remainedrelatively
flat at around 12 percent until 1998, only to increase to almost 20 percent after the 1998
crisis and to ease slightly only after 2002. These trends are not surprising given those in
the labor force and employment. Figure 3.11, which decomposes the 1999-2004changes in
unemployment rate into their main determinants, points to the growth in the working age
population as a major factor behind the increase in the unemployment rate (demographic
effect). This pressure was particularly strong in 1999,2000 and 2004. Growth ineconomic
activity (participation effect) worked inthe same direction, with the exception o f 2002 and
2004, when the average activity rate fell. Finally, in 1999-2000, employment contraction
was a main cause o f the increase in the unemployment rates, while in the subsequent
period o f 2001-04 growth in employment contributed to the unemployment reduction.
Nonetheless, in2001 and 2004 employment growth, although positive, was insufficient to
offset demographic and participation effects.

Unsurprisingly,UnemploymentI s IncreasinglyConcentratedamongtheUnskilled
3.18 The right-hand panel o f Figure 3.11 shows differences in trends in unemployment
across different educational groups. From the discussion it i s somewhat to be expected
that unemployment may have increased more among the unskilledthan among the rest of

Chapter 3                                                                                               27


the labor force. However, the extent of the differential is striking. The unemployment rate
for the unskilled nearly doubled over the last six years, while that o f workers with higher
education remained grossly unchanged. In2004 over 50 percent o f the unskilledlabor force
was unemployed, compared to the average EU15 and EU8 rates for this population of 12
percent and 23 percent, respectively. The unskilled unemployment rate in 2004 was almost
three times higher than the rate for the entire population, compared to a value o f this ratio of
1.4and 2 inthe EU15 and EU8, respectively.

          Figure3.11:DecompositionofUnemploymentRate Changes, 1999-2004

     Unemployment rate growth                                         Decomposition of employment
                      ezzzzzaparticipationgrowth effect         .^I       growth, thous. people
                      O p o p u l a t i o n growth effect
     s. 3             -employment            decline effect




                                                                ! O H CII


                                                                                    TERT empl growth
                                                                                    SEC empl,growth
                                                                              O B A S e m p l growth
      5-2 I                                                                   &TOTALempl,growth
            1999 2000 2001 2002 2003 2004                      -80
                                                                    1999 2000 2001 2002 2003 2004

                 Source: EUROSTATand Slovak LaborForceSurvey, authors' calculations.



            D.BUTTHE              UNEMPLOYED ARE INCREASINGLY                   EXCLUDED

3.19 O fparticular concernamongrecent trends is the growthinthe share o funemployment
that i s long term, as the long-term unemployed are far less likely than others to find
employment and are at aparticularly highrisk o fbecoming excluded from the labor market
and the society at large. It i s therefore worrisome to find that inthe SR nearly 65 percent
o f the unemployed are unemployed for more than a year, and that nearly 12 percent o f the
labor force has been unemployed for at least that lengtho f time-a                share that i s double the
average for the EU8 (6 percent) and nearly four times that for the EU15 (3.5 percent).16
Thus, unsurprisingly,the average unemployment duration inthe SR in2004 (27 months)
was well above that o f the EU15 and the EU8 (17 and 20 months, respectively).


~




 l6 To some extent the high incidence o f LTUcanbe related to the relatively high participation rates of
    the Slovak labor force. The argument here would be that Slovak workers have a higher propensity
    to remain unemployed than in other countries where workers that have been unemployed for
    any length o f time would leave the labor market thanks to workers generous disability or early
    retirement schemes. While this is indeed a possibility, the relatively low participation rates of
    unskilledmales-exactly the category of workers with the highest incidence of LTU-in the SR
    would argue against this hypothesis.

28                                           The Questfor Equitable Growth in the Slovak Republic



3.20 Moreover, the long-term unemployment (LTU) pool i s increasingly stagnant and
unresponsive to economic recovery. Figure 3.12 shows increasing at an almost constant
rate until2002, whenitreachedthe 12percent markandstabilizedaroundthat value. Thus,
while short-term unemployment (STU) shows an obviously cyclical nature, LTU seems
to grow at times o f increasing unemployment but does not decline when employment
opportunities improve. This conclusion i s further strengthened by the analysis of labor
market dynamics that follows which supports the hypothesis o f a strong "hysterisis"
effect with a highdegree o f labor market duality.

                        Figure3.12: STU and LTUCycles, 1998-2004
                                   ST & LTunemploymentrate,
                                  LTUis stubborn, STUis cyclical


                                   ST & LTunemploymentrate,


                              12

                              11

                              10

                              9

                              8

                              I

                              6

                              5
                                 1998  1999 2000 2001 2002 2003 2004



                Source:EUROSTATand Slovak LaborForceSurvey, authors' calculations.



               E.ANDEMPLOYMENT               MOBILITY        REMAINS      LIMITED

3.21 Inaddition, the Slovak labor market has traditionally been characterized by very low
mobilityinterms ofjobturnover, spatialmigrationand/or changesinthe labormarket statuses.
Figure 3.13 shows the persistence o f geographical disparities over the period 1998-2004 by
plotting the district levelunemploymentrates in 1998 versus those in2004.At 66 percent, the
correlationbetween these rates i s very highandsuggeststhe differencebetweenregions to be
long-lasting and of a structural nature.I7




l7 There was also little change in the coefficient of variation of district level unemployment rates
   duringthat period.

Chapter 3                                                                                   29


            Figure3.13: UnemploymentRates,DistrictLevel, 2004 versus 1998
                          1          -r)88.2004     unemployment correlation

                               4 0 3 -45.degree     line
                                                                            I

                               3 5

                               30

                               25

                               20

                               15

                               XI

                                5


                                   0    5     x)  15    20 25    30     3 5 40




                 Source:EUROSTAT and Slovak Labor Force Survey, authors' calculations.

3.22 Mobility betweenjobs and between employment statuses also remains very limited.
The transition matricespresented inPolicyNote 1o fVolume I1suggest, for example, that on
average only 2.7 percent of workers lose their jobs and only 21 percent of the unemployed
find a newjob within a year. It is perhaps more worrying that nearly one-fifth of the exits
from unemployment are out o f the labor force (20 percent), and only 3.2 percent o f the
inactive find ajob within a year (with the majority being school-leavers). The size of these
flows is nearly half o f that experienced in the EU, suggestingthat employment dynamics
are very limited.

FlowsInto and Out of Employment: TheTale of Education
3.23 I t i s worth noting that the single most important correlate of the direction and
extent o f the flows into and out o f employment over the last few years is the individual's
level o f education. A typical worker with a university education i s unlikely to ever leave
employment. The worker has less than a 2 percent probability o fbecomingjobless within a
year, and shouldthis worker become unemployed, he/she finds work relatively rapidly: over
one-third of the high-skilled unemployed find a newjob within a year.

3.24 Byway ofcontrast, the typical low-skilledworker hasamuchhigher probabilityoffalling
intounemployment andhasonly a9.6 percent chanceoffindingajob withinayear. The worker
i s then likely to become discouraged and to move into inactivity, rarely rehuning to the labor
market. This is insharp contrastwith the experienceo fthe rest of the EUwhere the association
between skill levels and flow ratestends to be weak.

30                                             The Questfor Equitable Growth in the Slovak Republic


                     F. BUTJOBDYNAMICS RELATIVELY    ARE                     HIGH

3.25 Data from the enterprise survey "Information SystemonAverage Earnings" (ISAE)
show that a substantial job restructuring has occurred in recent years.18 Job flows are
significantly higherthan the average inthe euro zone and more on the order o f magnitude
experienced in other Central European transition countries. This i s the case despite
the fact that the sample design makes it impossible to calculate the total job flows in
the economy. It gives an indication o f job creation and destruction only in firms that
remain in the sample for the full period of the analysis, and i s not able to assess the full
extension o f net job creation because new firms entering the market (and the sample)
cannot be adequately analyzed. This is an important limitation that i s likely to result ina
considerable underestimation o f the netjob creation incountries like the SR which have
experienced considerable growth and positive inflows o f foreign companies over recent
years. However, this i s a common problem with this type o f studies and one that cannot
be easily overcome.

3.26 The averagejob creationrate inthe sample o ffirms covering the 2000-04 period i s 7.2
percent and the average rate o f destruction is 10.9 percent, which leads to an 18.1 percent
average job turnover rate and an excess job reallocation rate of 14.4 percent. The net job
creation is negative at an average o f 3.7 percent, and individual yearly rates vary from -3.3
percent in2001to -4.1percent in2003.19This overall negative trendseems somewhat at odds
with the positivetrends inemploymentas evident fromTable 3.1, Two mainjustifications can
beadvocated for what appearsto bea contradiction:

        0  The fact that the enterprise-level analysis focuses on"continuing firms" andignores
           theeffectso fenterprisecreation, which, inthecontext ofthe SRinrecentyears, could
           be substantialand is probably inexcess of enterprise destruction. By comparingthe
           average net job creation (-3.7 percent) with the average change in the number of



    ISAE i s a part o f the Information System on Labor Costs that is collected by a private firm for the
    MinistryofLabor, SocialAffairs andFamily ofthe SR.The databasecontains detailedinformation
    on enterprises and their workers. The data are collected on the worker level on a quarterly basis
    by software integrated into a respondent enterprise informationsystem. However, it is not possible
    to track a particular employee over time, which prevents us from building a panel. The sampling
   methodology used is in no way perfect (for more information, see Policy Note 1, Volume 11).To
    correct for the sampling biases, we have restrictedthe datato enterprises with at least 20 employees.
    The data were cleaned from outliers, and records with missing observations were excluded. The
    samples contain two sets o f variables, one describing the enterprise and the other describing the
    employee. For the enterprise, the number o f employees, legal form, branch o f economic activity
    (coded according to NACE), geographical location and ownership type are reported. Ownership
    types are divided into private, cooperative, foreign, international private and public, state owned,
    and municipal. For the employee, average hourly wage, gender, age, education, occupational
   classification (in the national extended version o f the International Standard Classification o f
   Occupations) and some other characteristics are reported. Inprinciple, the data cover all industrial
   sectors and occupational categories except for the self-employed.
I 9 See Policy Note No. 1, Volume 11, for more details.

Chapter 3                                                                                       31


          employed persons (-0.4), we can quantify the underestimation ofjob creationinour
          analysis to be approximately 10percent.
       0 .The recent marked change inthe employment structure over the period which has
          seen a reduction in the number of employees but a sharp rise in the number of
          self-employed persons (22 percent in 2004). As the self-employed and the small
          enterprises are excluded from the analysis, thepositivetrends inthis dynamic sector
          are not pickedup inthejob flows described here, while the reduction inthe number
          of employees is.

Table 3.1: Numberof EmployedPersonsinthe Slovak Republic,2000-04
(in thousandsofpersons)
                                   Employedpersons                                  Employment

                                                                           employed
                                                                            %        rates, 15-64
                  Change
                    %       Employees       Change
                                              %       Self-employed Change
                                                                     %                  (%I
 2000     2,101.7                 1,931.0                      167.4             8.0           55.9
 2001     2,123.7      1.o        1,943.4         0.6          177.9    6.3      8.4           56.3
 2002     2,127.0      0.2        1,940.9        -0.1          183.1    2.9      8.6           56.0
 2003     2,164.6      1.8        1,947.6         0.3         210.9    15.2      9.7           56.5
 2004     2,170.4      0.3        1,904.2        -2.2          256.8   21.8     11.8
 2000-04 2,137.5       0.8        1,933.4        -0.4          199.2   11.6      9.3
Source:Labor Force Survey, Statistical Office of the SR

                                         JOBFLOWSREGIONS
                                                       BY

3.27 According to Table 3.2, large regional disparities injob flows have existed. The most
dynamic regions over the period 2000-04 were Zilina andBratislava with ajob turnover rate
that was 19 percent higher than the sample average, and the most stagnant was Nitra, with
ajob turnover rate that was 14 percent below the sample average and ajob creation rate (5
percent) that was only half of the job destruction rate.2oThe driving force behind regional
differences injob flows i s the job creation rate. The job creationrate inBratislava, the most
successful region, was twice as many newjobs (11percent versus 5 percent) as inNitra, the
least successful region. However, once the three most successful regions (Bratislava, Zilina
and Trencin) are excluded, the differences across the other regions becomes minimal.




*O  See Policy Note No. 1, Volume 11, for more details.

32                                            The Questfor Equitable Growth in the Slovak Republic


Table 3.2: Job Flow Rates by Regions, average over 2000-04 (%)

 Region           Job creation Job destruction Netjob creation   Job turnover      Excess job
                                                                                   reallocation
 Average SR
 2000-04                   7.2            10.9             -3.7           18.1                14.4

                           100             100             100             100                 100
 Bratislava              147.6           100.4            192.5          119.2                131.1
 Trencin                 111.2            97.2            130.3          102.8                96.5
 BanskaBystrica           94.1            86.8            127.7           89.7                77.6
 Zilina                  119.4           120.0             78.9          119.8                111.9
 Tmava                    75.8            92.0             76.2           85.6                67.7
Nitra                     63.5            91.6             53.0           80.4                63.5
 Presov                   77.8           102.2             49.6           92.5                77.8
 Kosice                   76.8           103.1             45.3           92.6                66.3
Note: The average over 2000-04 is computed for each region and the average over 2000-04 in the
sample considered is taken as a base (100 percent). Results are presented as a percentage change,
relative to the definedbase.
Source: Authors' calculations,

Job Flows by Industries
3.28 Job turnover also varies substantially by industry (59 percent of the sample average in
utilities versus 47 percent above the sample average inwholesale andretail trade), butthe net
employment change emerging from the sample is negative inthe majority of sectors (Table
3.3). Sectors with the highest rates of restructuring, - measured in terms of the excess job
reallocation rate above the sample average-include construction(45 percent) and retail (33
percent). The highestjob creationrates were found inretail (106 percent above the average)
followed by bankingandinsurance, andconstruction. Conversely, majorjob destruction took
place inmining(76 percent above the average) and the agriculture sectors (50 percent above
the average), the two sectors with the highestamount ofjob closures.

Chapter 3                                                                                                     33


Table 3.3: Job Flowsby Industries,Average over 2000-04 (in %)

 Industry                            Job            Job                                           Excessjob
                                   creation     destruction     creation
                                                                Net job        Job turnover       reallocation
 Average SR 2000-04                       7.2           10.9            -3.7              18.1               14.4
                                       100,o           100.0          100.0              100.0              100.0
 Agriculture, Forestry, Fishing        107.2           149.7          -33.4              132.7               87.9
 Mining, Metallurgy                     25.6           176.1         -272.4              116.1               25.6
 Manufacturing                           88.0           89.6          107.1               89.0               87.0
 Utilities                              68.1            52.9          177.0               59.0               30.6
 Construction                          158.1           127.9          131.6              139.9              145.3
 wholesale, RetailTrade               205.9            107.4          286.2              146.7              132.6
 Hotels and Restaurants                149.4           121.8          132.5              132.8               63.5
 Transport, Communications             106.9           124.4            41.2             117.4               86.5
 Banking and Insurance                 158.1            75.8          286.0              108.6               84.7
 Business Services*                    143.8            95.3          200.2              114.6               79.0
 PublicAdministration                  112.8            85.9          167.1               96.5               51.6
 Education                               80.4           76.5          131.2               78.1               79.2
 Health and Social Work                  70.5           57.1          169.4               62.4               38.8
 Other. Personal Services              140.4            98.2          185.0              115.0              130.8
Note:The average over 2000-04 iscomputedfor eachregionandthe averageover 2000-04 inthe sampleconsidered
i s taken as a base (100 percent). Results are presentedas percentages, relative to the definedbase.
Source:Authors' calculations.


3.29 Inaddition, two growing sectors emerge from this analysis21:


         0   The construction industry, with a fast growing job creation rate (from 5 percent in
             2001 to 15 percent in2004)
         0   Thebankingsector, with (i)         anaveragejob creationrateo f 13percent over theperiod
             2000-04, despite a decline of 6 percent in2004, and (ii)the lowest job destruction
             rate o f all sectors (an average of 4 percent if2001 i s excluded).22

Job Flowsby Enterprise Size
3.30 I t may be somewhat surprisingthat over the period 2000-04 the enterprises with 20-49
employees (small enterprises) seemto be the most dynamic enterprises, with a 26 percentjob
turnover compared to a job turnover o f 15 percent for the largest establishments. However,
this highjob turnover is induced by highjob destruction rates in the sample considered,
which decline with the increasing number o f workers: from 22 percent in small enterprises
to 7 percent in extra large companies. Job creation rates vary less, with 4 percent in small
enterprises compared to 8.5 percent in larger establishments. Thus, only in the largest
enterprises i s net employment change positive and approaching 1 percent. For the other



21    See Policy Note No. 1, Volume 11, for more details.
22   The year 2001 i s not representative because inthe period 2000-01 four of the Slovak banks went
     intobankruptcy.

34                                                The Questfor Equitable Growth in the Slovak Republic



categories, net employment change is always negative and rises uniformly, peaking at -18
percent for the smallest enterprise^.^^

Job Flows by Ownership Categories
3.31 A significant difference intrends emerges between state-owned and domestic private
enterprises on one side and foreign enterprises on the other. The former experienced a
substantialgapbetweenjob creation(onaverageat 6percent) andjob destruction (at 11percent
and 12 percent, respectively), which led in both cases to negative net job creation growth.
Consequently, their job turnover and excess job reallocation rates are similar inmagnitude.
However, the dynamics were significantly higher among foreign owned enterpriseswhere the
averagejob turnover rate reached 27 percent. At 15 percent the job creation rate was more
than twice that indomestic enterprises (both state and private), leading to a positive netjob
creation growth of 2 percent.

3.32 The dynamics of the contribution of differentownership categories to job creation and
destruction are presented inFigure 3.14. Inparticular, the following canbe observed:


      0  The developments inforeign enterprisesare different fromthe other two categories:
         job creation and destruction shares were almost constant during the fist three
         periods, with the job creation share always well above the job destruction share. In
         the last period, the job creation share rose significantly, while the job destruction
         share experienced a decline.
      0  The employment share inforeign enterprises i s risingconstantly, and is accompanied
         byauniformdecreaseintheemployment sharefor state-ownedenterprises. The share
         o fprivate enterprises inthe sample stays almost constant over the periodanalyzed.

  Figure3.14: Job Creation, Job DestructionandEmployment Shares by Ownership
                                         Categories in2000-04

    a) Job CreationShare                    b) Job DestructionShare         c) Employment Share
             (inYO)                                    (inYO)                      (in%)




   "I
   0                               0'                                   0-




                                         Source: Authors' calculations.




   For detailedresults, see Policy Note No.1inVolume 11.

Chapter 3                                                                                                     35


Job Flows by EducationalLevel of the Labor Force

3.33 Finally, job turnover i s highest inenterprises with more than 60 percent o f educated
workers (5 percent above the sample average). As Table 3.4 shows, this difference i s driven
predominantly by the relatively highrate o fjob creation (32 percent above the average) in
this category which more than compensates for the relatively lowjob destruction. Indeed,
this is the only category o f enterprises where the net job creation is zero as opposed to
negative.

Table 3.4: Differences in Job Flows by Share of Skilled Workers, Average over 2000-04
(in YO)

 Skilled Workers (in YO) Job Creation         Destruction
                                                       Job                                            Excess Job
                                                                 Creation
                                                                  Net Job      Job Turnover         Reallocation
 Average SR 2000-04                    7.2             10.9            -3.7               18.1              14.4
                                     100.0           100.0            100.0              100.0             100.0
 L o w [0,30]                         98.6           102.0            91.2               100.6              98.6
 Medium [30,60]                       86.7           104.1            61.7                97.2              86.7
 High[60,100]                        131.6             87.2          200.2               104.9             109.9
Note: Skilled workers are those with completed secondary and tertiary education. The percentage share of the
skilled labor force is computed inevery enterprise and enterprises are then categorized into three groups.
The average over 2000-04 is computed for each group and the average over 2000-04 inthe sample considered i s
taken as a base (100 percent). Results are presented as percentages, relative to the defined base.
Source:Authors' calculations.

Job Flows by Gender Decomposition of the Labor Force
3.34 Job turnover rates are significantly higher in enterprises with a higher fraction o f
employed females (20 percent compared to 16 percent); this is due to the much higherjob
creationinthese enterprises (10 percent as compared to less than 5 percent).

What Are the M a i n Sources of Differences inJob Reallocation?
3.35 Over the last years, clear differences in the patterns o f job reallocation are
found across different groups o f enterprises. These differences can be summarized in
employment shifts that occur within and between groups. Empirical evidence suggests
that a misallocation o f resources in the economy implies that employment shifts occur
mainly between groups - from the declining groups to the growing groups.24However,
job reallocation in the SR seems to be accounted for predominantly by within group
reallocation according to a pattern similar to that found in other Central European
transition countries (Table 3.5).25




24   According to BilsenandKonings (1998), at the beginning of the transitionprocess inCEEcountries,
     thejob reallocation was identifiedmainly betweengroups; later on itwas identifiedprimarily within
     groups. For more details on the methodology used, see Policy NoteNo. 1, Volume 11.

36                                           The Quest for Eauitable Growth in the Slovak Republic


Table 3.5: The Share of Excess Job Reallocation Due to Employment Shifts within
Groups (in YO)
                               Region          Industry      Enterprise Size Ownership Category
 2000/2001                       69.7              69.4                47.8                  87.7
 2001J2002                       50.9               71.5                72.6                 84.8
 2002/2003                       56.9               73.5                54.1                 81.1
Source: Authors' calculations.

3.36 Iti s interestingto notethat, while for regions, industriesandenterprisesizesno significant
patternofevolution over time inthe decomposition ofexcessjob reallocationcanbedetected, an
increaseinthe cross-group excessjob reallocationhas occurred, inaccordancewith the findings
ofthe analysis ofjob flows, where ashift inthejob creationandemployment shares betweenthe
state-ownedandforeign enterpriseshasbeenreported.


       G. ANDWAGEINEQUALITIES                 REMAIN      Low BUTAREGROWING

3.37 Unfortunately, the Slovak Labor Force Survey does notprovideinformationon earnings,
which prevents accurate monitoring of wage dynamics and inequalities. Inaddition to being
somewhat outdated (2002), the Microcensus contains information on individual monthly
earnings but not the number of hours worked and, as discussedextensively inChapter 5, it is
not statistically representativeofthe Slovak population. Thus, the only source ofinformationon
earningsisthe Slovak employer surveyknownasthe "Information SystemonAverageEarnings"
(ISAE) which, although not fully free of samplingbiases, has the advantage of being carried
out on a quarterly basis.26It is important to note, however, that the levels o f wages reported in
this survey are likely to be overestimations ofthe average wages inthe economy, because (i)   the
survey covers only registeredenterprises anddoes not include any informal activities, and (ii)
owing to the potentialsample bias the smaller enterprisesare excluded from the analysis.

What Happened to Wage Levels after 1998?
3.38 Overall, meanhourly wages in2003 were 5.3 percent higher inrealterms thanin 1999,
with the increasefor maleworkers being 7.2percent compared to 4.8 percent for females. The
average wage level varied according to enterprise characteristics, employee characteristics
and occupation. Average wages were higher inthe following:


       0   Foreign enterprises, followed by state-owned enterprises and private domestic
           enterprises
       0   Enterprises with more than 1,000 employees
       0   Bratislava, where average wages reached 16.5 percent above the nationalaverage.

3.39 Moreover, the average wage differential between foreign companies and others and
betweenthe Bratislavaregion andtherest o fcountry increasedrapidly over the last five years.


25 See Faggio and Konings (2001).
26 For more details, see footnote 18, above.

Chapter 3                                                                                           37


Average wages in Bratislava increased by 17.8 percent innominal terms versus an average
increase for the country as a whole of 5.3 percent. This can be at least inpart attributedto the
higherjob creationexperienced inforeign companies and inthe Bratislava region.

3.40 At the individual level, unsurprisingly, wages increase with age, education and
occupational ranking. Managers rank first and are followed by professionals. While skilled
agriculturalworkers receive wages that are comparable to those ofunskilled workers inother
sectors, the share o f skilledworkers intotal employment has increasedover time. In2003,54
percent o fworkers inthe sample had at least completed secondary education (3 percent more
than in 1999). This pattern applies inparticular to the middle occupation classes (technicians,
clerks and service workers).

3.41 Femaleearnings aresignificantlylowerthanmaleearningsacrossallageandeducational
categories. Unskilled female workers earn the lowest wages and males with only basic
education in2003 continue to receive, onaverage, wages comparable to those o fwomenwitha
secondary education. However, because ofthe generalized increaseinreturns to education, by
2003 women with university degrees were paidmore than menwith a secondary education.

And to Their Distribution?
3.42 Table 3.6 highlights a considerable increase in wage inequality over the period
considered, with the overall Gini coefficient rising from a very low 0.236 to 0.279, an 18
percent increase and a value much more inline with international comparisons. The increase
i s particularly high for men (24 percent as compared to 13 percent for women), following a
patterncommonto all advancedeconomies.Overthree-quarters ofthis increase is attributable
to an increase in wage dispersion within foreign companies, with the dispersion within the
domestic private sector declining slightly and that within the state sector increasingslightly.
This is particularly true of inequalities among women which inforeign companies reached a
Ginicoefficient of0.311.

Table 3.6: GiniCoefficientsby Ownership Type and Gender, 1999 and 2003
 OwnershipType                      All                       Male                      Female
                              1999        2003          1999         2003          1999       2003
 All                         0.236        0.279        0.229        0.283          0.218      0.247
 State                       0.213        0.223        0.202        0.23 1         0.210      0.205
 Private                     0.263        0.254        0.263        0.251          0.218      0.225
 Foreign                     0.277        0.316        0.267        0.290          0.252      0.3 11
Source:Authors' calculations basedon Information System on Average Earnings (ISAE).

WhatAccountsfor Wage Inequalities?
3.43 Table 3.7 represents the results o f an OLS regression o f hourly wages over a set o f
variables capturing characteristics of the worker, the company and the occupation. The
findings o fthis exercise are not surprising.

38                                                     TheQuestfor Equitable Growth in the SlovakRepublic


Table3.7: OLSEstimationResults-All Types of Ownershipas aDummy, 1999 and2003
                                   (Dependent Variable: Log of Hourly Wage)
                                                   1999                                     2003
                               Coef                       t-Stat            Coef               t-Stat.
Constant                       3.138                      141.600           3.114              167.258
Worker's characteristics
Female                         -0.190                     -140.906          -0.205             -197.175
Education (Elementary omitted)
Apprenticeship                 0.066                      31.295            0.058              32.352
   Secondary                   0.172                      74.830            0.136              69.994
University                     0.346                      102.812           0.314              109.606
Age                            0.036                      21.138            0.064              43.497
Age"2                          -4.2E-04                   -10.040           -1.1E-03           -31.007
AgeA3                          '7.5E-07                   2.221             6.6E-06            21.795
Union                          0.002                      44.316            0.218              113.880
Occupation (Unskilled workers omitted)
  Managers                     0.780                      180.399           0.065              60.444
  Professionals                0.435                      125.073           0.855              255.457
  Technicians                  0.378                      142.195           0.547              176.439
  Clerks                       0.236                      71.661            0.413              190.778
  Service workers              0.175                      51.846            0.200              82.158
  Skilled agricul. workers     0.195                      32.255            0.044               18.742
  Craft and trade workers      0.203                      81.240            0.145              31.404
  Operators and assemblers     0.254                      105.884           0.164              82.804
Enterprise characteristics
Regions (Bratislava omitted)
   Banska Bystrica             -0.097                     -42.095           -0.182             -105.124
   Kosice                      -0.043                     -19.353           -0.151             -83.827
   Nitra                       -0.089                     -39.950           -0.137             -74.618
   Presov                      -0.113                     -44.100           -0.187             -102.340
   Trencin                     -0.065                     -27.251           -0.142             -80.499
   Trnava                      -0.066                     -22.500           -0.060             -31.605
   Zilina                      -0.089                     -38.210           -0.133             -74.803
Nr employees                   4.1E-06                    75.876            4.4E-05            226.989
Industries (Industries (C-F) omitted)
Agriculture (A-B)                        -0.162           -48.018           -0.163             -57.407
Business Services (G-K)                  -0.071           -34.036           "0.001             0.557
Public Services (L-0)                    -0.406           -166.207          -0.187             -104.126
Ownership type (State omitted)
    Domestic private                     -0.032           -18.979           -0.018              -13.873
Foreign    ,                             0.044,            15.257           0.146               87.200
R-squared                                0.495                              0.591
Adj R-squared                            0.495                              0.591
F-statistic                              8489.2                             21728.7
Prob(F-statistic)                        0                                  0
Observations                       ,     251,446                            436,274
Note: Estimated coefficients are statistically significant at the 1percent level.
(*) Stands for significance at the 5 percent level (**) indicates that the estimated coefficient is not significantly
different from zero at the 10 percent level.
Source: Authors' calculations based ISAE.


3.44 They show the following:
        0   Therearehighretumstoeducation(15 percent for secondaryand39percentfor tertiary
            education in2003) andsome returns to age and to being a member of a union.
        0   Higherwages prevail for professionals andfor employees oflarge firms.
        0   Working outside Bratislavaandinpublic services or agriculture hasanegative effect

Chapter 3                                                                                         39


            onan individual's earning potential. This is particularly true inthe easternregiono f
            Presov and the central region of Banska Bystrica.
        0   The returns to education, age and unionization have not increased substantially
            over time.
        0   However, the differentialbetween occupations and size of enterprise has increased.
        0   Itpays increasingly moreto work inBratislavathan inother parts ofthe c0unt1-y.~~

I s Gender an ImportantCorrelate of Wages?
3.45 At around 24 percent, the gross gender gap is comparable to that of other EU8
countries. Figure3.15 shows that on average the percentageof the gender wage gap that can
be explained bythe differences inthe observedcharacteristics is rather low (around one-third)
and it has remained unchanged over time. However, the explained component has increased
considerably inforeignenterprises anddeclined inthe state sector andamong domestic private
companies. This suggests that the female labor force in foreign companies i s becoming, on
average, comparatively less skilled.

             Figure 3.15: Decompositionof GenderWage Gap in 1999 and 2003

                      -n%exolained
                          .           I%unexolained        % explained      % unexplained
                  100%                                 100%

                   80%                                 80%

                   60%                                 60%

                   40%                                 40%

                   20%                                 20%

                    0%                                   0%
            I            All     State Private Foreign         All    State Private Foreigm


                                       Source: Authors' calculations.

To What ExtentDoes OwnershipCount?
3.46 The descriptive section above concluded that ownership o f enterprises was an
important determinant of wage levels, with foreign enterprises paying much higher wages
than others. However, it is also evident that they employ different types o f workers, on
average more qualified but younger. Table 3.8 decomposes the wage differentials between
different enterprises with different ownership into two components: one that i s explained by
differences inthe workers' characteristics - i.e., the "explained" component -and one that is
due to differences inrewardsto these characteristics.

3.47 The results are extremely interesting. They show that in 1999 the foreign firms in
the sample were paying higher wages for workers who on average had lower productivity-
related characteristics than those working in other enterprises. However, this had changed
considerably by2003, withover 61percento fthewage differentialbetweenforeignenterprises


27  It is, however, important to note thatpart ofthe wage differential betweenBratislava andthe restof
    the country may be due to differencesinlivingcosts.

40                                             The Duest for Equitable Growth in the Slovak Republic


and private domestic enterprises (and nearly 45 percent of that between the foreign and the
state sector) being accounted for by the higher characteristics o f the workers. It also shows
that the domestic private sector may have lost ground in terms of workers' productivity-
related characteristics when compared to the state. This is a somewhat worrying scenario
suggesting two possible explanations: (i) foreign companies have "cream-skimmed" the
                                               the
Slovakprivate sector bybuyingout the most productiveenterprises-i.e., those withthe most
productive workers -or (ii)       the domestic private sector i s losingthe most productive workers
and with them competitiveness relative to the foreign sector.

Table 3.8: Decomposition of the Wage DifferentialAccording to Ownership Type,
 1999 and 2003
~                                                                           1999               2003
 State and Domestic Private
 Mean LogWage Gap:                                                         0.056              0.075
 Differences inworkers' characteristics                                    0.004              -0.042
 Differences inrewards to characteristics                                  0.052               0.117
 Foreign and State
 Mean LogWage Gap:                                                         0.004               0.419
 Differences inworkers' characteristics                                   -0.492               0.185
 Differences inrewards to characteristics                                  0.496               0.219
  Foreign and Domestic Private
 MeanLog Wage Gap:                                                         0.059               0.494
 Differences inworkers' characteristics                                   -0.069               0.300
  Differences inrewards to characteristics                                 0.129               0.180
Source: Authors' calculationsbasedon ISAE.



                   H. THE REFORMBEGINNINGTO BEARFRUIT?
                         I S


 3.48 As pointed out above, the implementation of the most recent, and most far-reaching,
round of reforms in January 2004 marked a watershed in the development o f the Slovak
Republic, which was followed shortly after by EUaccession. A year and a half i s clearly too
 short a period for a full evaluation o f the impact o f a structural far-reaching reform of the
 labor market and the social assistancesystemwhose effects shouldbe fully evident only after
the economy has adjusted to a new higher-employment equilibrium.However, inview of the
importance of an adequateevaluationfor the sustainability ofthe reform process, Policy Note
 1 in volume I1provides some analysis o f the short-term impact using seasonally adjusted
quarterly data, predominantly from the LFS.

3.49 These data suggest that, after an initial negative effect, the full implementation o f the
reforms may have been associatedwith a positive trend inbasic labor market indicators. I t i s
clear from Figure 3.16, for example, that the beginning of the second phase o f reforms that
followed the 2002 election saw a relatively highincrease inemployment, with a year to year
 seasonallyadjusted growtho f 1to 2 percent, drivenprimarily (but notonly) by growth inself-
employment. Indeed, self-employment continued to increase even after the implementation

Chapter 3                                                                                    41


of the 2004 reforms and stabilizedat around2 percent per quarter from the second quarter of
2004. By way of contrast, the number of employees declinedby as muchas 4 percent inthe
second quarter of2004.

3.50 Nevertheless, after a sharp drop in the first half of 2004, overall employment began
to increasewith the year-to-year growth reaching2 percent by the first quarter of 2005. This
positive trend is also confirmed by the quarter-to-quarter data, which show an increase in
overall employmentbeginningfrom the secondquarterof2004 andsuggest that the reduction
inthe first quarterof2004 were the result ofevents inthe lastmonthsof2003 ratherthanthe
direct effect ofthe reform. To what extentthis is thebeginningofapositive trendandwhether
it is attributable to the 2004 reforms is too early to determine, but the signs are encouraging.
The fact thatthisgrowth is drivenpredominantlybygrowthinself-employmentandinsmaller
enterprises i s also worth noting, as it points to economic liberalization as the main driver of
growth andhas important implications for policymaking.

              Figure3.16: SeasonallyAdjustedTrends inEmploymentbyType




                       I

                                        Source: EUROSTAT.


3.51 At the same time, the active population grew steadily by around 1percent in2004 as a
result of the growth inthe working age population, as shown inFigure3.17. Unsurprisingly,
giventhetrendsinemploymentandactivepopulation,theunemploymentrate spikedin2004ql
to 19.5 percent (15-64 age bracket), the second highest figure in the records (after 2001ql).
However, it fell rapidly after that and throughout 2004. The key drivers of unemployment
dynamics in2003 had beenemployment creation. In2004, asjob creationcame to a halt, the
increase in the active population translatedinto higher unemployment.Employment creation
resumed inthe third quarter of 2004, and, together with a slightly lower activity rate, it offset
the populationpressures.

42                                          The Ouest for Eguitable Growth in the Slovak Republic



                 Figure 3.17: Sources of Growth in the Active Population




                                        Source:EUROSTAT,



                                   I.EMERGING          ISSUES


3.52 In a nutshell, the most striking result o f this detailed analysis o f the Slovak labor
market is the degree o f segmentation that has historically characterized the SR and the
fact that this duality has deepened over time. Traditionally, labor economists talk about a
"dual" labor market with reference to a single labor market where two distinct groups o f
workers face very different conditions bothpre- and post-entry indifferent segments of the
labor market. Inthe SR the selection criterion for entry into the two segments o f the labor
market i s the skill level. The reforms o f recent years have helped skilled workers to take
full advantage of the increased competitivenesso f the economy and the rapid growththat
it has generated.Although positive, the progress for the unskilledhas been slower and this
has resulted inincreased inequalities.

Bleak Prospectsfor the Unskilled
3.53 Accordingtotheevidencereportedinthischapter, theprospectsfortheunskilledpopulation
inthe SRarenotpromising, asthey face asignificant disadvantagecomparedto the rest ofthe
populationinall indicators used andthat disadvantageis increasing rapidly. An alarming-by
international standards-proportion     o f unskilled working age individuals, especially males,
leaves the labor force never to go back. Ofthose who remain economically active, over half
are unemployed and most have been unemployed for more than 12 months (LTU). This is
particularly worrying, since the re-employment probabilities for the unemployed ingeneral,
andthe LTUinparticular, are very low by Europeancomparisons and are particularly low for
the low skilled. This is partly due to the relatively lowjob creationo f the types of enterprises
that tendto employ ahighershareo funskilledlabor, which i s clearly amajor concernnotonly
for the low-skilled unemployed but also for school-leavers without adequate qualifications.
Finally, those who succeed infinding and maintaining a job despite their low skills face an
increasing wage gap compared to other workers, given the growing inequality in earning
distribution.

Chapter 3                                                                                  43


3.54 This suggests a crucial, and growing, role for education indetermining labor market
opportunities and curbing social exclusion. The "new" SR i s increasingly becoming a
"knowledge-based economy" where the returns of education are seen not only in terms
o f higher wages (although these clearly exist) but especially in terms o f better access to
jobs and economic opportunities. This also signals the risk o f considerable inefficiencies
emergingfrom the existing skillmismatches and the threat that they may pose to economic
growth. If, for example, the economy and employment were to grow by roughly 20 percent
(that is, to the point at which there were enough job places to eliminate unemployment
completely), and the skill composition of employment and the labor force were to remain
unchanged, some 80 percent o fthe low-skilled (namely, one-fifth o fthe unemployed) would
not find employment. At the same time, there would be a shortage o fhigh-skilled persons to
filltheopeninghigh-skillvacancies.

Ample Evidence of the GeographicalSegmentation of the Labor Market
3.55 Another salient feature of the Slovak labor market is provided by the magnitudes of
the existing regional disparities in labor market indicators in a relatively small country. At
around 8 percent, the unemployment rate in Bratislava i s less than one-third of that in the
disadvantaged regions o f Banska Bystrica, Kosice andNitra, and at the NUTS3 level, the SR
has the second largest variance in the regional unemployment rate in the EU25 after Italy.
Other labor market characteristics such as the employment rate or long-term unemployment
also vary substantially by region.

To What ExtentAre Skills and RegionalMismatchesRelated?
3.56 Using two major indicators of labor market performance, Figure 3.18 suggests the
existence of a strong correlation between the skill composition o f different regions and the
prevailing labor market conditions, so that regions with a relatively high share of unskilled
population also have worse labor market opportunities. An important issue from a policy
point of view, then, i s the extent to which the poor educational indicator can be considered a
cause of the poor labor market outcomes inthose regions.

              Figure 3.18: Negative Correlation between Employment Rates
                                   and Skills by Districts, 2004


                                 75,                                         1
                                 70--

                                 65--

                                 60 --

                                 55 --

                                 50 --

                                 45                           . -
                                    --

                                 40 -                *

                                 35               %low skillin population 2004
                                    7




                         Source: Slovak Labor Force Survey, authors' calculations.

44                                             The Questfor Equitable Growth in the Slovak Republic



3.57 Thesimplestreasonwhyskillcompositioncanproducespatialdifferencesinunemployment
rates is that high-skilledworkers tend to have a lower natural rate of unemployment (since job
matching - that is, hiring, training and firing- is much more costly for them).28If one of the
factors was not mobile and was unequally distributed, then high-skilledworkers would cluster
inhigh capital regions (or vice versa) and a difference innatural unemployment rates would
emerge. However, cross-regionalvariationsintheunemploymentrateinthe SRare too largeto be
explained inthis way.

3.58 This situation points to other factors being at play, factors that are strong enough to
preventthe equalizationof unemployment rates. These effects include: agglomeration (firms
tendto cluster to exploitnetwork externalities), thetransportation advantagesofsomeregions,
the marginal tax imposed under the old system by the combinedwelfare-tax system, sorting
effects (where high-skilled workers are reluctant to move to low-skill regions), rigidity in
relative wagesbetween highand low skills, imperfectioninthe housingmarket, especially for
rented accommodations. Also, it is plausible to assume that, because of technology, various
externalities and network effects, modern economies require skills in certain proportions.
The proportion is weighted increasingly towards high skill. As more mobile high-skilled
workers migrate to more dynamic regions, the stock of human capital inthe lagging regions
deteriorates andthe skill mix becomes sub-optional.

3.59 For a more precise answer to the question of how much o f a regionalvariation inlabor
market outcomes (if any) can be explained by regional skill mismatches, two models are
estimated inwhich the probability o fbeingjobless i s definedas depending onbothindividual
andregionalcharacteristics, includingvariables relatedtotheskillmismatch(see Box3.1). The
findings point to the individual's level of education as the most important single determinant
o f employment status (apart from age) with the secondmost important determinant beingthe
region inwhich the person lives. Regional skillmismatches also matter, with the regions with
the lowest skillmismatcheshavingthe highestemploymentrates.However, giventheplausible
range ofvalues for these skill mismatches andcoefficients, these effects arerelatively smaller.
These results suggest the importance of tackling inequalities in labor market opportunities
via a combination o f individual specific policies and a regional development strategy. At
the individual level, improving the skills of the unemployed is paramount for better market
outcomes.At the regionallevel, geographically targeted interventionsmay berequired.These
interventions should be designedto address the current barriers to a more effective strategy
for regional "cohesion."




28  Assumingthe idealworldo fclassicalproduction function, competitivemarketsandfullemployment
    (subject to turnover), but constraining the mobility o f one production factor: if capital and labor
    were perfectly mobile, factor movement would make the spatial distribution o f skill uniform and,
    hence, would produce no unemployment differences.

Chapter 3                                                                                           45



                  Box 3.1: Are RegionalInequalities inthe EmploymentSituation
                                    a Matter of Skill Composition?
 One way of assessingthe extent to which regional skill composition affects regional employment
 levels through channels other than individual characteristics i s to model the probability o f being
 employed as a function o f other regional and individual characteristics. To this end we estimate
 two models, both including some measures of regional skill mismatches andor composition. In
 the first model we use two "skill-mismatch'' variables: deficit o fhighskills and oversupply o f low
 skills. Other region-specific variables include: share o f industry and services and dummies for
 regions. Individual characteristics include an age and sex profile meant to account for age-related
 nonlinearities. The reference education category is basic education (low skills). In the second
 model, share of high skill and share o f medium skill in the region have been included to control
 for the regional skill structure. The models are estimated as a binary choice model (probit) and are
 intended to describe the probability o f beingjobless, given individual andregional characteristics,
 for individuals not ineducation. Results are presented inthe table.

  Estimation of RegionalMismatchesversus Skill Mismatches
  Model1: Joblessness& askill mismatch                 Model2: Joblessness& askill composition
  Dep. variable: not-employed   Coeff.    z-stat       Dep. variable: not-employed  Coeff.   z-stat
  Sex and age profile                                  Sex and age profile
  Woman? [dummy]                0.097     6.9          Woman? [dummy]               0.099    7.0
  age if 15<age<30              -0.026    -30.5        age if 15<age<30             -0.026   -30.3
  &woman                        dropped                &woman                       dropped
  age if30<age<55               0.018     11.1         age if30<age<55              0.018    11.1
  &woman                        -0.018    -8.7         &woman                       -0.018   -8.5
  age if55<age<65               0.158     14.6         age if 55<age<65             0.159    14.6
  &woman                        0.069     30.1         &woman                       0.069    30.2
  age 55-60? [dummy]            -2.099    -11.4        age 55-60? [dummy]           -2.101   -11.4
  &woman                        dropped                 &woman                      dropped
  age 60-65? [dummy]            -1.104    -4.7          age 60-65? [dummy]          -1.103   -4.7
  &woman                        -1.224    -16.2         &woman                      -1.227   -16.2
  Individual characteristics                            Individual characteristics
  Heado f household? [dummy] -0.179       -11.4        Heado f household? [dummy] -0.177     -11.2
  medium education? [dummy] -0.997        -59.0         medium education? [dummy] -0.978     -57.4
  higher education? [dummy]     -1.671    -62.7         higher education? [dummy]   -1.639   -60.8
  Regional characteristics                              Regional characteristics
  Tmava? [dummy]                0.223     7.8           Tmava? [dummy]              0,159    5.2
  Trencin? [dummy]              0.203     6.1           Trencin? [dummy]            0.178    5.1
  Nitra? [dummy]                0.631     21.3          Nitra? [dummy]              0.538    16.8
  Zilina? [dummy]               0.465     15.8          Zilina? [dummy]             0.427    13.9
  Bystrica? [dummy]             0.579     21.2          Bystrica? [dummy]           0.475    15.6
  Presov? [dummy]               0.552     18.7          Presov? [dummy]             0.467     15.3
  Kosice? [dummy]               0.662     24.4          Kosice? [dummy]             0.569    19.2
  share o f industry            -0.004    -4.6          share o f industry          -0.005   -5.6
  share o f services            -0.011    -5.9          share o f services          -0.010   -5.6
  Skill mismatch                                        Regional skill composition
  deficit o f highskills        -0.007    -0.3          share o f highskill         -0.013   -8.3
  oversupply o f low skills     -0.006    -3.8          share ofmedium skill        -0.010   -7.1
  N=70266, ps.R2=0.2694                                 N=70266, ps.R2=0.2647

 ResultsofEstimation 1are presentedinthe table. All coefficients are signed as expectedandmost
 are significant at the 5 percent and 1 percent levels. Results indicate that the skill composition o f
 the region inwhich a person lives plays a role indetermining this person's labor market status, on
 top of the individual's skilllevel and pure regional effects. However, given the plausible range o f
 values for skillmismatches andthe respective coefficients, these effects seem too small to account
 for and explain a significant part o f the differences inregional employment rates.


                                    CHAPTER 4 -

    HAVE RECENTREFORMSWORSENEDTHE
         ODDS FORTHE MOSTVULNERABLE?


4.1    Any policy change is likely to worsen the welfare of some while improving that of
others. The "potential Pareto" criterion advocates in favor of implementing those policies
that increase the overall welfare of the population- i.e., to the point where the sum of the
benefits to the winnersexceedsthe sumo fthe costs to the losers. The policy will thenbecome
an "actual Pareto improvement" ifthe losers can insome way be compensated for the loss of
welfare they face so that their final welfare i s unchanged.

4.2    There i s little doubt that, overall, the Slovak reform represented a "potential Pareto
improvement". The rationale for the reforms was clearly andfully consistent with the overall
objective o f increasing employment opportunities and strengthening the incentives to take
up ajob for the overall population. As such, it addressed many of the criticisms of the old
system as they had been pointed out by both Slovak and international                 Thus the
reform package was necessary and long overdue, since the old system was outdated and
economically unsustainable. This was clearly awidely sharedview, as according to anopinion
poll conducted inJune 2004 by the Markant Agency 82 percent o f respondents were infavor,
to varying degrees, o f the changes to the tax code, while 57 percent were in favor of the
changes insocial benefits.30This conclusion is also supported by the results of the qualitative
analysis conducted as backgroundto this report, which was basedon detailedinterviewswith
main stakeholders in 11sites,

4.3    Nevertheless,whenthereformswere introducedinJanuary 2004, the SRexperienced
considerable social unrest due to the concern that the reforms had shifted resources away
from the most vulnerable and towards groups o f individuals that, while poor, were not
as socially excluded as those that lost out. Evaluating the full impact o f the changes is,
therefore, critical to the sustainability o f the reform process, which could potentially be


29 See also World Bank (2004), Doing Business in 2005, for a more detailed description and an
   international context.
30 Miroslav Beblavy, ?Reform o f the Social System: Motivation, Targeting and Consolidation,"
   presented September 2004.

48                                            The Questfor Equitable Growth in the Slovak Republic


jeopardized by the concerns about its effect onthe livingstandards o f the most vulnerable
groups.

4.4   This chapter attempts to master this task, which is made difficult by a number of
          The first is the very broad scope of the reform which, as described below, was
extremely comprehensive and far-reaching. This makes it extremely difficult to evaluate its
full impact inview of the potential interaction of its many components. This report takes a
"partial equilibrium" approach-that is, it focuses on two prominent aspect of the reformand
attempts to evaluate their impact on the level of and inequality inthe living standards of the
population under different assumptions about what happens to the rest o f the economy. The
aspects analyzed are the following:


      0   The introductionof the flat-rate tax
      0   The changes inthe structure andthe average value ofthe social benefits.

4.5   Data limitation i s the other reason why a satisfactory impact evaluation cannot
be carried out. Detailed information on income is only available for 2002-in            the 2003
Microcensus. Thus, the data predate the reforms and cannot beusedto evaluate their impact.
However, despite the limitations on the accuracy o f the income reporting, the data could
provide a useful basis for comparing the income levels and distribution before and after the
reform.


       A. WHYREFORM INCOME      THE               TAXAND     THE   BENEFITS     SYSTEM,
                                            AND HOW?


4.6   There was broad agreement among both Slovak and international observers that
the social benefits system in operation in the SR over the 1990s acted as a reasonably
effective safety net mechanism, but also that (i)it was fiscally unsustainable; (ii)             it
provided a financial disincentive for individuals with low earning potential to actively
seek employment; and (iii) hightax wedge that it generated resulted inbarriers to job
                              the
creation.

4.7   Theneedfor far-reaching reformshadbeenarguedconvincingly inthe previousWorld
Bank Living Standards Assessment (World Bank, 2002), which had reached the following
conclusions:


      0   The pre-2002 social benefits system was an effective safety-net mechanism.
      0   However, its generosity generatedbotha"poverty trap" andan"unemployment trap"
          which removedthe financial incentive for some groups of workers to actively seek
          employment. This was because the value of the benefits received by workers with
          low earning potentialwhen out o fwork (and inpoverty) was higherthanthe amount
          that they could have received inemployment -i.e., the replacement ratio was close

                                       ~~~~




31 For more details on the analysis, see Policy Note 2 inVolume 11.

Chapter 4                                                                                           49


          to       This is clear from Figure4.1, which plots total income (including benefits)
          against gross wage for selectedgroups of householdsunder the two different social
          benefit regimes that were in operation before the reform.33The last two graphs in
          particular show the lack of incentive generated by the old benefit system for the
          second adult to work ina low-income two-adult household with children.
          Thehighratesofunemployment andinactivitywereassociatedwithlowjob creation
          (especially in some regions) and with the existence of severe mismatches between
          the labor suppliedand that demanded.Among the mainreasonsfor lowjob creation
          was the comparatively high tax wedge on wages required to finance the generous
          benefits system.Thiswas particularlybitingfor the lower levelofwages and inlow-
          productivity/low-employmentregions.

4.8   Inresponsetotheweaknesses ofthesystem,the"New SocialPolicy" ofthegovernment
elected in the autumn of 2002 focused on employment creation, poverty reduction, and the
improvement of the fiscal sustainability of the benefits system. The reform addressed three
interrelated sectors: (i)social assistance and family policy; (ii)     the labor market and active
labor market policy; and (iii)the pension and social insurance system. It had three guiding
principles: (i) and simple rules, (i) individuals, and (iii) w o r k ~ a y . ~ ~
               clear                          active                        making




32 The "replacement ratio" is the ratio o f income out ofwork to potential work income.
33 Under the old system individuals in material need were divided into two distinct groups (for
   objective reasons and for subjective reasons) depending on their recent job search history. Those
   who were inmaterial needfor subjectivereasons-i.e., who were able to work butwere not looking
   for ajob, hadbeen deleted fiom the unemployment registry, or were not paying alimony -received
   lower benefits than the rest (generally, the elderly and the sick). I t is worth noting that the total
   number o f beneficiaries under the two regimes was approximately equal. For details, see Policy
   Note 2 inVolume 2.
34 See Concept o f Pension Insurance, approvedby the government onApril 2,2003; Strategy o f Labor
   Market Reform to Generate Growth inEmployment, approved on April 30, 2003; and Strategy of
   Public Sector Employment, approved on June 11,2003.

50                                                                              The Questfor Equitable Growth in the Slovak Republic


    Figure 4.1: The "Poverty Trap" for Different Family Typologies under the 2002
                                                                         Benefits System

                                                                      Single hou hold, withjob


    225,000 .                                                                          225,000  .
   205,000  .                                                                          205,000 -

    185,000 -                                                                          165,000 -

    165,000 .                                                                          165000.

    (45.000 -                                                                          U5,000 -

    125,000 -                                                                          t25.000  .

    m5.000  .                              - - -                                                                                     -
                                                    -Netlncoma2002obj                  M5,OOO .  a                             II      .Nellncama2002obj

     85,000 .                                                                           85,000 .
                                          -NeIIncome2002subj                                                                   -Net         Income 2002 subj
     85,000 .*                                                                          65,000 -

     45.000   . , , , , . , , , . , , , , , , ,   . , , ., , , , ,. , , . ,, 1
             0    72,000 98,000 120,000 U4.000 188.000 (92,000 218,000240,000
                                   AnnuaIGmiiWsgl)

                                                                                  0thwith job
                                                                                                                       3 children
                                                                                       285.000
                                                                                       265,000
                                                                                       245,000
                                                                                       225,000
                                                                                       205,000
                                                                                        185,000
                                                                                        185,OO 0
                                                                                        115,000

                                                                                        125.000
                                                                                        105.000
                                                                                                                                         .
                                                                                        85.000
                                                                                        65.000   I.    , , , ,, , , .           .                       . .    . ,
                                                                                                                      I . . , ,   , , , ,  , , , , , , ,   , ,
                                                                                                  0      U4,OOO 188,000 192,000 2~.000240,000284,000288,0003~2,000
                                                                                                                         Annual Gm11 Wage


                                                   A couple, one withjob, other long-termunemployed


     215,000 .                                                                         225,000 -

     205,000 .                                                                         205.000   .
     ms,oao -                                                                           185.000  .

     165000.                                                                            165,000 -

     (45,000 .                                                                          UP,OOD

     125.000 -                                                                          U5,000   .
     105.000 -a                                      -Net Income 2002 obi               m5,aao   .Di
I     85.000 -                             -NetIncome2002subl                           85,000   .                             -Netlncome2002aubj

      55,000   . , ,.  , , , , , , ,. , , ,  . . .,   , ,. . . . . . . . . .
                                                           ,                 I          8 5 , O O O r , , ,, , , ,, . , , , , , , , , , ., .  , , , ,. ,. , , . , I
              0    72,000 96,000 t20.000 144.000 88.000 9 2 0 0 0 216,000240,000
                                    Annual Gioili Wage


Note: Whenthere is zero gross wage, netincomeof the unemployedperson(couple) is consideredwiththe same
number of children.A gross wage of 72,000 roughly reflects the minimumwage in 2002 (in 2004 prices). Total
grosswage ofthe workingcoupleassumesthat one ofthe partnersreceivesonly the minimumwage.
Source:Authors.

Chapter 4                                                                                             51


        B. THEMILESTONESREFORM ITS FEATURES
                                      OF THE                    AND       MAIN

4.9    The reform was implemented in two stages. The first was primarily a fiscal maneuver
designedto addressthe concernsofexternalobserversaboutthe sustainabilityofthebudgetdeficit
and to reduce fiscal pressure by cutting child benefits and other social transfers.35Meanwhile,
the New Social Policy was developed intwo concept papers on (i)pension reform and (ii)                  a
"Strategy of GrowthinEmploymentBasedonthe Labor MarketReform". The mainelements and
milestonesofthe New SocialPolicy are highlightedinBox4.1.

I           Box 4.1: The Milestonesofthe Reform:Action Plansand New Legislation
  The concept papers were elaborated in two National Action Plans and a Joint Inclusion
  Memorandum (JIM):
  a The NationalAction Plan on Social Inclusion 2004-2006 was adopted by the Government o fthe
    Slovak Republic on July 14,2004.
  a The National Action Plan for Employment 2004-2006 was adopted by Governor o f the Slovak
    Republic on September 22,2004.
  a The Joint Inclusion Memorandum was prepared with support from the European Commission,
    Directorate-General for Employment and Social Affair. Memorandum. It outlines the principal
    challenges in tackling poverty and social exclusion, presents the major policy measures taken by
    the Slovak Republic andidentifies the key policy issues for future monitoring and policy review.
  The Action Plans served as background materials for designing ten new legislative pieces, which
  reformed the social system and labor market from January 2004. These are:

  0 Flat personal income tax (PIT) at 19percent, and corporate income tax at 19 percent.
  a Single value added tax at 19 percent.
  0 Introduction o f a single basic social benefit for the poor, allowances for health care and housing,
    and activation andprotection benefits.
  a Introduction o f a single child benefit for all parents paidin cash and as a tax credit.
  a Simplifiedcalculation o fthe base for andpayment o f social benefits andincreases inthe benefits
    for families with children.
  a New active labor market policies requiring the registeredunemployed to visit the Labor Office
    every two weeks, long-term unemployed every seven days, and activation workers once a
    month. The new instruments include training, services for families with children, the "activation
    allowance," the mobility allowance and allowances for disabledworkers.
  a The merging of social and labor agencies into Offices o f Labor, Social Affairs and Family (Labor
    Offices).
  a More transparent social insurance. The act on social insurance made social insurance more
    transparent reduced certain contributions, re-defined sickness and unemployment benefits and
    prepared the conditions for pension reform.
  During 2004, several amendments to existing legislation increased the amounts of the social
  benefits, housing allowance, child benefit, child birth grant and multiple birthbenefit, foster care
  benefits and funeral allowance (government decrees nos.790-795 fromAugust 18,2004).




35  The cut inchild benefits was from Sk 480-Sk 1,290 per monthto Sk 270-Sk 890 per month. Social
    benefits were reduced in size but most important was the introduction o f a monthly ceiling per
    household set at Sk 10,500.

52                                             The Questfor Equitable Growthin the Slovak Republic


4.10 The focus of this report i s on the second component of the reform - the reform of
benefits andtaxes -which was approved by the government onApril 30,2003 and developed
into new legislation (see Box 4.1 for a list and description) which substantially reshapedthe
tax and benefits systems inthe following ways:

       a   The tax system by introducing (i)what is increasingly known as the flat tax on
           personal income, and (ii)   a single bracket value addedtax (VAT)
       a The social benefits system        by (i)  lowering the base benefit; (ii)supplementing
           it with an additional benefit conditional on employment; (iii)           changing the child
           benefit from a means-testedbenefitto a flat benefit tax credit; and (iv) introducing
           more proactive labor market policies.

Introducing the Flat Tax
4.11 Table 4.1 summarizes the changes introduced by the reform o f the personal income
tax. The initial five brackets and the corresponding rates between 10 and 38 percent were
replacedby a 19percent flat tax with a sharp increaseindeductibles (more than double) and
the introductiono f a tax credit on dependentchildren. The flat tax i s a relatively new concept
inpublic finance that hasbeengainingincreasing support andprominence after its seemingly
successful implementation inCentral European and some CIS countries (see Table 4.2). Its
mainadvantagesare seen to be (i) ease of implementation, (ii) highertransparency, and
                                     its                                 its
(iii) increasedcomplianceratethatitappearstogenerate.Although-asisevidentfrom
     the
Table 4.2 - countries vary substantially inthe rate they adopted, the overall evidence i s of a
very successfulalternative to a more complex tax system incountries with a relatively young
systemof collection and enforcement and generalized issues of governance and compliance.

Table 4.1: Personal Income Tax in 2004 and 2002
                  2004                                    2002
                                                          Incomerange, rateonthe rate+fixed amount
                                                          Sk 0-Sk 90,000
                                                                                         20% + Sk 9,000
                                                                                         10% + Sk 0
Tax rate           19% flat                               Sk 90-Sk 180,000
                                                          Sk 180,000-Sk 396,000
                                                                                         35% + Sk 87,480
                                                                                         28% + Sk 27,000
                                                          Sk 396,000-Sk 564,000
                                                          Sk 564,000 andmore             38% + Sk 146,280
Deductibleitems
 Taxpayer         Sk 80,832*                               Sk 38,760 (Sk 48,000 ifhandicapped)
 Spouse           MAX (Sk 80,832 minus spouse's income, 0) Sk 12,000 ifhadnoincome
 Old-age                                                   Sk 0 (Sk 8,400 ifpartially disabled; Sk 16,800
 pensioners       Sk 0                                     ifdisabled)
 Dependent child Sk 0                                      Sk 16,800 (Sk 22,800 ifhandicapped)
Tax credit
                  Sk4,800 only ifparent's annualgross
 Dependentchild incomeexceeds 6 x minimumwage, i.e. Sk Sk 0
                  36,480
* The amount is set to 19.2 times the minimum subsistence level as of January every year (see Act No.
601/2003).
Source: Act on IncomeTax No. 595.

Chapter 4                                                                                                                                                                 53


Table 4.2: The Flat Tax in Central Europe
Country                                                                                          Rate                                 Year
Estonia                                                                                           26                                  1994
Lithuania                                                                                         33                                   1994
Latvia                                                                                            25                                   1995
Russia                                                                                            13                                  2001
Serbia                                                                                            14                                  2003
Ukraine                                                                                           13                                  2004
Slovak Republic                                                                                   19                                  2004
Georgia                                                                                           12                                  2005

Source: The Economist (2005).

Changing the Social Benefits System
4.12 Recent reforms have maintainedthe traditional categories o f benefits and introduced
new types o f interventions, conditional on the individual's taking a more active role inthe
labor market and the economy, the so-called "activation program". Nevertheless, the overall
expenditure on social benefits has declined from about 11.5 percent of GDP in 2002 to 9.9
percent o f GDP in2004. As shown inFigure 4.2, the main reason for the decline is the sharp
reduction in social assistance and in sickness benefits (more than 50 percent) but spending
on active labor market policies has also declined by two-thirds, and benefits to the severely
disabled were reducedby more than one-third.

                       Figure 4.2: Social Expenditures in the Slovak Republic, 2001-04

            Social expenditures 2001-2004, percent of GDP                                              Social expenditures 2001-2004, total = 100 percent
           Pensions                                       E Social assistance benefits                  Pensions
                                                                                                        Fami benefits                 0 Sickness benefits
                                                                                                                                        Social assistance benefits
           Benerits to severei disabled
           Famii benefits                                 0 Sickness benefits                        rn Benevits to severetv disabled
        rn Active labor marke?policies                      Unemptoymnt benefits                     iActivelabormarkefpolicies         Unemutovrrant benefits
                                                                                                                                              . .

       12                                                                                              mo%
                                                                                                       90%
       10                                                                                              60%

        6                                                                                              70%
                                                                                                       60%
        6                                                                                              SO%
                                                                                                       40%
        4                                                                                              30%
        2                                                                                              20%
                                                                                                        m?4
        0                                                                                               0%
                     Y O 1                      Y O 2             YO3               YO4                           Y O 1          Y O 2         YO3             YO4

            Social expenditures 1q2003-1q2005, SK million                                              Number of recipients 1q2003-1q2005, `000 persons
      8,000 ,                                                                                        2,500 1 2003under-nported, see 4.U
      1,000
                                                                                                     2,000
      6,000

      5,000                                                                                          1,soo
      4.000

      3,000                                                                                          1,000

      2,000                                                                                            so0
       1.00 0

            0
                 l q 0 3   2 1 0 3       3 q 0 3   4 1 0 3 1904   2 q 0 4 3 q 0 4 4 q 0 4 l q O S            lq03  2103   3q03  4q03   lq04  2q04     3q04  4q04   lq05

          oFamilybenefiis                                  01 Social assistance benefits              0 Familybenefits V                   Socialassistance benefits
           m O l h a r S o c i a l S t a t e Benefits      sEensfits lo ssvereiydisabled              Q Other SocialState Benefits      t3 Benefits lo severelydisabied21

     I/Under-reportedin2003becauseofunavailablenumberofrecipientsofchildallowances,
     21Under-reported in 2003 becauseof unavailable number o f recipients o f purpose benefits.
     Other Social State Benefits include allowance for families o f soldiers, and burial and housing allowances


      Sources: Statistical Office of SR, Ministry ofLabor and Social InsuranceAgency, Authors' calculations.

54                                                The Questfor Equitable Growth in the Slovak Republic



Table 4.3: SocialAssistance Benefits Before andAfter the Reform, 2004 and 2002
                           I Description/Income   related   I2004                I2002 (in2004 prices)
 Family-related
   Child birthgrant 11       one-off payment / Not           Sk4,320 or less      Sk3,370 or less
   Multiple birthbenefit     annual I Not                    Sk2,340-Sk3,070      Sk2,340-Sk3,080
   Child benefit             monthly / 2004 Not, 2002 Yes    Sk540                Sk520-Sk970
   Parental allowance        monthly / 2004 Not, 2002 Yes    Sk4,llO or less      Sk2,970
   Foster care allowances    three allowances / Not
 Poverty-related
                                                                                  depends on past
   Social benefit            monthly / Yes                   Sk4,850 or less      behavior andfamily
                                                                                  size, no cap
   Housing allowance         monthly / Yes                   Sk1,670 or less      NIA
   Lump-sum benefit          annual, no entitlement / Yes    Sk13,740 or less
   Health-care allowance    monthly / Yes                    Sk50                 N/A

   Activation allowance     monthly during activation
                            works / Yes                      Sk1,500              N / A

   Back-to-work allowance   monthly / Yes                    special formula      N/A
   Protectionallowance       monthly, exceptional /Yes       Sk1,500              N/A
   Subsidy for catering and  daily and annual, respectively, Sk25 catering,
 school needs                paid to school I Yes            Sk1,OOO school needs N/A

   Scholarships for pupils   monthly, depends
                             on grades / Yes                 Sk500 or less        N/A

   Scholarships for          monthly, depends
   students                  on grades I Yes                 Skl,200 or less      NIA

 Housingbenefit              monthly I Yes                   special formula      special formula
11Lower amount paid for second and subsequent children. Payments per child.
Sources: Law N o 23511998 and authors' calculations.


4.13 Family-related benefitsconsistofthe childbirthgant,multiplebirthbenefit, childbenefit,
parental allowance, andfoster care allowances. The most important i s the child beneJiP6which
in2002 was income-related but is now afixedamountperchildindependentofthe household
income (see Table 4.3).37Poverty-related benefiW8are provided to all inmaterial need - i.e.,
those under a minimumsubsistence level which i s defined by the structure and composition
of the household.39The reform increased the minimumsubsistence level for adults, but left
it almost unchanged for dependent children. However, the amount of social benefits paid to
households below this level was substantially reduced (see Box 4.2 for minimum, maximum
and meanbenefits paid in2001-04). Between2002 and 2004 the minimumsocial assistance
benefit per recipienP declined by 59 percent, the meanbenefit declined by 56 percent andthe
maximumbenefit declined by 34 percent. The special housingbenefit in 2002 was intended
for low income families andwas phasedout in2004 and replaced by a more systemic housing
allowance.


36 Act No. 60012003 on the childbenefit.
37 For details o f the payments see Policy Note 2 inVolume 11.
38 Acts on Social Assistance no. 59912003 and no. 195/1998. Note that social assistance is in the
   competency o f municipalities and the Labor Office.
39 The details of the benefits paidbefore and after 2002 are given inPolicy Note 2 inVolume 11.
40 Recipient defines a householddependent on social assistance.

Chapter4                                                                                                               55



                      Box 4.2:Average SocialAssistance BenefitsandNumber of Recipients
 The minimumsubsistencelevel increased inreal terms for adults, andremained almost unchanged
 for dependent children. However, the amounts o fthe social benefitswere cut so muchthat between
 2002 and 2004 the minimum social assistance benefit per recipient declined by 59 percent, the
 mean benefit declined by 56 percent, and the maximum benefit by 34 percent (see below).
                           Figure(i).AnnualSocialBenefitsPaid,perRecipient,2001-2004
                                          (minimum, maximumandmeanvalues)
                         Nominal amounts, Sk                                          In2004 prices, Sk


 26,000  -                                                                  26,000  -
 26,000  --                                                                 26.000 T

 24.000 --                                                                  24.000 .-
 22.000 --                                                                  22,000 --

 20.000
  UI.000 ---                                                                20,000 --
                                                                             *DO0   --
                                                                                    --
  1.000  --                                                                  8.000
                                                                                    --
  y1.000 ..                                                                  11.000
                                                                             Q.000  --
                                                                                    4                                 I
  13,000 4                                                                   8,000
                                                    I                                   YO!      YO2      YO3    YO4
               YO1         YO2       YO3      YO4



 (annual paymentper recipient in Sk)                           (annual payment per recipient inSk)
 Maximum             22,505     22,066      19,479    20,247   Maximum     27,116          25,738     20,940   20,247

 Minimum             15,205     15,313       13,998    11,520  Minimum     18,320          17,861     15,048   11,520

 Mean                19,289     19,504       17,255    17,145  Mean        23,240          22,749     18,550   17,145

 (in percent o fannual payment per recipient in2004)           (in percento f annual payment perrecipient in2004)

 Maximum              111.2      109.0          96.2    100.0  Maximum        133.9         127.1      103.4    100.0

 Minimum              132.0      132.9        121.5     100.0  Minimum        159.0         155.0      130.6    100.0

 Mean                 112.5      113.8        100.6     100.0  Mean           135.6         132.7      108.2    100.0

 Note:Maximum, minimumand meanpayments were calculated from an average annual benefitper recipient
 (dependent household).
 However,onemayarguethatfor analyticalpurposesitmakessensetocomparedatafor the fourthquarterofeachyear
 toallowforthephasinginperiodofchangesinsocialassistancein2004(includingthe introductionandimplementation
 of the changes). Therefore, belowwepresentacomparisonoffourth quartersfor 2001-2004.Itisclearthatinnominal
 terms averagebenefits didnot change. Inrealterms, howev&, benefits in4q2004 were lower by 20 percentthan in
 4q2001, by 18percentthan in4q2002, andby6percentthanin4q2003.

                      Figure (ii). Annual SocialBenefitsPaid, perRecipient, 4q2001-4q2004
                                         (minimum, maximum and meanvalues)
                         Nominal amounts, Sk                                            In2004 prices, Sk


    28000   -
    26 000  ..
                                                                              26 000
    24000   --
                                                                              24,000
    22000        f                                                            22,000
    20,000                                                                    20,000
     18000'                                                                    #,OOO
     S 000  -.                                                                 8 000
     Y.000  -.
                                                                                 000
     TZOOO  .-                                                                 12,000

     N""" 7
                4qYO1     4qY02    4qYO3    4qYO4



                                        Source:Ministry of Labor andauthors' calculations.

56                                                          The Questfor Equitable Growth in the Slovak Republic


4.14 Becauseoftighter entitlement rules, the number ofrecipients declined also, by as much
as 38 percent, between January 2001 andDecember 2004 (see Box 4.3). This was inline with
a 34 percent decline inthe registered unemployment rate. However, large disparities existed
across districts in the extent of the reduction in number of recipients. While the median
change between January 2001 and December 2004 was large (a 41 percent reduction), there
were districts that remainedwell below that rate (such as Medzilaborce and Poltar) owing to
a persistent lack ofjobs inthose districts and limited labor mobility.


         Box 4.3:Reduction inthe Number ofRecipientsofthe SocialAssistanceBenefitsbetween
                                               January 2001 andDecember2004
  Because of the lower socialbenefits, the number ofrecipientsalso declined: by 38 percentbetweenJanuary
  2001and December 2004. 1/ This seems to be very muchin line with a 34 percentdecline inthe registered
  unemploymentrate (althoughthe correlationbetweenthe two variables calculatedon 79 districts was 0.49
  only) andinlinewitha 15percentdecline intheLFSunemploymentrate.However,the registryofdependents
  containsdataonly onthosehouseholdsthatwerereceivingsocialassistancebenefits(Le., thosewhose income
  was lower than the sum of the benefitsfor which they were eligible), and not on those that were below the
  MSLpoverty line. Becausethe benefitswere cut significantly,previouslyeligible personswould drop from
  the registry independentlywhether or notthey found anew source of incomeCjob).

  Below we presentthe distributionof districtsby percentagereductioninthe number ofrecipientsbetween
  January2001 andDecember2004, andfive districtswith the greatestand leastreduction.While the median
  changebetweenJanuary2001 andDecember2004was large(41 percent reduction), there are districts that
  stayedwell below that rate, suchas MedzilaborceandPoltar.
                               Numberof Districtsby percentReductionof Recipients


                                                                 Notes:
                                                                 Leastchange: 10percentreductionina number of

                                                                             nge:41 percent reduction ina numberof

                                                                             hange: 75 percentreductioninanumber of

                                                                 Tables below:districtsrankedby percentchangeinnumber
                                                                 ofrecipients. Registeredunemploymentrate (LFS not
         20 -25 -30 4 5 4 0 4 5 -50 -55 -60 -65 -70 -75          availableby district).
             % reduction In number of recipients                 Source: Ministryof Labor and authors' calculations.




                 Districts withmost change,                                     Districts with leastchange
                       percento f              Unemployment                           percent of             Unemployment
                       dependents                      rate                           dependents                    rate
                     Dec04         JanOl       Dec04       JanOl                    Dec04        JanOl       DecO4      JanOl
  Namestovo              3.4          13.8         10.4     23.1 Medzilaborce          11.4        12.6          20.6    26.1
  Myjava                 2.8           9.4         7.7      15.6 Poltar                11.5        12.8          24.4    27.7
  Tvrdosin               3.3          10.0         11.7     22.6 R. Sobota            21.8        24.4           28.7    37.6
  Puchov                 2.1           5.8         6.9      13.4 Revuca               21.1        23.7           28.4    34.4
  Cadca                  3.6           9.8         10.8     20.4 B. Stiavnica.          7.7         8.8          17.3    22.8
   I'The fact that unemployedyouth are no longercountedindividually ifthey are still livingwith their
  parentsis a factor inthis reductionas well.

Chapter 4                                                                                            57



4.15      An importantpart ofthenewsocialpolicyisthe introductionofanumberofactive labor
marketpolicies:l designedto helpmatchjobs andjob-seeker~~~ to support the income o fthe
                                                                       and
poorwhile inwork. These policies, which are summarizedinTable4.4, vary intheir character and
inthegoupthattheyaredesignedtotarget.Thebroadestareopento alloftheunemployedwhile
others are targetedto the long-termunemployed, andto youth, parents andthe disabled.

Table 4.4: Focus ofActive Labor Market Policies in 2004 and 2002
Contribution                          2004                                   2002           Beneficiary
For commuting expansesrelatedto
job interview                         All unemployed                         NIA            Unemployed

For labor marketpreparationand        Re-trainingofpreviouslyunemployed      Allowance
trainingofjob seekers                 person                                 for re-trainingEmployer

For self-employment                   Allunemployedfor morethan4 months,     All
                                      who start andmaintaintheir ownbusiness unemployed     Unemployed

Forjob mobility                       All unemployedfor more than6 months NIA               Unemployed
 Activation (subsidy)                 All unemployedinpoverty                NIA            Employer

 For graduatepractice                 Youthbellow25                          Similar
                                                                             schemeexisted  Both

                                                                             Long-term
 For employment ofdisadvantaged        Long-termunemployed,youthor elderly,  unemployed,
job seeker                             singlepersontakingcare of child,                     Employer
                                       disabledperson                        elderlyand
                                                                             graduate
 For child-carecosts ofjob seeker
 attendinglabor marketpreparation      Unemployedparentundertakinga          NIA            Unemployed
 andtraining                           training
            -
 For establishingandmaintainingofa Disabled
 shelteredworkplace andworkshop                                              Disabled       Employer
 For self-employmentofdisabledjob Disabledunemployedfor morethan4
 seeker                                months                                NIA            Unemployed
 For productioncosts ofsheltered
 workplace andworkshop andfor
 transport-relatedcosts ofdisabledjob  Disabled                              NIA            Employer

 seekers
 For anassistantat work for disabled
 employee(s)                           Disabled                              NIA            Employer

Note: For active labor market policies in 2002 see also Act No. 38711996 on Employment.NIA stands for not
available in2002.


4.16 Intotal, Sk 1.9 billionwas spent on active labor market policies in2004 (as opposedto
Sk 3.5 billion in2002), and270,000 people took advantage of them (figurenot available for
2002).43The mainpart of the expenditure was usedfor activation contributions: 82 percent of


    Act No, 512004 on EmploymentServicesandAct No. 38711996 onEmploymentServices.
    Ajob-seeker is apersonobligedto visit the labor office at the following frequencies: once aweek
    iflong-termunemployed, once amonthifparticipatinginactivationprograms, or once every two
    weeks otherwise (Act No. 512004, Art. 8).
    The comparativelylow spendingonALMPs in2004 couldbeexplainedby agradualphasinginof
    the newpolicies:the activationsubsidy was implementedfrom February, while other policieswere
    implemented from Marchto May. Thus, while inthe first half of the year only Sk 0.4 billion was
    spent, in the secondhalf Sk 1.5 billionwas spent. Using the latter figure, we estimate a potential
    cost of Sk 3.0 billion in2004, which is asomewhathigher thanthe Sk 2.8 billion spent in2003.

58                                                                                                                The Questfor Equitable Growth in the SlovakRepublic



all beneficiaries of the active labor market policiesparticipated (almost 220,000 people, or 46
percent of all unemployed in 2004) and 53 percent of total expenditure went to this benefit
(see Figure 4.3). Contributions for the child-care costs ofjob-seekers attending labor market
preparation and training were paid to 10 percent of the beneficiaries and accounted for 11
percent of total expenditure. Contributions for graduatepractice were paid to 5 percent of the
beneficiaries andcost 11percent.The maindisproportion seems to bebetweenthe percentageof
beneficiaries ofthe contributionfor self-employment (2 percent) andits highcost (17 percent).
All other active labor market policies benefited only 1percent of the beneficiaries and cost 7
percent oftotal funds (for more details see Policy Note 2 inVolume 11).                                                                                     -
             Figure 4.3: Active Labor Market Policies in 2004: Beneficiaries and Costs


           Numberof beneficiariesand costs (SK million)

  2 so ,0110                                 -No.of                  participants (LH-y)                     1.200         90 -
                                                                                                                           a o .                             B I A L M P participants

  200.000                                   - C o s t               (S k m , R H -y)                         1,000                                           O%ALM P T o t a t c o s t
                                                                                                                           70  .
                                                                                                             800           a 0 .
   150 .oo 0                                                                                                                50 -
                                                                                                             6 0 0
   100.000                                                                                                                 40 -
                                                                                                             4 0 0         3 0 -

    50.000                                                                                                                 2 0 .
                                                                                                             200            10 .
           0                                                                                                 0               O  r
              ACtlvsliln         Contribution Canlribvlian Canlribulion                           Olhei
                subsidy             f o r sell-       lorchild.care        foigraduile
                                 emplaymant            C l l l l i OfJob     PtICtiCl
                                                            seeker


    Numberof beneficiaries,cumulativemonthly(persons)                                                                      Cost of the programs,cumulativemonthly (SK million)
            -9-5       4 6 C o n t r l b ~ t i o nf o r c h i l d - c a r e c o s l s o f j o b s e e k e r                       -8-5    46 contribution forchild-care c o s t s o f job s e e k e r
            -6         B I C o n t r i b u l i o n f o r g r a d u a t e o r a c t i c e                                          -5      BlContribution torgraduate practice
            -5         49 Contribution f o r s e l f - e m p l o y m e n t                                                        -5      49 Contribulio n tor self-em ployment
  35J00     -Other                                                                                                                -Other
  30,000   ]                                                                                                400,000       '0°

            - ~ 5 Z A c I ~ v r t i o n r ~ b r i d y ( R H S ) ~ ~ ~ 6~0 0~ ~ -~$ 5~2 A~c t 1 v a t i o n s u b s i d y ( R H S )

                                                                                                                          50 0                                                                          1.400

                                                                                                                          400                                                                           1.200
                                                                                                                                                                                                        1.00 0
                                                                                                                          300                                                                           a o o

  10,000                                                                                                    100 .o 0 0    200                                                                           S O 0
                                                                                                                                                                                                       400
   5,000                                                                                                    50,ooo        100                                                                           200
        n       - -                                                                                         0                0                                                                          0
          Fab23   Apr29         Jan25        Aug21          00129        Dec31       Fa125 Apf25                              Feb23  Apr29   JunZS  A u g Z l OclZ9    Dee31     Fnb2S      AQIZS


                     Average cost per beneficiary(SK)
                 - § 5 2 A c t i v a l i o n s u b s i d y
                 & $ 4 6 C o n t r i b u t i o n              forchild-care c o s t s o t j o b s e e k e r
                 -5             SlContrlbulion forgraduate practice
   90.000        -A-$           49 C o ntributio n for self-em PICym en1
   80.000        &OIher
   70,000
   60,000                                                                                                                 Average cost per beneficiary, note:
   50.000                                                                                                                 because different programs                          imply different
   40,000
   30,000                                                                                                                 duration o f contributions, the average cost per
   20,000                                                                                                                 beneficiary providesbiasedpicture. Unfortunately,
   10,000                                                                                                                 information on the duration o f each program was
          0                                                                                                               not available.



                                                 Source: Labor Office, Ministry o fLabor and authors' calculations.

ChaDter 4                                                                                      59


         C. HAVE      THE REFORMS       REDUCED THE        INCOME      OF THE   POOR?

4.17 Earlier parts of this chapter outlined clearly the need for the reform and its rationale.
However, as mentioned earlier, when the reforms of the tax and benefits systems were
introduced in January 2004, the SR saw considerable social unrest which received wide
media coverage bothnationally and internationally. The source o f the unrest was the concern
by some groups that the reforms had shiftedresources away from the most vulnerable and
towards groups of individualsthat, while poor, were not as socially excluded as those that lost
out. This concern was particularly vivid among the long-term unemployed and in the most
vulnerable communities but was also sharedby other parts o f the society (see Policy Note 2,
Volume 11).

4.18 On the surface this appears to be a somewhat justified concern, as the basic benefits
available to those families that traditionally had been most dependent on the benefits had
declined substantially. However, this is an excessively simplified conclusion that ignores the
complexity of the reform and may fail to quantify its full impact. Thus, evaluating the full
impact of the changes i s clearly o f critical importance to the sustainability o f the reform
process, which could potentially bejeopardized by the risk of social unrest. Unfortunately,
the limitations of the data - as highlighted extensively in Chapter 5 - make this a rather
difficult task. Indeed, detailed information on individual and household income after the
reform i s not available and the informationthat i s available for the pre-reform period (ie., the
2003 Microcensus) suffers from serious limitations insample design and responserates (see
Chapter 5 for more details).

4.19 This chapter attempts to fill this gap usingtwo different quantitative methodologies.
The first uses purposely built software to compute the total net income corresponding to
different levels o f gross earnings for a number of household typologies before and after the
reform. By comparing the net income which corresponds to a given level o f earning pre-
reform with its post-reform counterpart, the model quantifies the net effect of the reform for
a selectednumber o f household typologies. The final part o f the chapter uses a standard tax-
benefitmodel to quantify the effect o fthis impact onpoverty and income inequality.

HouseholdTypologies
4.20 A number of different household typologies are analyzed. These can be summarized as
follows:
       0   Single headedhousehold
           o with or without employment-ifwithout, then long-term unemployed
           o with different numbers o f children o fdifferent ages.
       0   Couple
           o wherebothareinemployment/onlyoneisinemploymenthone areinemployment
              (then long-term unemployed)
           o with differentnumbersof children o f different ages

60                                                   The Questfor Equitable Growth in the Slovak Republic



Definitionof Income andModelSpecification
4.21 For the purposeof this model, net income is definedas original income(without pensions)
minuspersonalincometax (PIT) andsocialcontributionsplussocialbenefits. The modeldoesnot
include pensions, sickness and maternity allowances, and unemploymentbenefits (as it does not
consider the short-termunemployed).44It is, however, designedto capture differencesbetweenthe
two schemesofsocialbenefitsthatexistedin2002: that for individuals whowerepoor for subjective
reasons andthat for those who were poor for objective                        The main structureof the model
is defined in Table 4.5. The focus is on households whose m u a l gross wage is ina range of Sk
72,00046 to Sk 240,000, which-accordingto Microcensus2003 -covers 61percentof all Slovak
households(see PolicyNote2 inVolume II).

Table 4.5: Assumptions ofthe BenefitImpactModel
 Contribution              Assumptions
                           2004: includes child benefits and parental allowance (two amounts considered, de-
                           pending onjob status o f the parent)
 Family benefits           2002: includes child benefits and parental allowance
                           Parental allowance is received when child aged 0-3 years i s not inkindergarten
                           2004: includes social assistancebenefit, activation, health care and housing allow-
                           ances
                           2002: includes social assistance benefit
                           Householdeligible when inmaterial need, eligible for activation allowance when
                           without job, eligible for housing and health-care allowances when inmaterial need
 Social assistance benefit Note that while housing allowance in2004 was a part o f the social assistance
                           benefits, in 2002 the housingbenefit was considered independently o f the social
                           assistance benefits. However, duringthe transition periodo f 2004, those eligible for
                           the housing benefit inDecember 2003 continued receiving the same amount in2004,
                           providedtheir net income was below the MSLand the total sum of social assistance
                           benefits equaled zero.
                           2004: includes food subsidy, school needs subsidy and average scholarship (which is
                           basedon grades, otherwise)
 School assistance         2002: NIA
                           Household eligible when inmaterial need and only iftotal social assistancebenefits
                           paid are greater than zero
 Tax + Tax credit          Tax credit applicable only for 2004
 Pensions                  N o pensions assumed, no pensioners inthe household modeled
 Sickness and maternity    No sickness and maternity allowances modeled
 Unemployment benefits No short-term unemployment modeled, thus no unemployment benefits considered




44  The model has been sharedwith the Ministry ofLabor and the Ministryof Finance for comments,
   butalso to serve as a simple tool for simulating certain changes inthe socialpolicy.
45  See Footnote 33, above, and Policy Note 2 inVolume I1for definitions and more details on these
    categories.
46  That is, the annualminimumwage inSeptember 2004 was Sk 72,960 andthe average wage in2004
   was Sk 189,900.

Chavter 4                                                                                   61


WhoAre theWinners andWhoAre the Losers?
4.22 The findings o f the analysis of the net impact of the reforms on selected household
typologies (see Figures4.4 (a) and 4.4 (b)) conclude that:


      0   The combined effect o f the reform on the tax and benefitregimehas increased the
          disposable income of the majority o f households that are entitled to benefits
      0   The only households that may have lost out as aneffect o f the reformare those with
          three or more children where all adults are unemployed and those that before the
          reform were entitled to the higher benefit schedule
      0   Whether the householdshave lostout dependspredominantly ontheir entitlementto
          school assistance
      0   Evenfor the most affected households the maximum potential loss in disposable
          income i s limited.

4.23 These conclusions are confirmed by the findings of the tax-benefit model, which
concludes that the reform has only a small potential effect on the incidence o f poverty, with
the poverty headcount declining by 2 to 3 percent according to the simulations based on
the 2003 Microcensus. This is the case despite an overall reduction of around 16 percent in
transfers paid. What this implies i s that the change in the tax-benefit regime in itself (i.e.,
inisolation from the reduction intransfers) would have significantly reduced poverty by as
much as 17 percent had it not been accompanied by a simultaneous reduction inthe average
transfer.

62                                                                      The Questfor Equitable Growth in the Slovak Republic



                Figure 4.4(a): Examples of Results of Simulations, 2004 versus 2002:
                                        InMaterialNeedfor Subjective Reasons
I
                                                       Single household, withjob, 3 children
                   Net income in 2004 and 2002                                 Netgain in income in 2004 by different type of income(%) I/
                                                                                25
                                                                                20
                                                                                 15
                                                                                 IO
                                                                                  5
                                                                                  0
                                                                                  5
                                                                               -10
                                                                               -15
                                                                               .2 0                                                                    -
                                                                               -2 5
                                                                               -30                                       AnnualGart Wa(e(Sk)

                                                                                    72.000         96,000         120,000    164.000        166,000    192,000    216.000            240,000


            72,ooo 96,ooo  QO,OOO  u4,ooo   BB,OOO 182,000   2m.000 24o.000           U a 3 n F i m I l y benellIr       n S C h O O l a m s l m o            Sailal a i i l i l i n ~ e
                                                                                      m H I u l i n g bmelil             ~ C o n l t i b u l i o n l    0 T . X   * Tax   madit
                               AnnualGtoas W a p (Sk)                                 -Nil           Income


                                                          A couple, bothwith job, 2 children


     290,000 ,     Net income in 2004 and 2002                                 Net gain in incomein 2004 by different type of income(%) I/
                                                                                 6.0 1




                                                                                      164,000        88,000        192.000    216,000         240.000  264.000    268,000             312,000




                                      A couple, one with job, other long-term unemployed, 4 children
                   Net income in 2004 and 2002                                 Net gain in incomein 2004 by different type of income (%) I/
       260 000                                                                  60 0
                      -Net    Incame 04  -Ne1      Incoma 02                    50 0
       240 000
  E                                                                             40 0
  L?.  220000                                                                    30 0

  -c8
   E                                                                             20 0
       200000                                                                     I O 0
        i50000                                                                     0 0

   =                                                                            -10 0
        180000                                                                   20 0

        160 000                                                                  30 0
                                                                                       n o a a          96000      uoooo       n 4 o o o       m6ooo    192000     2mooo               2aoooo




1/ Percentage contribution ofparticular income to the net gain intotal income.
Source:Authors' simulations.

Chapter 4                                                                                                                                                           63


                Figure4.4@): Examplesof Resultsof Simulations, 2004 versus 2002:
                                           InMaterialNeedfor Objective Reasons


                                                            Single household,withjob, 3 children
                      Net income in 2004 and 2002                                Net gain in income in 2004 by differenttype of income (%) I/
     260,000 7                                                                    30.0 7

                                                                                  20.0
 5e  220,000  -                                                                    10.0

     200,000 -                                                                      0.0
 20
     ra0.000
 5
 5
     180,000  --                                                                 .20.0
  J                                                                              -301


     t2o.ooo  -                                                                  40.0
                                                                                        7z.ooo  86,000   i2o.000   n4.ooo    !68.ooo      192.000 218.000    z ~ o . 0 0 0




                                                             A couple,bothwithjob, 2 children


     290,000  ,       Net income in 2004 and 2002                                Net gain in income in 2004 by different type of income (%) I/


     270,000                                                                        6 .O
                  -Nal    lncoms'04 -Ne1     IncOma'OZ
 5E8
 -                                                                                  4.0
     250,000  -                                                                     2 .o
     230,000  -                                                                     0 .o
                                                                                   .2.0
 5Iim
 5 28.000 -
                                                                                  4.0
      80.000  .                                                                   -6.0
                                                                                   4.0
                                                                                  .O.O                           AnnualGlosr W a w I S k l

                                                                                        144.000 166.000  192,000   28,000    240.000      264.000 268,000     312,000




                                 AnnualGrors W l g e ( S k )

                                       A couple, one withjob, otherlong-termunemployed,4 children
                      Net income in 2004 and 2002                                 Net gain in incomein2004 by differenttype of income(%) I/
        260,000                                                                    30

        250.000
  a                                                                                20

  v,    240.000
  5                                                                                 8
  ~     230.000

  =I    220,000                                                                      0
         28,000
  -                                                                                .a
  3     200,000

         190.000                                                                  -20
  2      180,000                                                                  4 0                            AnnualGrosr Wage(Sk1
         i70.000                                                                        7zooo   96ooo    ~ O O O O n4ooo     1 8 0 0 0    82000    2 1 0 0 0   ZPOOO
         80,000




li Percentagecontribution ofparticular incometo the net gainintotal income.
Source:Authors' simulations.



       D. WHATHAS                        HAPPENED                  TO  POVERTY INCOME       AND                         DISTRIBUTION?

4.24 A different approach to evaluating the impact of the reform is to use a "Tax-Benefit
Model", which compares the household income levels before the reform with the simulated

64                                         The Questfor Equitable Growth in the Slovak Republic



post-reform levels. Then it becomes possible to identify the gainers and losers as well as all
o f the distributional andpoverty changes (Orcutt, 1957). The use of a static microsimulation
model makes it possible to focus on the effect of specific policy changes on poverty and
income distributionwhile maintaining all of the other aspects of the reality constant.

ModelDescription
4.25 The framework usedconsistso ftwo stages.Thefirst involves computingthe PITdueby
eachindividualandhisherbenefit entitlement onthebasisofpre-reformtax benefit schedules.
Inthe second stage, reforms are simulated by changing the parametersof the model and its
specification as requiredand recomputing the entitlement /payment. The model assumes no
tax evasion, full take-up ofbenefits and inthe first instancenobehavioralresponses(for more
details on these assumptions and other technical issues, see Policy Note 2 inVolume 11).

4.26 Formally, the model structure for individual i i s givenby:




       where y     i s the disposable income, y     i s the gross income, t 0") i s the tax rate
(depending on gross income) andB, @,a)are income from differentkindso fbenefits that can
be universal (Bo)or means-tested. The latter type of benefit varies by income type (y) and
socio-demographic characteristics (a) relevant for eligibility (for further details, see Policy
Note 2, Volume 11).

4.27 Owing to data limitations it was impossible to simulate every fiscal and benefit
instrumentof the Slovak taxes andbenefits system. For example, the unemploymentbenefits,
maternitybenefitand sicknessbenefit, as well as specific deductions suchasthose for disabled
pensioners, handicapped tax payers and disabled dependent children, were not modeled
becauseo fthe lack of information onrelevant variables. The policies simulated inthe model
are:


        0  Child Benefits
        0  Scholarships
        0  ParentalAllowance
        0  Social Assistance Benefits (for Households inMaterialNeed)
        0  Personal Income Tax.

Data and Empirical Specifications
4.28 The simulation relies on micro-data collected at the individual level. The data set
used is Microcensus 2003. It is a sample extracted from Census 2001 and representative
at the national level. The sample size i s 19,000 households (approximately 1 percent of all
Slovak households) and 61,6 11individuals. The data set contains individual and household
characteristics as well as gross wage, and income from capital, pensions and benefits. All
incomes are yearly (they refer to fiscal year 2002) and inSlovak koruna.

ChaDteer 4                                                                                             65


4.29 In calculating the poverty and inequality indicators, the unit of observation i s the
individual, and, when not otherwise specified, the StandardOECD Equivalence Scale is used
to assign household income to each member.47Relative as well as absolute poverty lines are
used: a commonly used relative line equal to 60 percent of the median per equivalent adult
income (43,422 SWyear); and three absolute poverty lines (US$2.15 PPP per day [10,776
SWyear], US$4.30 PPP per day [21,552 SWyear], and the nationally set Minimum
Subsistence Level [MSL]).48

4.30 The poverty indicators used include: (i)the poverty rate (or Head-Count ratio); (ii)
the income gap ratio; and (iii)the Foster-Greer-Thorbecke (FGT) index o f the severity of
poverty. The income quintile ratio (S80/S20) and the Gini coefficient are usedas indicators of
income inequality andthe Kakwaniand Reynolds-Smolensky indices areusedto evaluate the
redistributive effect o f taxes and benefits pre- and p o ~ t - r e f o m . ~ ~

Evaluatingthe RedistributionalImpactof Implementingthe FlatTax
4.31 While the potentialbenefit o fimplementingaflat tax onthe efficiency o fthe tax system
is well knownand widely recognized, many are concerned about the redistributionaleffect of
movingfrom a traditionalPITsystemto onebasedonwhat i sbasically aproportionaltax.The
assumption here i s that the reductionintax brackets and the abolition o fhighertax rateswill
reduce the progressivity o fthe system-Le., the well-accepted principle that "equity" requires
the better-offto pay a higherproportionoftheir income intaxes than the poor.

4.32 However, simulation o f the distributional impact of the move to the flat tax inthe SR
show that it i s possible to actually increasethe progressivity ofthe tax systemwhile replacing
five tax brackets (and a fairly steep tax structure) bya singleone. This is clear fromFigure4.5
which shows that (i)   in2004 thePITburdenwas reducedfor everybody-except avery small
rangeofincome-but (ii)       particularly for the households on lower incomes. This i s due to the
combined effect o f an increase inthe value and a change inthe structure o f deductible items
and tax credits.




47 Thisscaleassignsavalue of 1.O to thefirst adult, 0.5to childrenbelow 18, and0.7 to other household
   members. To investigate the sensitiveness o f the results to the equivalence scale used some results
   are also presentedusingthe Modified Equivalence Scale or per capita values. The Modified OECD
   Equivalence Scale assigns a value o f 1.0 to the first adult, o f 0.3 to children below 14 and of 0.5 to
   other householdmembers.
48 The MSL differs by family structure with the lowest level equal to 47,160 S W y e a r per single
   adult.
49 See Policy Note 2 inVolume I1for more details.

66                                                     The Questfor Equitable Growth in the Slovak Republic



             Figure 4.5: Effective PersonalIncome Tax Rates in2004 and 2002
                                         for a SingleWage-earner




                                  72,000 144,000 2B,000 288,000 360,000 432,000 504,000 576,000
                                                    Annual gross w age (Sk)

                             Effective tax calculated as ratio of tax to gross income.
                                          Source: Authors' calculations.

4.33 An alternative way of analyzing the change i s presented in Table 4.6 which shows
the impact of personal income tax reform on the tax burden and net income for different
income decciles. On average, the change inthe tax structure, ceteris paribus, has reducedthe
tax burden on households by 37 percent but the reduction was as highas 80 percent for the
households inthe first decile of original income, by around 60 percent for those inthe second
andthirddeciles andby only 20 percent for the upper deciles. This confirms the conclusionof
Figure 4.5 that the poorest households benefitedmost from the tax reform.

Table 4.6: Simulated Personal Income Tax by Income Deciles (yearly SKK), 2002
and 2004
 Inc,omedeciles'     ,                                                          Tax         ,  ,



                                        ,  ,2002                               2004, .             % Change
 I                                            16                                  3                 82.00
 I1                                          299                                 117                 60.86
 I11                                         820                                393                  52.01
 IV                                        1,204                                726                  39.64
 V                                         1,42 1                              1,042                 26.67
 VI                                        1,420                               1,108                 21.96
 VI1                                       1,603                               1,270                 20.76
 VI11                                      2,296                               1,858                 19.08
 IX                                        4,324                               3,538                 18.17
 X                                        13,816                               10,006                27.58
 Average                                   2,722                               2,006                 36.87
Note: Taxes are per capita.
* Income deciles are built using equivalizedgross income.
Source: Microcensus2003.


4.34 According to the findings inTable 4.7, the Slovak tax system was slightly progressive
before the reformandhasbecomemoreprogressive after 2004. Thisi sreflected inthe increase
inthe Kakwaniindexfrom 0.154 to 0.192 percent. Similarly, the Reynolds-Smolensky index
showsthat the mildredistributive impacto f2002 hasgrownstronger in2004 after the reforms.
This was due to the effect o f the increase in deductions and credits which has more than

Chapter 4                                                                                   67


compensated for the reduction inprogressivity associatedwith the move to a single tax rate.
The Kakwaniindex is halved or doubledifwe consider separatelythe effect onprogressivity
of reducing five tax rates to only one rate of 19percent or of introducing the new deductions
combined with the tax schedule of 2002. Inthe first case, the progressivity index falls from
0.114 to 0.060, while inthe secondcase the index increasesup to 0.225.

Table 4.7: Progressivityand RedistributiveImpactof the Tax System, 2002 and 2004
                                                                   2002            2004
 Gini coefficient of original income                              0.574            0.574
Concentration coefficient for taxes                               0.688            0.766
 Concentrationcoefficient for net income                           0.557           0.556
Kakwani index                                                      0.114           0.192
Reynolds-Smolensky index                                           0.017           0.018
Note: Households.
Source: Microcensus 2003.


4.35 However, before deriving strong conclusions on the distributional impact o f the tax
reform, it i s important to rememberthat this analysis focuses exclusively on the new PITand
thus does not evaluatethe full impact ofthe reforms on the overall tax system.The concerni s
that the overall reductioninthe tax revenue generatedby the newPIToverall may have ledto
a shift o f the tax burden away from the PIT and into taxes on consumption (VAT and excise
taxes), which tend to be more regressive.

Have BenefitsBecomeLess Pro-poor?
4.36 As emphasizedabove, the 2004 reformsbroughta generalized reductioninvalue ofthe
average transfer. According to these simulations the reduction was around 35 percent for the
child benefits and 51percent for social assistancebenefits and scholarships (see Table 4.8).
Inaddition, inthe case of social assistancebenefits (and scholarships) the greatest reduction
i s found at the top of the income distribution (more than 50 percent), while inthe case of the
childbenefitthe greatest fall i s for the bottom ofthe income distribution. However, this effect
i s compensatedby the introductiono f a tax credit for parents (as described above).

68                                            The Quest for Equitable Growth in the Slovak Republic



Table 4.8: Benefits by Income Deciles (yearly SKK), 2002 and 2004

  Income                                                                 SocialAssistanceBenefit +
 Deciles*              Child Benefit               ParentalAllowance            Scholarship
            I

               2002      2004       Change
                                     Percent     2002    2004   Change
                                                                percent   2002       2004  Change
                                                                                            percent
                                                            , ,            ,   ,    , , , ,

     I 12,982            7,531         -42       5,319   2,915    -0.45   58,025   36,829    -37
    I1        15,956     9,398         -41       5,012   4,785    -0.05   27,436   14,577    -47
    I11        11,698    6,924         -41       1,171   1,892    0.62    9,062     4,366    -52
    IV         8,997     5,428         -40       947     1,400    0.48    5,502     2,750    -50
    V          6,594     4,098         -3 8      499     1,089    1.18    2,606      1,217   -53
    VI         5,777     3,620         -37       705     700      -0.01   1,092       563    -48
    VI1        6,013     3,925         -35       741     822      0.11    1,259       666    -47
   VI11        5,955     4,110         -3 1      327     710      1.17     561         198   -65
    IX         5,340     3,981         -25       396     660      0.67     410         158   -61




andof0.7 to other householdmembers.
*Incomedeciles are builtusingequivalizedgrossincome,
Source: Microcensus 2003.

4.37 The overall benefit system, as well as its components, has become slightly less pro-
poor. This conclusionis confirmedwhen comparing indices o fredistributionsfromthe benefit
systembefore andafter the reform (Table 4.9). The redistributiveimpact o fbenefits, already
low in 2002, i s likely to have been further reduced by the 2004 reforms (the Reynolds-
Smolensky indexwas 0.121 in2002 and 0.089 in2004).

Table 4.9: Impact of Benefits on Income Inequality, 2002 and 2004 (percent)

                               NetIncome   + Child    NetIncome+Parental      Net Income Social
                                                                                           +
                                    Allowance              Allowance         Assistance Benefit and
                                                                                  Scholarship

 Measures                      2002         2004       2002         2004        2002        2004
 Giniindex                     36.6          37.4       37.4         37.8       33.1        33.9
 Standarddeviationof logs       80.8         89.9       69.3         71.8       60.7        63.9
 Theil entropy measure         23.4          24.8       25.3         25.9       18.6        19.6.
Note: Householdsdisposable income,
Source:Microcensus 2003.

Evaluating the Overall Impact of the Reform on Poverty
4.38 If the impact o f the overall reform i s evaluated-rather        than evaluating the tax and
benefit systemseparately-the     implementationof the new regime may have increasedpoverty
independentlyof theindicators andthe povertylineused.Inthe absenceo fbehavioralchanges,
the results of the simulations reported inTable 4.10 conclude that the regime introduced in
2004 has probably resultedina little change inthe occurrence o fpoverty, with the headcount
declining by 2 to 3 percent for households and increasing by 3 percent for individuals when
relative poverty i s considered and when the absolute poverty line i s the MSL.However, the

Chapter 4                                                                                   69


percentage change i s larger when extreme poverty i s considered (US$2.15 PPP), partly as a
reflection o f the small headcount for this type of poverty.

Table 4.10: Impact of Reforms on Poverty Indicators (% increase)

                                                                     Changein Headcount

Poverty Line                                                        HH              Indiv.
60 percent medianequivalent income                                   -2               3
M S L                                                                -3               3
US$2.15 PPPper personper day                                         25              20
Source:Microcensus 2003.


4.39 Evaluating the simultaneous effects o f tax and benefit reforms on income distribution,
we can conclude that the reformsundertaken in2004 may have increasedinequalitybut only
slightly (the Ginicoefficient calculated on disposable income rose from 0.3 17to 0.329). The
decrease in the pro-poor impact of benefits and the loss of progressivity due to the tax rate
partially compensated for the introduction of the tax bonus and higher deductible amounts
from taxable income.

How Muchof the IncreaseinPoverty I s Due to Changes inthe Structureof the System?
4.40 According to the simulations, the average transfer paid to individuals dropped by 16
percent between 2002 and 2004. However, as noted earlier, this resulted in a small decline
inthe poverty headcount (3 percent). This suggests that the change in regime in itself has
increased the targeting o f the system and has protectedthe living standards o f the poor from
the potential effects o fthe decrease intransfers which was dictated by the needto increasethe
fiscal sustainability o f the system.

But the Objective of the Reformis to Alter Labor Market Outcomes
4.41 Moreover, the modelsonwhichthese results arebasedare static andassumenodynamic
behavioral change (in other words, they assume that the employment status of the poor did
not change after the reform). This i s clearly a strong assumption, especially inthe context of
a reform that was designed to remove disincentives to work and to make work pay. Inthe
absence o f longitudinal data that can provide information on behavioral changes o f the type
the reform i s trying to encourage, a speculative assumption can be made to try to incorporate
the potential behavioralresponsesinthe simulation o f the impact of the reform.

4.42 One o f these assumptions i s the possibility that the new policies succeeded inhaving
one extra person per each poor household (with more than one adult) in employment. The
results o f this exercise are presented in Table 4.12. The good news i s that, if this should
be the case, the poverty incidence would be reduced drastically-with        the headcount ratio
dropping from 16.8 to 3.1-and      the poverty gap andFGTindex wouldbecome close to zero.
Unfortunately, this would require the creationo f over 75,000 newjobs that would require the
skill levels of the household members who are currently unemployed.

70                                          The Questfor Equitable Growth in the Slovak Republic




                                             2004 Regime (No        2004 Regime (Allowing for
                                          Behavioral Changes)          Behavioral Changes
 Povertyrate                                       16.8                        3.1
 Povertygap                                         5.5                        0.7
 Foster-Greer-Thorbecke(FGT) index(a=2)             2.9                        0.3
 Number of additionaljobs required                                           74,878




4.43 This i s a challenging but not impossible task. According to data from the Labor Force
Survey (LFS), in the second and third quarters o f 2004 the net increase in employment
was 72,000 jobs. However, given the current patterns o f job creation-as      evident from the
enterprise survey-only a small proportion of the newly created employment opportunities
were for the types of workers who are most likely to be inpoverty (Le., the low-skilled in
lagging regions). Thus, reducing labor market mismatches i s of great importance to making
the current growth and employment strategy aneffective poverty reduction strategy as well.

4.44 A more optimistic way of looking at this result is to compute the minimumincrease in
employmentrequiredto maintainthe poverty headcount at the simulated levelfor 2002, given
the new benefit level (Le., for the reform to be poverty neutral). Ifwe assume no structural
unemployment and no regional or skill bottlenecks, the economy would need to generate
around 15,700 new jobs for the reform to be poverty neutral. This i s less than one-third of
the net increase inemployment experienced inthe third quarter of 2004 andjust over halfo f
that inthe second quarter. Thus, even allowing for structuralunemployment, this should not
be an impossible target, especially ifthe targeted ALMPs implementedas part of the reform
were to prove successful in reducing labor market mismatches and promoting employment
opportunities for the most vulnerable groups.5o




50 Early empirical evidence is encouraging: according to the Labor Office, subsidies for the
   employment of disadvantaged job seekers (which includes the long-term unemployed) made
   it possible to generate 1,800 jobs in 2004 and 1,900 jobs in 1H2005. Contribution for self-
   employment (for the unemployed for more than 4 months) made it possible to generate 5,618
  jobs in 2004 and 5,323 jobs in 1H2005. Furthermore, according to the MinistryofLabor, Social
   Affairs and Family records, since February 2004, 20,000 previously long-term unemployed
   persons have become eligible for activation Gob-start) allowance by finding ajob or becoming
   self-employed (see law no.599/2003, art. 12(8-9)).

                                 CHAPTER 5 -

            WHY ITREMAINS IMPOSSIBLETO
      ADEQUATELYMONITORPOVERTYAND
      INEQUALITY INTHE SLOVAK REPUBLIC


5.1   In evaluating the impact of the reforms, it is important to keep in mind that the
quantitative information on poverty and living standardsinthe SR i s limitedandunreliable.

5.2   Inthe past there has been some reluctance to acknowledge the existence ofpoverty
inthe SR.The Slovaklegislationneverusestheterm"poverty" andofficialstatistics do not
record basic poverty indicators. Synonyms for poverty - such as "socially underprivileged
population," "low-income households" and "material distress" -have typically beenused.
However, recent years have seen a sharp change inattitude, with the government explicitly
committing itself to fight poverty and social exclusion ina number o f strategic documents
linkedto entry into the EU, such as the Joint Inclusion Memorandum (JIM) (2003) and the
National Action Plan on Social Inclusion (NAP) (2004). Consequently, research onpoverty
inthe SR is relatively new, although there are now a numberofstudies onthis topic.


    A. DATA     LIMITATIONS     PREVENT ACCURATEPROFILEOF POVERTY
                                            AN


5.3   Untilthe releaseof the Microcensusdata last summer, the last quantitative indicators
referredto 1996. Then, accordingto World Bank figures--withaheadcount at 8.6 percent and
a US$4.30 ppplday poverty line - the SR had a comparatively low poverty rate, but it had
a comparatively high incidence (2.6 percent) when the poverty line was lowered to that o f
extreme poverty (US$2.15 ppp/day).S*This suggests the existence o frelatively small butvery
deeppockets o fpoverty compared to other EU8 countries. With a Gini coefficient o f0.250 in
1996, the SR also hadavery low levelo fincome inequality, slightly higher than, for example,
Denmark (Gini=0.227) but lower than Spain (0.340), Greece (0.35 l),     or Portugal (0.368).


   World Bank (2000), Making Transition Workfor Everyone: Poverty and Inequality in Europe and
   Central Asia.

72                                           The Questfor Equitable Growth in the Slovak Republic


However, inequalities increased in the first half o f the 1990s with a rise o f over 35 percent
since 1992, when the Luxembourg Income Study calculated an extremely low Gini of 0.187.

5.4    As emphasized inthe previous chapter, the concern o f the critics o f the reform is that
the process of economic liberalization and widespread reforms, while fostering GDP growth
and average household consumption, may have further increasedbothpoverty and inequality
over recent years. The initial rationale for this study was to verify the empirical relevance of
this claim by quantifying and analyzing the changes in the levels of poverty and inequality
since the last study. Unfortunately, the lack of reliable, nationally representative data with
which to calculatethe basic indicators made it impossible to derive aprofile o fpoverty which
was multidimensional in nature, representative of the Slovak population and in anyway
comparable with that emerging from the 1996Microcensus.

5.5    The limitations o f the data are neither a new development nor a new discovery. The
World Bank (2002) report pointed out at some length andmade extremely clear the fact that
the figures obtainedfor the basic indicators were unlikelyto be totally reliable and should be
treatedwith caution. Nevertheless, these figures are still widely usedfor lacko fbetteroptions.
The risk was that any further numbers generatedby this report would also be usedwidely as
a quantitative descriptiono f the current living standards inthe SR. Inaddition - andperhaps
more worrying-the comparison betweenthe values o fthe indicators in 1996and2002 would
be usedto make comparisons over time. This would clearly be highly undesirable since (i)
neither the 1996 nor the 2002 data arerepresentative ofthe Slovak populationand their living
standards; (ii) lacko frepresentationdoes notarise from the same limitations ofthe sample
               the
designandis unlikely to affecttheresultinthe same directions; and(iii) itisnotonly the
                                                                           thus,
absolute levelof the indicatorsthat i s likely to bebiasedbut also the magnitudeand direction
o f any change that occurred between the two periods.

5.6    Inorder to avoidthistype ofmisunderstanding, thisreportdoesnotattempttoprovidea
profile ofpoverty and living standards inthe post-accession Slovak Republic and it definitely
does not intendto fully assesshow welfare distribution has changed inthe "new" SR. Rather,
it focuses on buildingthe capacity to appropriately monitor poverty and living standards in
the future so that:


       0  The impact o f the continuingprocess o f transformation can be properly evaluated
       0  The findings can feed into the policymaking o f a government that i s extremely
          conscious o f the importance o f monitoring its performance in a number o f
          directions
       0  The requirements o fmonitoring employment and social exclusion under the Lisbon
          agenda can be adequately satisfied.

              B. ASSESSINGTHE CURRENT CAPACITY TO MONITOR
                             TkENDS IN LIVING STANDARDS


5.7    The SR has the basic framework of surveys needed to monitor poverty and social
exclusion, to evaluate the impact o f economic reforms and to deliver evidence-based

Chapter 5                                                                                    73


policymaking. However, the execution o f the surveys within this framework shows varying
degrees o f success, and potential problems with sample design and the actualization of
sample designs may limit the usefulness o f the data collected. In addition, at this time the
data for some of the core variables are not beingcollected. This includes information on the
non-monetary aspects of living standards and representative data on household income. The
forthcoming Survey of Income and LivingConditions (SILC) may resolve this problem.

5.8    TheexistingdataframeworkisdescribedinBox5.1. Inadditiontothisbasicframework,
there are a few other surveys that could be useful for poverty monitoring (among other
uses) but that are not currently included in the work of the Statistical Office. These include
specialized health surveys such as the Demographic and Health Survey or the Reproductive
Age Mortality Surveys, and the Multiple IndicatorCluster Survey(MICS) which is designed
to look at women's and children's issues. Inaddition, surveys of facilities-such     as schools
and health posts-would     be useful to assess infrastructure availability and needs, and could
becarried out by the line ministriesinvolved.


               Box 5.1: ExistingData Capacityfor Monitoring Povertyin the SR
  The data currently available inthe SR to monitor poverty include:
        The Census, which was carried out in2001
      0 The Labor Force Surveys (LFS) undertaken quarterly since 1993
        A Microcensus carried out in2003 which collected informationabout 2002
      0  The Household Budget Survey (HBS) which was re-designed in2004
  A Survey ofIncome and LivingConditions (SILC) planned for 2005.


                       C. AREASOF POTENTIAL              IMPROVEMENT

5.9    There are several common concerns with the surveys currently implemented, which
can be grouped broadly into five categories: (i)   questionnaire design; (ii) design; (iii)
                                                                            sample
field work; (iv) responserates; and (v) comparability over time.

Questionnaire Design
5.10 There i s no systematic attempt to field-test alternative specifications o f surveys before
their implementation and this often leads to excessively complex surveys which may not
give the answers they were designed to address. This i s particularly obvious inthe case o f
the HBS. Thus, the coverage and quality o f the data could be greatly improved by putting
inplace a system to pilot test any andall changesto survey questionnaires. The pilot testing
system would determine whether the information provided was accurate and could show
where interview techniques neededto be modified or improved.

TheHouseholdBudget Survey (HBS)
5.11 The questionnaire for the HBS, which was re-designed in2004 from earlier versions,
consistsofthreeschedules:(i)    ashortquestionnaireonhouseholddemographics; (ii)     asummary
diary for income and expenditure for the first month o f participation; and (iii) a family diary
to record daily expenditures for the second month of participation. The questionnaires are

74                                            TheQuestfor Equitable Growth in the SlovakRepublic



delivered to the participating households along with detailed instructions for completion,
but they are complicated and difficult to fill out and there are no mechanisms to ensure the
completeness of the data. This i s of particular concern inthe recordingof income.

5.12 Inaddition, inits current format the information contained inthe HBS i s not sufficient
to measure consumption, as opposed to expenditures, as the HBS does not:


       0  Collect information onthe consumption of home production and o f goods received
          inexchangefor goods andservices
       0  Limit the recording of expenditures to those that will be consumed in the same
          period inwhich they were purchased
       0  Collect price information to permitregional adjustments.

5.13 Inaddition:

          There is no way to estimate the amount of consumption from subsidized
          government services such as educationor health. Only those education and health
          itemspurchasedby the householdare included inthe survey.
          Noattempt i smadeto estimate the consumption value ofdurable goods andhousing,
          items of particular importance inmiddleincome countries.
          Most important, the diary is completed by one household member on behalf o f the
          whole family-Le., aFamilyDiary(FD).Althoughthisi sfairly typicalo fexpenditure
          surveys inEastern Europe and CIS countries, it is unlikethe expenditure surveys in
          most Western countries, includingthe EU15 and the other EU8, where each family
          member keeps hisiher own diary -Le., a Personal Diary (PD).52
          Finally, a major limitation for the monitoringofpoverty is the fact that the HBSdoes
          not cover non-monetary poverty indicators, such as access to education and health
          care services, availability o f cleanwater supplies, indicators o f social exclusion, etc.

TheLabor Force Survey (LFS)
5.14 The questionnaire for the Labor Force Survey (LFS) has been inplace since 1993 and
will bere-designedin2006 to include the new EUROSTATcodification. The survey collects
mostofthe informationneededfor analyseso fthe employment situationo fallindividuals aged
15years or older, except for informationonwages or salaries. This i s amajor limitationsince
the only information available onwages inthe SR i s derivedfrom an enterprise survey whose
sample design appears to be questionable. Moreover, enterprise surveys tend to overestimate
earnings since they do not have information for the self-employed, small businesses and the
informal sector generally. The exclusion of the wage question from the LFS survey is based
onthe aprioriandnever testedbeliefthat accurateinformationwillnotbeprovided.However,


s2 PDs are generally considered better for obtaining complete household expenditure data because it
   is unusual, inmost societies, for any one householdmember to know the expenditure of every other
   member, especially expenditures on items such as alcohol, tobacco, daily travel, personaltoiletries,
   daily newspaperslmagazines, meals (especially snacks and lunches) eaten outside the home, etc.
   Recent research inGeorgia and Russia has confirmed that FDs do infact result inunder-recording
   of expenditure - around 7 percent overall inthe RussianHBSand about 15 percent inGeorgia.

Chapter 5                                                                                    75


proper testing o f this hypothesis should be carried out, as wage informationi s includedinthe
LFS ofall other EUcountries.

TheMicrocensus
5.15 The 2003 Microcensus questionnaire was administered in April 2003 and collects
information onincome and other indicators for the calendar year 2002,with arecallperiodfor
information up to 16months long. Recallperiods o f this lengthleadeither to underestimation
o fhousehold incomebecauseit is difficult to remember one's incomeover this lengthyperiod,
or to educated guesses inwhich respondents estimate their income over the whole year using
their current rate of income. Inaddition, serious errors withrespectto items for which there are
no writtenrecords, such as casual work, income from self-employment, or own consumption
o f self-produced food, are likely to occur.

SampleDesign
5.16 The surveyshave alltakenadvantageo fthe 2001Censusto re-designtheir samples,but
information on sample designs i s somewhat difficult to assess.53The most serious issue - at
least inthe case o f the HBS-is that the sample seems to be drawn so as to include the areas
that are closest to the sites inwhich the 8 regionaland42 sub-regional offices are located. This
could result in a strong geographical bias in the implemented samples. Households distant
fromregionaland sub-regional offices maybeunderrepresentedalongwith any characteristics
associatedwith livinginremote areas. Inaddition, the different surveys allegedly use different
samplingtechniques and different sampleframes, which makes itimpossibleto integrate their
results.

FieldWork
 5.17 Interviewers, who are permanent SOSR employees, receive minimaltrainingand little
 or no supervisioninthe field. The interviewersareresponsible not only for the data collection
but also for the key entry of the data, and the verification of the key entry is minimal.
 Supervisors check approximately halfofthe questionnaires, butthe procedures usedfor these
 checks are not known.

 5.18 The centraloffice inBratislavahas few ifany oversight fimctions. Decisions onhow to
 implement the field work are left to the regional offices. There i s no information onwhether
 there are commonly enforcedrules for decisionmakingamongthe regionaloffices. Ifregional
 offices use different rules, the resulting data could differ from region to region.

 ResponseRates
 5.19 The very low response rate is probably the most worrying aspect of current data
 collection. The surveywith the lowest responserate i s the HBS.Despitean incentivepayment
 of 1300Sk (about US$43) paid to each cooperating household, response to the HBS i s very
 low, at about 25 percent. Response i s even lower in Bratislava and among the very rich and
 very        Ofparticular concernis the fact that the effect o fthe incentive payment i snegated
 for those households that receive social benefits because the benefit is offset against the


 53  Formore details, see Volume I1ofthis report.

76                                              The Questfor Equitable Growth in the Slovak Republic



payment (although recipients o f unemployment benefits are not affe~ted).~~ suggests aThis
non;response bias which would considerably reduce the number of poor households in the
sample.Attempts havebeenmadeto havethe regulations changed, butthey have so far failed.
However, although these factors will undoubtedly contribute to the response problem, the
extent o f non-response is too great to be explained by these factors alone.56

5.20 The LFS has a response rate close to 93 percent. The Microcensus had an overall
responserate of approximately 73 percent, which i s quite a low rate considering that response
i s meant to be compulsory and that this figure also includes partial responses.

Comparabilityover Time
5.21 Another important limitation of the data for policymaking i s the fact that the surveys
currently inplace-andtheir findings-cannot be compared to similar surveys carried out in
previous years. This i s the result, primarily, o f the fact that the previous sample frame was
based on the 1991 Census, which was undertaken prior to the dissolution of the federation
o f the Slovak and Czech Republics. However, the 2001 Census didnot completely solve the
problem. Until2003, the HBSuseda quota sample madeupofthe following social groups (as
definedbythe household head): manualworkers; employees,the self-employed, farmers; and
pensioners. Householdsheadedbyunemployedindividualswere not included. Quotasamples
do not permitthe calculation o f selection probabilities, which precludes the weighting of the
results to make them representative of population totals. This i s not the case with the LFS,
which i s therefore more easily compared over time.

5.22 Inaddition to changesinthe sample frame, comparability across differentwaves o fthe
Microcensus surveys i s limitedby changesinthe questionnaire design, as the categories used
to collect information are more detailed in2003 than inthe 1997 q~estionnaire.~~            Inaddition,
the period o f time between the Microcensus surveys i s too long to make this a u s e h l tool for
monitoring the income inthe country.

The Issues of the RomaPopulation
5.23 The Roma are among the most vulnerable groups in the Slovak Republic. However,
measuring the living standards o f the Roma and the characteristics o f the communities in
which they live i s difficult usingthe existing surveys. The SOSRisnotpermittedbylawto ask
ethnicity insurveys. Nationality canbe includedina questionnaire, butnot all Roma identify
their nationality as Roma. The 2001 Census identifies approximately 80,000 Roma living in
approximately 700 communities using the nationality question. Roma groups estimate the
number of Roma as closer to 350,000. A social mapping exercise identifiedmore than 1,000
communities inwhich Roma are living, but there i s no reason to believe that the discrepancy


    Although the official line is that non-respondinghouseholds are not substituted, the fact that the SO
    selectsa sample o f about 19,000 addresses inorder to achieve a cooperating sample o f4,700 seems
    to amount to a substitution strategy.
    This at least seems to be the view ofthe Statistical Office andthe public despite existing legislation
    that allows for the possibility of social benefits being unaffected by one-off income items such as
    the mentioned incentive payment (Article 5, paragraph4e of the Act 59912003.)
    Unemployment benefits are not reduced by the amount o f the incentive payment.

Chapter 5                                                                                         77


arises from more than 300 communities beingexcluded from the census rather than from the
inhabitants not having been identifiedas Roma.

5.24 Nevertheless, because o f problems with the implementation o f the sample designs
and because o f the difficulty o f identification in the Census, the Roma are probably
underrepresented in every survey. I t i s possible to use other information as proof o f the
extent o f the problem. For example, in the 1997 Microcensus, few households reported
having more than 3 children while the average number o f children for Roma families was
4.2. However, these discrepancies have been rectified inthe 2003 Microcensus.

Lack of Informationon the Non-incomeDimensionof LivingStandards
5.25     Currently no information on some of the core indicators ofwelfare i s beingcollected
inthe SR. Periodic specialized surveys, such as a Demographic and Health Survey, should
be undertaken on a regular basis to provide more detailed information on such topics
as women's and children's health. Administrative data as collected by line ministries and
specialized research institutes should also be used. Examples of the type of indicators for
which information can be obtained from administrative data and the potential sources of that
informationare provided inthe Technical Note inVolume 11.

5.26 However, it is widely accepted that the best tools for collecting information on living
standards are surveys that collect information on more than one subject, inorder to capture
the multiple dimensions o f living standards and their recognized connections--for example,
between health and housing conditions or between education and fertility. While single topic
surveys are important and provide much useful information, for the purposes o f measuring
living standards their focus is too narrow to address broader issues of social and economic
development.58

5.27 Unfortunately, none o f the surveys currently implementedinthe SR i s multi-topical in
nature. The Survey of Income and Living Conditions (SILC), which is shortly to be carried
out by the SOSR, will represent the first multi-topic integrated household survey in the SR
and, as such, it provides the first opportunity to obtain a complete multidimensional profile of
poverty and living standards inthe SR.Thus, it i s extremely important that special attentioni s
given to the sample design and the implementation of this survey, so that it will generate the
highquality integratedinformationthat isrequiredto (i)       adequately evaluate the impact o fthe
reform process on growth and living standards; (ii)inform policymaking at this challenging
but promisingtime; (iii)  prepare a meaningful andrealistic strategy for reducingpoverty and
social exclusion along the lines o f the NAP requiredby the EU; and (iv) monitor progress
towards the implementation o f the strategy.




57  With the long recall period, it is better to use more detailed questions in order to obtain a more
    complete recollection. However, because o f changes to the questions, the answers cannot be
    compared completely because the responses cover different units o f measurement.

78                                         TheQuestfor Equitable Growth in the Slovak Republic


                                     D. CONCLUSIONS

5.28 Essentially, if correctly carried out and expandedto cover the non-income dimensions
of living standards, the surveys currently in place in the SR would provide much of the
informationneededfor monitoringpoverty, evaluating the impact ofreforms, andformulating
evidence-based strategies for social inclusionand poverty reduction. However, the execution
of the surveys shows varying degrees of success. This chapter has highlighted a number o f
potentialproblems, o f which o fthe following are of particular concern:


       0 Questionnairedesign-This is anissueparticularly for the HBSwhich (i)sbased   i
         on a very complex self-completion questionnaire; (ii)uses Family Diaries rather
         than Personal Diaries for the recordingof expenditure; and (iii) does not collect the
         informationrequiredto compute householdconsumption, as opposedto expenditure.
         Thus, the mainconcerns arethat:

         (a) The informationon income may be highlyunreliable
         (b)The use of FamilyDiaries leads to under-recording of expenditure
         (c)No information on a very important indicator o f household welfare - i.e.,
              household consumption--is available for the SR.

         More generally, the lack of systematic piloting of the questionnaires i s cause for
         concern.


       0 Designandimplementationofthe sample-The largest issue inthe design ofthe
          sample, at least inthe case of the HBS, i s that the implemented samples seem to be
         drawn so as to include the areas that are closest to the sites inwhich the 8 regional
         and 42 sub-regional offices are located. This could result ina strong geographical
         bias.


       0 Very low responserates on some surveys-The survey with the lowest response
          rate i s the HBS for which only approximately one-quarter of the households that
          were asked to participate agreedto do so. The Microcensushadan overall response
         rateo fapproximately 73 percent, which i s alsoquite low consideringthat responsei s
         meant to be compulsory and this figure also includes partialresponses. By contrast,
         the LFShas aresponserate close to 93 percent.


       0 Lackoftrainingofinterviewersandofsupervisioninthefield O fconcernalso
                                                                             -
         i s the poor quality control exercised over the data collection process. For the HBS
         and the Microcensusthere i s very little propertraining of, or field supervisionover,
         the interviewers. Although the LFS interviewers do seem to receive systematic
         training and re-training, the concern about lack o f field supervision remains.


       0 Lack of informationon the non-income dimensionoflivingstandards -At this


58 Chander, Grootaert, andPyatt, p. 5.

Chapter 5                                                                                 79


         time, information on some of the core indicators of welfare i s not being collected.
         While other ways of collecting this information are possible, it i s widely accepted
         that the best tools are multi-topic surveys that simultaneously collect information
         on more than one subject. These surveys have the capacity to capture the multiple
         dimensions of living standards and their recognized connections--for example,
         between income and education, health and housing conditions or education and
         fertility. The Survey o f Income and Living Conditions (SILC) which i s shortly to
         be implemented is one of these surveys and will represent the first opportunity to
         obtain a complete multidimensional profile of poverty and living standards in the
         SR. Thus, it is extremely important that special attention is given to the sample
         designandthe implementationofthis survey so that it will generate the highquality
         integratedinformation that i s required.


                                  CHAPTER 6              -
   CANCOMPETITIVENESSAND INCENTIVES
       BREAK THE CIRCLE OFPOVERTYAND
                               DEPENDENCE?


6.1   Perhaps the clearest mandate emanating from the 2002 election inthe Slovak Republic
was the reductioninunemployment that followed the economic restructuring and the structural
changes in the enterprise sector. This mandate was reinforced in the preparation of the Joint
Inclusion Memorandum (JIM) prior to EU accession and later in the National Action Plan
for Employment (NAPE) where, according to the LisbonTreaty, it was reformulatedinterms
of growth in employment and an increase in competitiveness. Therefore, it i s not surprising
that economic liberalization and improved labor market indicators became a priority for the
government that emergedfrom the 2002 elections.

6.2   These objectives were pursued with determination and thoroughness. Over the
period2002-04 the government introducedanumber o f measures designed to increase the
demand for and supply o f labor. On the demand side, the emphasis has been on fostering
job creation by improving the business environment through a package o f measures that
covered both the labor market and the economy more widely. These measures included
the following:


      0  Making the taxation and regulatory frameworks more supportive of new enterprise
         creation andbusiness sector development
      0  Reducing non-wage labor costs (by as much as 13 percent on average, from 38
         percent to 33.2 percent o f total labor cost
      0  Increasinginflexibility inlabor contracts andlabor relationsincluding easingmany
         restrictive regulations inboththe labor market and the product market.

6.3   On the supply side, on the other hand, the focus was onmakingwork pay and reducing
skill mismatches through the following measures:


      0  Increasingthe value o f the minimumwage relative to the average wage
      0  Reducingthe disincentiveto work generatedbythe pre-reform socialassistancesystem

82                                             The Questfor Equitable Growth in the Slovak Republic



       e   Making the public education systemmore efficient, effective and responsive to the
           needs of the twenty-first century environment.

6.4    Inthe view of many Slovak and international observers, the comprehensive reform
package was long overdue and was very muchinline with recommendations from a number
of sources.59I t also representedthe mainelement offirst the JIM and thenthe NationalAction
Plan prepared by the Slovak government incooperation with the EU as part of the reporting
requirements for the Lisbonagenda. Indeed,the spirito fthe reformwas very muchinline with
the principlesunderlyingthe EULisbonTreaty of2000and, inparticular, with the objectives
o f"making work pay" and increasing employment opportunities for the overall populationby
increasingthe competitiveness of the economy.

6.5    At the same time, the reformpackagemetwith vigorous oppositioninsomeparts ofthe
country and among some groups. The mainconcern o f the critics was that, while promoting
economic growth and increasing the returns from employment, the reform deepenedthe gap
betweenthose who could benefit fully from better economic opportunities and the rest of the
population, who riskeda considerable reductionintheir income. This concern i s ofparticular
relevance inthe context of a country such as the SR with deep-rooted historical inequalities
and i s reinforced by the findings that to date the growth inGDP has not been fully reflected
in more favorable labor market outcomes for those groups with the strongest barriers to
employment, such as the unskilled and those living inlaggingregions.


          A. HAS STRATEGYWORKED DATE FOR WHOM?
                       THE                              TO          AND


6.6    While it is too early to assess the full impact o f the reform, the findings o f this report
suggest that overall the "Strategy of GrowthinEmployment" seems to have beenreasonably
effective inachieving the following targets:


       0   Eliminating the disincentive to work provided by the highreplacement ratio faced
           by individuals with low earning potential underthe old system
       0   Increasing the disposable income of the majority of households that are entitledto
           benefits
       e Providing better employment opportunities and access to ALMP programs for the
           average worker and for particularly vulnerable groups.

6.7    However, questions remain regarding the magnitude of these effects and their
sustainability over time. Inaddition, it is important to note that, giventhe short time-frame o f
the analysis, it is difficult to establish whether the recent positivetrends insome indicators are
the result of the reforms or the effect o f short-term trends.

6.8    Inaddition, owing to the lack of adequate post-reform data, the impact of the reform
on income and poverty cannot be evaluated but can only be simulated. This report uses


59 Amongthem were the WorldBankreports (2002 a) and OECD (2004).

Chapter 6                                                                                     83


two different approaches. First, it computes pre-reform and post-reform income levels for
different household typologies and calculates the net impact o f the reform by comparing
these two levels. The secondapproach uses a "Tax-Benefit Model" to compare, for all sample
households, the income levels of the household before the reform with the simulated post-
reformlevels so as to identifythe gainers and losers as well as the distributional and poverty
changes. The first approach draws the following conclusions:

          The combined effect of the reformonthe tax andbenefits regime has increased the
          disposable income of the majority of households that are entitled to benefits
          The only households that may have lost out throughthe reform arethose with three
          or more children where all adults are unemployedand that were entitled before the
          reformto the higher benefits schedule
          Whether households have lost out depends predominantly on their entitlement to
          school assistance
          Even for the most affected households the maximum potential loss in disposable
          income i s limited.

6.9    These conclusions are confirmed by the findings of the Tax-Benefit Model, which
concludes that the reform has only a small potential effect on the incidence of poverty, with
the poverty headcount decreasingby 2 to 3 percent according to the simulationsbased onthe
2003 Microcensus. This i s the case despite an overall reduction of around 16 percent inthe
average amount ofthe transfers. What this implies i s that the change inthe tax-benefit regime
initself (Le,, in isolation from the reduction intransfers) would have significantly reduced
poverty, by as much as 17percent, had it not been accompanied by a simultaneous reduction
inthe averagetransfer.

6.10 Aprimary factor behindthisresultisthereductioninthetaxburdenonthe lower income
groups which has been associated with the move to a flat-rate personal income tax (PIT)
combined with a substantial increase in deductions and credits. This i s an important result,
since it shows that concerns about the potentially negative distributional effect o f a flat-rate
PIT may be misplaced. Inthe Slovak case the flat rate tax has not been merely an efficiency
measure. The reform has increased the progressivity o f the tax system while replacing five
tax brackets (and a fairly steep scale o f tax brackets) by a single one. Indeed, in2004 the PIT
burden was reduced for everyone but particularly for households with lower incomes. This
was due to the increase indeductions and credits, which has more than compensated for the
reductioninprogressivity associatedwith the move to a single tax rate.

6.11 However, while the incidence o f poverty has declined, its intensity and the related
poverty gap have increased considerably-by         81 and 62 percent, respectively-suggesting
that the income levels o f those inpoverty have declined as result of the change intax-benefit
regime.The simulations suggestthat the inequalityinhousehold incomehas also increased as
a result of the changes inthe structure o f taxes andbenefits.

6.12 Interms o f labor market indicators, the information presented inthis report refers to
2004 asawhole. However, these values hideaclear difference betweenthe first quarter o f2004
and the last part o f that year. Duringthe earlier period, the overall performance o f the labor

84                                           The Questfor Equitable Growth in the Slovak Republic


market deteriorated sharply with 41,000 fewer people employed and a 3 percent reduction in
the employmentrate. However, the labormarket's performanceimprovedconsiderably inthe
nexttwo quarters, with anoverallincreaseinemployment of73,000 people and a 2.6 percent
increase inthe employment rate, and remained largely unchanged inthe last quarter.

6.13 Thus, the use o f quarterly data suggests that after an initial negative effect the post-
2002 reform package may have led to a relatively high increase in employment with a year
to year seasonally adjusted growth of 1-2 percent, driven primarily, but not only, by growth
in self-employment. Self-employment continued to increase even after the implementation
of the 2004 reform and stabilized at around 2 percent per quarter from the second quarter of
2004. By way o f contrast, the number of employees declined substantially, by as much as 4
percent, inthe secondquarter of 2004. However, overall employment began to increase from
the third quarter of 2004, reaching 2 percent by the first quarter o f 2005. An inspection o f
quarter to quarter changesreinforces thisfinding.After a sharp drop inthe secondhalfo f2003
(-0.3 percentand-0.4 percent inthe thirdandfourthquarters,respectively), overallemployment
beganto increasein2004 withatotalincreaseof46,200 jobs-and          a0.4 percent increaseinthe
employment rate-in      the period from the second quarter of 2004 to the first quarter o f 2005
[all seasonally adjusted]).

6.14 It i s too early to say to what extent this is the beginning o f a trend and whether it is
attributable to the success o f the 2004 reforms, but the signs are encouraging. The fact that
thisgrowthis drivenexclusively by growth inself-employment-in         the contexto fa declining
number ofemployees-is        also worth noting, as itpointsto economic liberalizationasthe main
driver o f growth and has important implications for policymaking. This could be interpreted
as sign that the labor market i s moving in the right direction. However, it i s too early to
conclude with any degree of confidence that the 2004 reforms have increased employment
opportunities. The confirmation that this i s not simply a cyclical effect can be provided only
by continuous monitoring over the next months.

6.15 Finally, it is also too early to fully asses the effect of the ALMPs that have been
implemented. The available data show that the "activation contribution" i s by far the most
popular policy introduced to date in terms of number o f beneficiaries and that it is a very
successful measure, despite the lack o f employment opportunities in some areas. However,
the question of whether participation in this scheme will permanently increase a person's
employment opportunities remains open. The international evidence on the success o f these
types o f interventions i s contradictory and country-contingent,

6.16 I ti salsoimportanttorealizethat, inadditiontotheirpotentialeffectonfutureemployment
prospects,ALMPs can play a significant role as a strategy for reducing social exclusion. This
is particularly important ina dual labor market such as that inthe SR, where a small pocket
ofthe labor force i s at risko fbecoming increasingly marginalizedand "discouraged." Within
this context, the programs canbeseenas away ofkeepingthese workers engagedinthe labor
market and the economy rather thanhavingthem rely exclusively on social assistance.Inthis
context these policies would act more as an alternative form o f social assistance designed
to break the cycle of dependence on social exclusion rather than as a route to permanent
employment.

Chapter 6                                                                                      85


             B. PRIOFUTYACTIONSARENEEDEDCURBPOVERTY        TO


6.17 The overall conclusion of this report is that the government's strategy for increasing
employment and tackling social exclusion i s comprehensive and internally consistent. All
o f the right mechanisms are possibly in place to address poverty and social exclusion in
the long term by breaking the vicious circle of lack of skills/poverty/state dependence that
has prevented the poor from breaking out of poverty for generations. The findings to date,
although limited, also suggestthat this strategy i sbeginningto achieve the desired results and,
based onrecent trends, the expectations for the future are optimistic.

6.18 Nevertheless, it is important not to be complacent, as (i) progress to date has been
                                                                  the
somewhat slower than anticipated, and (ii)   the SR faces issues o f deep-rooted socioeconomic
inequalities which historically have always been of a much larger magnitude than those
experienced by other EU countries-with        the possible exception of regional disparities in
Italy. Within this context, the followingpriorities emerge:


       0  Maintaining the conditions for equitable growth
       0  Increasingthe skills of the poor and reducing the cost o f employment
          Strengthening the effectiveness of active labor marketpolicies
       0  Supportingthe living standards of the most vulnerable.

                     MAINTAININGCONDITIONSFOREQUITABLE
                                   THE                               GROWTH
6.19 Although a rigorous analysis of the link between growth and poverty reduction is
prevented by the lack of comparable data on poverty, the importance of growth for poverty
reduction i s clear. Inthe eyes of many, the comprehensive package o f reforms implemented
inrecent years has (i)  positionedthe SR well not only interms o f other New Member States
(NMS)oftheEUbutalsoincomparisonwiththeOldMemberStates(OMS), and(ii)                    provided
the SR with the chance to take full advantage of the new opportunities presented by its entry
into the EU.Thus, the prospects for sustainable growth over the next years are very good.

6.20 But the evidence from recent years shows that growth alone will not be sufficient to
move the SR closer to the Lisbontargets, where the responsivenesso f employmentto growth
i s substantially lower than that inthe EU15 countries, although in line with that inthe EU8.
For growth to become an effective engine for poverty reduction, its link with employment
creationneedsto be significantly strengthenedbymakingthe growth environment more"labor
friendly" and "equitable." In general, this would require interventions in both the demand
side and the supply side of the labor market as well as inthe area o f labor market mismatch.
The SRhas already introducedmany o f the changes required to improve conditions on these
different fronts. However, while the formula for the next few years is likely to be one o f
waiting for the benefits o f these measuresto bear h i t and for their full impact to be realized,
some marginal changes may help inmoving the country more rapidly to higher employment
equilibrium.

6.21    This is particularly true on the demand side, where a positive business climate has
been created over the last years by reducing the "tax wedge" (payroll taxes) and increasing

86                                          The Questfor Equitable Growth in the Slovak Republic


the flexibility and competitiveness of the labor market and the economy at large. However, a
further reductioninlabor costs may be requiredat the bottom endof the earning distribution
in order to increase the employment opportunities for those low-skilled/low-productivity
workers who are at the highest riskof unemployment. Inthis regard, it i s recommended that
the current policy ofraisingthe minimumwage more rapidly than the average-as         part ofthe
objective o f increasing the incentive to work-be   progressively abandoned, since increasing
the price o f low-paidlow-skilled workers relative to other workers and to their productivity
would reduce the demand for these workers. It i s also recommended that the tax wedge be
further reduced, particularly at the bottom end o f the earning distribution, by raising the
minimumlevelofearnings to which contributions are paidabovethe minimumwage andby
introducing a uniformdeductible level inline with the reform of the PIT.

6.22 On the other hand, on the supply side, the recent reforms o f the tax and social
benefits system have substantially increased the incentive to actively participate in the
labor market by (i)eliminating the previously substantial poverty trap, (ii)considerably
reducing the replacement ratio, and (iii)introducing financial incentives to take up
employment even if on a part-time basis. Inaddition, the issue of labor market mismatch
has been addressed via an imaginative package o f active labor market policies designed
to bring the long-term unemployment and discouraged workers back into employment
and to increase geographical mobility via targeted policies combined with more general
types o f interventions.

6.23 Finally, the government has addressed the issue o f regional inequalities within the
context o f the EURegional DevelopmentProgram. However, inorder to continue to utilize
the incoming funds effectively to reduce inequality, the SR needs to make an important
decisiononwhether to put efficiency or equityfirst initsmodel o fdevelopment. The diverse
experiences of some EU countries clearly show that a trade-off may exist in this area,
especially inthe case o f countries with largeregionalinequalities.The crucial questionhere
i s whether the available funds should be used to target low-income regions or to increase
the potential of already more affluent "growth poles" -that is, whether or not to choose the
route o f "balanced growth".60

Increasingthe Skills of the Poor
6.24 The low level o f skills inthe Slovak labor market, especially inthe poorest regions, i s
a major cause of concern for the economic development o f the country. Addressing this issue
comprehensively requires a multi-pronged strategy designed to undertake the following:


       0   Improvingschool enrollment andthequality ofeducation for youngpeoplecurrently
           inthe education system
       0   Enhancing the skills (and employmentprospects) ofthose who have already left the
           school system
       e Implementingas soonaspossiblea radical reformof the higher education system in
           line with that recently prepared by the government.



6o  FunckandPizzati (2003).

Chavter 6                                                                                    87


6.25 The governmento fthe SRhastakenactive measurestoaddressthese issues. Forexample,
among the mainelements o f the reform are (i)    the introduction of scholarships for children in
materialneedattending school; (ii) school subsidies program; and (iii)
                                     the                                   extensive provisions
for retraining the LTUand others. Inaddition, the MoLSAF andthe MoEare working closely
to improve the transition from school to work and to break the cycle of the intergenerational
transmission o fpoverty that i s so strong inthe SR.

6.26 Generally speaking, a certain degree of skepticism exists regarding the effectiveness
of retraining programs. However, recent evidence suggests that these programs are fairly
successfid, except for the retraining o f workers in cases o f mass layoffs. Inaddition, inthe
case of labor market duality, these programs can be considered as one way to effectively
reduce social exclusion by providing otherwise excluded workers with an opportunity to go
back into the labor market, at leasttemporarily. Inthis case, the programs wouldbe seenmore
as an alternative form o f social assistancethan as a route to permanent employment.

Strengthening the Effectiveness of Active Labor Market Policies
6.27 The findingsofthisreport indicate that the spreadoflong-termunemploymentinthe SR
is associatedwith the deteriorationo fthe relative positiono f specific vulnerable groups, such
asthe unskilledand individuals living inpoor areas.Whenunemploymentriskis concentrated
and the receipt ofbenefits discourages regional labor mobility, there may be a strong case for
active labor marketpolicies (ALMPs) onthe grounds that they canhardlyinvolve deadweight
costs, while substitutioneffects may evenbe desirable.

6.28 From the beginningo f the transition to a market economy, the Slovak government
has applied a wide range o f such policies to combat LTU. In terms o f expenditures, three
programs have been o f particular importance: training, "intervention works" and public
works. Training programs are designed to solve skills mismatch inthe Slovak labor market.
Workers with redundant skills or no skills are trained in those occupations where there i s
a strong labor demand in the expanding sectors o f the economy. "Intervention works" i s a
program that in essence gives employers wage or job subsidies inthe amount o f the level
o f unemployment benefits. These wage subsidies are given to firms inthe private or public
sector if they hire an unemployed person and are larger the longer this person i s kept on
in the firm. Public works jobs are jobs directly created by the government, in particular
the municipalities, and targeted mainly (but not exclusively) at the long-term unemployed.
Many o f these jobs are in construction and in the cleaning o f public buildings, parks, etc.
Inother words, they have low skills content, although inprinciplebothinterventionworks
and public works have been conceived as maintaining or enhancing the human capital o f
participants.

6.29 While evidence from the SR is limited, several studies have investigated the
effectiveness o f ALMPs in other Central European countries by drawing on micro data on
labor market transitions6' and using a variety o f techniques (e.g., estimates o f augmented
matching functions, simple duration models including program intakes, matching estimators



61  Puhani and Steiner (1996 and 1997); Boeri (1997); Puhani (2000).

88                                            The Questfor Equitable Growth in the Slovak Republic



substitutingfor randomizationinlabor market programs, etc.). Thekeyquestionaddressedby
these studies i s whether after participatinginanALMP apersonfinds himselfor herselfbetter
positioned inthe labor market than ifhe or she hadnot taken part inthe program.

6.30 The results are not altogether encouraging with respect to the effectiveness of these
programs in improving re-employment probabilities. However, there i s increasing evidence
at the macro level that ALMPs can help countries to grow and increase employment. For
example, EC (2004) notes that while trade openness i s a main determinant of employment
growth, the intensity o f spending on ALMPs also contributes significantly. In particular,
training programs only mildly affectjob findingrates. Better results are obtained from wage
subsidies, but, typically, public works schemes involve stigma effects which may actually
reduce the likelihood of finding ajob after participation in such schemes. A type of public
service that i s usually fairly successful inincreasing employment among those seeking work
is j o b search assistance. This is also a relatively cheap type o f initiative that often has a
very favorable cost-benefit ratio. However, it is effective inlabor markets with low structural
unemployment and low long-term unemployment. Hence, in itself such assistance i s likely
to be limited invalue inthe SR.Labor mobiZity grants, on the other hand, can also be useful
instruments for bringing labor force skills closer to the needs of employers and for reducing
labor market mismatches via increasedlabormobility, anissue that is clearly ofimportance in
the case o f the SR.

6.3 1 Nevertheless, policymakers should not attach excessive hopes to ALMPs and should
berealistic about what they are capableo f achieving. There is no standardALMP that works
consistently andnone o fthese policies i s apanacea.Someofthem couldwork insomeplaces,
for reasons which are not always fully identified. Successful programs often involve a non-
orthodox approach andmay be difficult to replicate or to expand. A summary of the evidence
onbest practices to date is presentedinBox 6.1.

ChaDter 6                                                                                           89



           Table 6.1: Which Active Market Policies Are Most Likely to Work?
 A reasonable consensus exists among researchers regarding key success features of successful
 ALMPs. The main conclusions are as follows:
 (i)comprehensivepackageofservices(forexample,acombinationofindividualizedjobcounseling,
   A
 retraining andtemporary employment subsidies) usually works betterthan policies separately
 (ii)Orientation towards labor demandhelps(linking actions to realworkplaces, providingon-the-job
 training rather thanformal classroom, etc.,) andALMPs work bestwhen the economy is growing
 (iii) design,includingprecisetargeting, andconsiderationoftheenvironmentinwhicha
      Careful
 program operates, is an obvious sine qua non
 (iv) Careful assessmento f incentives provided is necessaryparticularly when benefit disincentives
 play a role -making sure that consistently working pays betterthanwelfare
 (v) Finally, monitoring and evaluation are important to provide feedback on the effectiveness o f a
 program.
 The following broadALMP categoriesare foundto bemostpromising inputtingpeople backto work
   Interventions inthe unemployment spells, forcing a sustainedjob search effort and a continuous
   contact with the labor market
   Job intermediation including an individualized approach when possible and necessary (support,
   counseling, placement assistance, individual action plan, labor supply/demand analysis, use o f IT)
   Programs dealing with issues such as individual skill deficits, lack of work experience and
   informationbarriers inthe labor market.
 The pros and cons of the ALMPs currently used inthe SR are summarized below:

 0 Employment services: (such as counseling, placement assistance,job matching, labor exchanges
   etc.) are generally found to have a positive impact, particularly in transition countries, where
   many people have never learned how to actively and effectively look for ajob. A good thing i s
   that they are relatively cheap, so benefits could be large. The biggest problem i s that they are
   usually o f limiteduse at times o f weak labor demand or if the structural unemployment is high
   and entrenched. Separately, an open question in the area concerns the relative effectiveness o f
   public versus private delivery o f employment services.
   Training: Evidence shows that training o f the unemployed has the potential to be successful in
   bringing people to jobs. In particular, an on-the-job training with an active involvement o f an
   employer, or specialized training (possibly leading to formal qualifications), could have large
   positive impacts. The effects o f training programs on employment for transition economies are
   consistently foundto bepositive. Onthe negative side, youthtraining programs arevery often found
   to beunsuccessful,butthisresult concernsonly developed countries. Someinconsistent butpositive
   results, however, are found inother countries. Inany case, this would make a case for putting the
   money inearlier inthe process--i.e., inthe education systemandreductiono fdrop-outs.
   Employment subsidies: which used to be assessed as not likely to have a positive impact and
   were associated with substantial deadweight costs and substitution risks. However the evidence
   is more promising with "new" types o f employment subsidies paid directly to individuals as
   a supplement to their earnings, as an incentive to encourage work and (in contrast to welfare
   payments) as a way to achieve greater economic self-sufficiency. Also, subsidies programs are
   more likelyto work, to the extent that they are usedto compensate employers for their screening,
   orientation, andinitialtraining costs. Finally, design, precise targeting, monitoringandpackaging
    (Le., combining the subsidy with other components such as training orjob search assistance) are
    important to the successo f employment subsidies.

Source: G. Betcherman, K. Olivas, and A. Dar (2004), "Impacts o f Active Labor Market Programs," World
Bank,
OECDEmploymentOutlook 2002, The Ins andOuts o fLong-termUnemployment

90                                               The Questfor Equitable Growth in the Slovak Republic



6.32 The Slovak government introduced a wide package o f ALMPs as part o f the 2004
reform. Among the policies introducedto date, the activation subsidy i s the most popular
interms of both number of beneficiaries and total expenditure.The second most popular
instrument,the contribution for childcare costs for ajob seeker, involves about 90 percent
fewer beneficiariesandcosts about 90 percent less thanthe activation subsidy.Monthly data
indicate no major changes in the trends in numbers of beneficiaries, with the contribution
for child care costs for job seekers and the contribution for graduate practice developing at
the most rapid pace. The average cost per beneficiary seems to have stabilized. However,
while the three mostpopular policies cost betweenSk 5,000 and Sk 14,000, the contribution
for self-employment costs almost Sk 60,000, and other contributions cost more than Sk
70,000 (however, information of duration o f eachprogram is missingby now).62

6.33 Unfortunately, it i s clearly too early to fully assess the impact o f these policies on
the employment prospects o f those inthe scheme. However, it is important to realize that,
in addition to their potential effect on future employment prospects, ALMPs can play an
important role as a strategy to reduce social exclusion. This i s particularly important in a
highly dual labor market such as the Slovak one where a small pocket of the labor force is
becomingincreasinglymarginalto the labor market and i s increasingly"discouraged." Within
this context the programs can be seen as a way of keeping these workers engaged with the
labor market and the economy rather thanhaving them rely exclusively on social assistance.
Thus, inthis context thesepolicies would act more as an alternative form o f social assistance
designed to break the cycle of dependence and social exclusion rather than as a route to
permanent employment.

Supportingthe LivingStandards of the Most Vulnerable
6.34 It is also important to recognize that, even under the best possible scenario, in
which the reformproves extremely successful inpromoting employment opportunities and
"making work pay," pockets of unemployment and social exclusion will remain, especially
inacountrywithahighlysegmentedlabormarketlikethe SR.Inorderto improvethe living
standardso f the most vulnerable, a comprehensive and adequate system o f social assistance
i s needed. Whilethe 2004 changes inthe tax andbenefits structure havepotentially increased
the disposable income slightly for most low-income households, they seem to have had a
negative effect on the income level o f poor households with unemployed adults and large
numbers of children. These households are highly concentrated at the bottom end of the
income distribution (both pre- and post- 2004) and thus it would not be surprising if the
2004 reforms had left the poverty headcount basically unchanged but had increased the
intensity o f poverty, as the disposable incomes o f poorest households were the only ones
negatively affectedby the reform.




62  The comparatively low spending on ALMPs in 2004 could be explained by the introduction o f
    new policies and their gradual phasing in:the activation subsidy was implemented from February,
    while other policies were implementedfrom March to May. Thus, while inthe first half o f the year
    only Sk 0.4 billion was spent, inthe second half Sk 1.5 billion was spent. Using the latter figure,
    we estimate a potential cost o f Sk 3.0 billion in2004, which is a somewhat higher than the Sk 2.8
    billion spent in2003.

Chapter 6                                                                                         91



6.35 To some extent this was the unavoidable consequence o f the shift to work incentives
and the reduction in the basic benefits which was required to reduce the fiscal pressures
generated by the formerly overgenerousbenefits system. However, inview of the barriers to
employment that remain (at least inthe short run) for unskilledworkers and for those inthe
lagging areas, the government may wishto target these groups andcommunitieswith targeted
short-term measures designedto help them overcome the loss of livelihood.

6.36 In the context o f the emphasis on providing the right incentives, any individual/
householdlevelinterventionof this type would needto be designedas a conditional transfer
program(i.e., as one inwhichentitlement to thebenefiti s conditional on some formo fpositive
behavior). As the focus o f this intervention is likely to be households with large numbers o f
children, it may be advisable to make the condition relatedto children. Programs of this type
that have proved to be successful in similar contexts have provided transfers conditional on
child attendance and regular visits to healthclinics.63Such programs may directly addressthe
vicious circle of the intergenerational transmission o fpoverty that is so evident inthe SR.

6.37 The 2004 reform has already introduced a number of conditional transfers such as
the activation allowance and the secondary and higher education scholarships. It has also
introduced similarly targeted interventions at the community level via the training and
education grants and the other interventions. These were steps inthe right directionthat need
to be reinforced for a more perceivable effect and to ensure their long-term sustainability.

6.38 Inprinciple, funding for these additional types of interventions shouldnot be difficult
to obtain. The simulations in this report suggest that in its current format the new benefits
system maintains a relatively highdegree o f inefficiency, with benefits being paid not only to
the poor but also to households inthe top income deciles. This i s partly due to the shift away
from income contingency o f all family-related benefits. While child benefits tend not to be
income-related across the EU countries, it i s arguable whether this i s a choice in a country
with limited resources and very deep pockets of poverty. Inaddition, the removal of income
conditionality for the parental allowance i s more difficult to justify inthe context o f limited
resources for social assistance. Reverting to a system o f income-related family benefits may
generate enough savings to finance the increased emphasis on conditional transfers.




63  Programs o f this type that have receivedparticular attention are, for example, Progressa inMexico
    and BorsaEscolar inBrazil.

92                                           The Questfor Equitable Growth in the Slovak Republic



 C. IMPROVING         THE CAPACITY        FOR MONITORING          THE   IMPACT  OF RECENT
                        AND   FUTURE     CHANGESPARAMOUNT
                                                      I s


6.39 In the midst of such a comprehensive process of transformation, the government's
desire (i) evaluate the impact o f the reforms already implemented and feed the results
          to
of this evaluation into the next phase o f the reforms, and (ii),    more generally, to build the
capacity for an effective monitoring system and effective evidence-based policymaking is
very welcome and i s to be encouraged. Unfortunately, the next steps in this direction are
prevented by the lack of reliable data producedat regular intervals.

6.40 Inshort, for the economic transformationofthe SRto becomplete, iti s essentialthat (i)
the country's statistical system i s updated to the level o f the other EU25 countries and to the
minimumlevelrequiredfor adequatemonitoringandreportingtotheEUonpriorityindicators;
(ii)adequate capacity i s built within the government for evidence-basedpolicymaking; and
(iii) linksareestablishedbetweentheSOSRandtheSlovakgovernment-especially
    closer
the MoLSAF.These objectives are clearly a priority and boththe SOSR and the government
of the SR are working inthis direction.

                               REFERENCES

Beblavy,Miroslav. "Reform ofthe Social System: Motivation, TargetingandConsolidation,"
presented September 2004.

Betcherman, G., K. Olivas, A. Dar (2004). "Impacts of Active Labor Market Programs",
World Bank

BilsenV., andJ. Konings (1998), "Job Creation, Job DestructionandEmployment Growth in
Newly EstablishedFirms inTransition Countries: Survey Evidence from Romania, Bulgaria
andHungary," Journal of ComparativeEconomics, 26,429-445.

Boeri, T. (1997). "Learning from Transition Economies: Assessing Labor Market Policies
across Central andEasternEurope,'' Journal of ComparativeEconomics, 25, 366-384.

Chander, Ravi, ChriostiaanGrootaert, and GrahamPyatt (1980). "Living Standards Surveys
inDeveloping Countries," LSMSWorking PaperNo.1,Washington,D.C., p.5.

EC (2003a), "Employment inEurope2003: RecentTrends and Prospects."

EC (2003b), "Jobs, Jobs, Jobs: Creating More Employment in Europe," Report of the
Employment Taskforce.

EuropeanUnion(2003[a]). Oficial Journal of the European Union, L165,3.7.2003, pp 1-9.

EuropeanUnion (2003[b]). OfJicial Journal of the European Union. L298, 17.11.2003, pp
1-22 and 34-85.

Faggio, G., and J. Konings (200l),"Job Creation, Job Destruction and Employment Growth
inTransitionCountriesinthe ~O'S," IZADiscussionPaperNo. 242.

Funck, Bernard, and Ludovico Pizzati (2003), "European Integration, Regional Policy, and
Growth," the World Bank: Washington, D.C.

Gomez-Salvador,R., J. Messinaand G. Vallanti (2004). "Gross Job Flows and Institutions in
Europe," Labor Economics, 11,469-485.

Konings, J., 0.Kupets andH.Lehmann(2003). "Gross Job Flows inUkraine," Economics of
Transition, 11, 321-356.

Ministry of Labor, SocialAffairs andFamily (2004), Joint Inclusion Memorandum between
the Slovak Republic and the EU.A documentprepared with help of EuropeanCommission,
Directorate-Generalfor Employment and SocialAffairs.

OECD (2004), Slovakia, Economic Survey.

94                                        The Questfor Equitable Growth in the Slovak Republic



OECD Employment Outlook 2002. The Insand Outs of Long-Term Unemployment.

Orcutt, G. (1957). "A New Type of Socio-economicSystem," inReview of Economics and
Statistics, 58(2).

Puhani, Patrick, (2000). "Poland on the Dole: The Effect of Reducing the Unemployment
Benefit Entitlement Period during Transition," Journal of Population Economics, February,
134;1, 35-44.

Puhani, Patrick and Viktor Steiner (1996). "Public Works for Poland?Active Labor Market
Policies during Transition," ZEW DiscussionPapers, 96-01, Mannheim, 1996.

Puhani, Patrick, andViktor Steiner (1997). The EffectivenessandEfficiency ofActive Labor
Market Policies inPoland?" Empirica, 24, 3:209-231.

Rutkowski, J. (2003a). "Rapid Labour Reallocation with a Stagnant Unemployment Pool:
The Puzzle of the Labour Market in Lithuania," World Bank Policy Research Paper 2946,
Washington, D.C.

Rutkowski, J. (2003b). "Why I s Unemployment So High in Bulgaria?" World Bank Policy
ResearchPaper 3017, Washington, D.C.

Rutkowski, J. (2003~)."Does Strict Employment Protection Discourage Job Creation?
Evidence from Croatia," Background paper prepared for the Croatia Country Economic
Memorandum, World Bank, Washington, D.C.

Rutkowski, J. (2004). "Firms, Jobs and Employment in Moldova," World Bank Policy
ResearchPaper 3253, Washington, D.C.

UNDP (2004). Millennium Development Goals Report.

World Bank (2000). Making Transition Workfor Everyone: Poverty andInequality in Europe
and CentralAsia. The World Bank, Washington, D.C., 2000-524        pp.

World Bank (2002). Slovak Republic: Living Standards, Employment, and Labor Market
Study.

World Bank (2004). Doing Business in 2005.