Date post: | 08-Mar-2018 |
Category: |
Documents |
Upload: | phungnguyet |
View: | 217 times |
Download: | 3 times |
49ECB
Monthly BulletinOctober 2014
ART ICLES
the impact of the economic crisis on euro area labour marKets1
The economic crisis has had a heavy impact on euro area labour markets. A notable feature of the crisis throughout its duration has been the considerable degree of cross-country heterogeneity of labour market adjustments – with some economies emerging relatively unscathed, while others have seen steep and persistent increases in unemployment. This article analyses the impact of the crisis as a whole on euro area labour markets, paying particular attention to the different impacts of the two euro area recessions during the crisis and the interplay of sectoral and institutional features driving labour market outcomes. Despite ongoing structural reforms in some countries, progress has been partial and uneven across the euro area. Further reductions in labour market rigidities are necessary to increase the adjustment capacity of euro area labour markets and to speed up adjustment, thereby helping to reduce the current high levels of structural unemployment.
1 tWo recessions and their different impacts on euro area labour marKets
In comparison to the recessions experienced across the euro area countries since the 1980s,the impact of the economic crisis since 2008 has been particularly severe and long-lasting (see Chart 1). Six years after the start of the first euro area recession (which began in the secondquarterof2008),euroareaemploymentremains some 4% below its pre-crisis peak,five and a half million people have lost their jobs and the euro area unemployment rate,having risen from a pre-crisis low of 7.3% toa peak of 12.0% early in 2013, has declinedonlymodestlysincethen(seeChart2).Inpart,this strong impact reflects the systemic – and synchronised – nature of the initial economic crisis, financial crises typically having amuch larger and longer-lasting impact than non-financial recessions.2 However, it alsoreflects the interplay of sectoral and institutional featuresoftheeuroareaeconomies,whichhaveled to considerable cross-country heterogeneity inlabourmarketoutcomes,wherebytherehavebeen heavy and persistent job losses in some euroareaeconomies,butmodestandrelativelyshort-lived deteriorations in others.
1 This article summarises the work of an ad hoc ESCB task force charged with extending earlier analyses of the impact of the crisis on euro arealabourmarketstoincludethesecondeuroarearecession.Foramoredetailedversion,includingthebackgroundresearchinformingthisarticle, see“Comparisonsandcontrastsof the impactof thecrisisoneuroarea labourmarkets”,Occasional Paper Series,ECB,forthcoming.ThisarticlebuiltontheECB’s2012StructuralIssuesReportentitled“Euroarealabourmarketsandthecrisis”,Occasional Paper Series, No138,ECB,2012,whichwassummarisedinthearticleofthesamenamethatwaspublishedintheOctober2012issueoftheECB’sMonthlyBulletin,forwhichdatawereavailablegenerallyonlytotheendof2011andwhichthusomittedmuchoftheimpactof the second euro area recession.
2 SeeReinhart,C.M.andRogoff,K.S.,This Time Is Different: Eight Centuries of Financial Folly,PrincetonUniversityPress,2008,andmorerecently,“RecoveryfromFinancialCrises:Evidencefrom100Episodes”,NBER Working Paper,No19823,NationalBureauofEconomicResearch, January 2014.The economic and financial turmoil of 2008-09 affected virtually allwestern economies – albeittovaryingdegrees– concurrently,while earlier recessionshad tended tobemore localised, reflecting isolated economicor financialimbalanceswithinaffectedcountries.Inaddition,thecontractionineuroarearealGDPwasparticularlystrongoverthecourseofthecrisis (almost6%frompeaktotrough),andGDPhasstillnotreturnedtoitspre-crisislevel.
chart 1 euro area employment across recessions
(index: T=100 at cyclical peak in GDP; intervals are quarters)
92
94
96
98
100
102
92
94
96
98
100
102
T-8 T T+8 T+16 T+24
Q1 1980Q1 1992Q1 2008
Sources: Eurostat and ESCB calculations.
50ECBMonthly BulletinOctober 2014
To some extent, differences in outcomesreflect the different nature of the two recessionary “phases” of the crisis. The firstphase encompassed the deep and sharp global downturn in activity and trade (widely referred to as the five-quarter “Great Recession” of2008-09) and its aftermath, which affected alleuro area economies to some extent. The second phase refers to the longer-lasting “seconddip”(which resulted in a six-quarter recession for the euro area economy, beginning in the finalquarter of 2011, following the emergence ofsovereign debt concerns in some countries), in which adjustment was principally concentratedinthemost“stressed”economies.
Over the course of the Great Recession, allcountries saw some deterioration in their unemployment rates, with national increasesranging from 0.2 percentage point in Germany to 9.8 percentage points in Latvia. Six yearson, however, the range of outcomes is moremarkedstill(seeChart3).Bymid-2013,attheupperend,nationalunemploymentrateshadincreasedbysome19percentagepointsinGreeceand16percentagepointsinSpain,translatingintounemploymentratesof27%and26%,respectively.Overall,sevencountries(Ireland,Greece,Spain,Italy,Cyprus,PortugalandSlovenia)standoutashaving seen particularly large and persistent increases in their unemployment rates since the start of the crisis.Together,thesecountriesformthegroupmoststronglyaffectedbythefinancialmarketstress
chart 2 unemployment developments in the euro area
(percentages of the labour force; quarter-on-quarter changes in thousands)
-1
0
1
6
7
8
9
10
11
12
13
1999 2001 2003 2005 2007 2009 2011 2013
euro area recessions
quarterly evolution in the unemployment rate (right-hand scale)
euro area unemployment rate (left-hand scale)
Sources: Eurostat and ESCB calculations.Notes: Shaded bars indicate euro area recessions, defined interms of negative quarter-on-quarter GDP growth.
chart 3 changes in unemployment rates across the euro area
(percentages of the labour force; countries ordered according to their unemployment rate in 2008)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
EA NL CY AT SI LU EE MT IE FI IT BE FR DE GR LV PT SK ES
2008 2011 2013
Sources: Eurostat and ESCB calculations.
51ECB
Monthly BulletinOctober 2014
The impact of the economic crisis on
euro area labour markets
articles
(and are henceforth collectively referred to as the “stressed economies”). However, a simplecomparison of changes in unemployment rates over the course of the crisis obscures an important facet of the country-level developments observed. In particular,while all countries experiencedincreases(atleast,initially)intheirunemploymentratesasaconsequenceoftheGreatRecession,over thecourseof the secondphaseof thecrisis, fourcountries (Germany,Estonia, IrelandandLatvia)managed to reduce their unemployment rates. In Germany, these declines are likely toreflect ongoing improvements to labour market flexibility as a consequence of comprehensive reformsintroducedinadvanceofthecrisis.InEstonia,IrelandandLatvia,theyreflecttheearliertiming of the downturn and the swift and comprehensive measures introduced in response to the adverse labour market effects of the crisis.3
2 the concentration of Job losses over the crisis
The marked rise in euro area unemployment over the course of the crisis has been heavily concentrated temporally,sectorally,demographicallyandbycountry.Whilevirtuallyalleuroareaeconomieswereaffectedtosomeextentduringthefirstrecession,overthecourseofthesecondeuroarearecessionthebrunt of the job losses was (almost exclusively) borne by the stressed economies (see Chart 4).
TheGreatRecessionhadastrongsectoralbias(seeChart5),withahighproportionofemploymentlosses resulting frommarkedcontractions in industry-dependent sectors (suchasmanufacturing,transport and business services) and, in particular, in the construction sector. All the euro area
3 BothEstonia and Ireland increased spendingon active labourmarket policies to retrain and reintegrate theunemployed. In addition,employmentprotectionlegislationwaseasedinEstonia,whilesectoralwageagreementswerereformedinIrelandtomakethemmoreresponsivetoeconomicconditions.InLatvia,publicsectorwageswerecutsharply.
chart 4 euro area employment – stressed economies versus other economies
(index: Q1 2008=100)
70
80
90
110
70
80
90
110
100100
2008 2009 2010 2011 2012 2013
euro area recessions
euro areastressedothers
Sources: Eurostat and ESCB calculations.Notes: Shaded bars indicate euro area recessions during the crisis, defined in terms of negative quarter-on-quarter GDPgrowth. The stressed economies comprise Ireland, Greece,Spain,Italy,Cyprus,PortugalandSlovenia.
chart 5 euro area employment by sector
(index: Q1 2008=100)
70
80
90
100
110
70
80
90
100
110
2008 2009 2010 2011 2012 2013
euro area recessions
euro area totalmanufacturingconstructionservices
Sources: Eurostat and ESCB calculations.Notes: Shaded bars indicate euro area recessions during the crisis, defined in terms of negative quarter-on-quarter GDPgrowth. The stressed economies comprise Ireland, Greece,Spain,Italy,Cyprus,PortugalandSlovenia.
52ECBMonthly BulletinOctober 2014
economies were hit more or less proportionally, albeit with differences reflecting the sectoralcompositions of each economy. The downturn in the industry-dependent sectors reflected the strong downturninglobaltrade.Meanwhilethecreditcrunchhittheconstructionsectorparticularlyhard,leading to a sharp fall in construction activity across the euro area. The most acute impact was seen in countries undergoing the consequences of recently burst housing bubbles. During the second phaseofthecrisis,however,virtuallyallofthejoblossesobservedwereconcentratedinthestressedeconomies,whileemploymentremainedlargelystableorevenincreasedelsewhere.Inthestressedeconomies, job losses continued largely unabated in the industry and construction sectors, butintensifiedstronglyintheservicessector.Indeed,whereasnon-marketservices–includingpublicadministration and predominantly publicly provided activities (such as education and health care) – had continued to contribute positively to employment developments in virtually all countries during thefirstphaseofthecrisis, fiscal consolidation during the second phase led to a notable downturn inpublicsectoremploymentinsomeoftheeconomiesundertheseverestmarketstress,reinforcingthe employment contraction seen in the other sectors.4
employment losses by WorKer attributeEU Labour Force Survey data allow further breakdowns of employment and unemploymentdevelopmentsbygender,age,qualificationlevel,professionalstatusandcontracttype(seeChart7).5 Overall,men,youngerworkersandthelow-skilledhavebeenparticularlyhardhitbythecrisis.Thestronger impact on men than women doubtless reflects in part the heavy concentration of the crisis insectors(industry,construction,transport)inwhichmenaretypicallystronglyrepresented.Thispattern is repeated across countries and over the course of the crisis.
4 Seetheboxentitled“Theeffectofthecrisisonemploymentandwagesinnon-marketservices”,Monthly Bulletin,ECB,FrankfurtamMain,December2013.
5 Labour force surveydata forLuxembourgneed tobe interpretedwithparticularcaution,not least since they typicallyexcludecross-borderworkers,agroupthataccountsforroughly40%oftotalemploymentinLuxembourgandwhichwasparticularlyhardhitbythecrisis.Asaconsequence,employmentgrowthmaybeoverstated,andunemploymentdevelopmentsmaybeunderestimated.
chart 6 euro area employment by sector – stressed economies versus other economies
(index: Q1 2008=100)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
50
60
70
80
90
100
110
2008 2009 2010 2011 2012 2013
constructionnon-market services
industrymarket services
(i) Stressed economies (ii) Other economies
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
50
60
70
80
90
100
110
2008 2009 2010 2011 2012 20130.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
50
60
70
80
90
100
110
2008 2009 2010 2011 2012 2013
Sources: Eurostat and ESCB calculations.Note:ThestressedeconomiescompriseIreland,Greece,Spain,Italy,Cyprus,PortugalandSlovenia.
53ECB
Monthly BulletinOctober 2014
The impact of the economic crisis on
euro area labour markets
articles
A breakdown by age shows that young workers (aged under 25) and prime age workers (aged 25-54) have been considerably harder hit than older workers (aged 55 and over). (See also Box 1 on youth labourmarketdevelopmentsoverthecourseofthecrisis.)Tosomeextent,theongoinggrowthinemployment of older workers is likely to reflect increased financial needs as they replace wealth lossesexperiencedasaresultofthefinancialcrisis,6 as well as ongoing changes in several euro area
6 SeeDuval,R.,Eris,MandFurceri,D.(2011),“TheEffectsofDownturnsonLabourForceParticipation:EvidenceandCauses”,OECD Economics Department Working Papers,No875,OECDPublishing,Paris.
chart 7 employment developments in the euro area – disaggregated results
(cumulative losses; percentage point contributions)
Euro area Change in employment 2008-13
(i) Breakdown by gender
-4
-3
-2
-1
0
1
2
-4
-3
-2
-1
0
1
2
2008 2009 2010 2011 2012 2013
femalemaletotal
-251 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
-20
-15
-10
-5
0
5
10
15
20
-25
-20
-15
-10
-5
0
5
10
15
20
malefemale
total
1 EA2 LU3 MT4 DE
13 EE14 SI15 IE16 PT
17 ES18 LV19 GR
5 AT6 BE7 FR8 CY
9 NL10 FI11 SK12 IT
(ii) Breakdown by age group
-4
-3
-2
-1
0
1
2
-8
-6
-4
-2
0
2
4
2008 2009 2010 2011 2012 2013
55-64total
15-2425-54 25-54
15-24 55-64total
-251 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
-20
-15
-10
-5
0
5
10
15
20
-25
-20
-15
-10
-5
0
5
10
15
20
1 EA2 LU3 MT4 DE
13 EE14 SI15 IE16 PT
17 LV18 ES19 GR
5 AT6 BE7 FR8 CY
9 NL10 FI11 SK12 IT
Sources: Eurostat and ESCB staff calculations.
54ECBMonthly BulletinOctober 2014
chart 7 employment developments in the euro area – disaggregated results (cont’d)
(cumulative losses; percentage point contributions)
Euro area Change in employment 2008-13
(iii) Breakdown by status and contract type
-5
-4
-3
-2
-1
0
1
2
-5
-4
-3
-2
-1
0
1
2
2008 2009 2010 2011 2012 2013
temporary employeesself-employed permanent employees
total
-251 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
-20
-15
-10
-5
0
5
10
15
20
-25
-20
-15
-10
-5
0
5
10
15
20
1 EA2 LU3 MT4 DE
13 EE14 SI15 IE16 PT
17 ES18 LV19 GR
5 AT6 BE7 FR8 CY
9 NL10 FI11 SK12 IT
temporary employeesself-employed permanent employees
total
(iv) Breakdown by educational level
-8
-6
-4
-2
0
2
4
6
-8
-6
-4
-2
0
2
4
6
2008 2009 2010 2011 2012 2013
hightotal
lowmedium
-25
-20
-15
-10
-5
0
5
10
15
20
25
-25
-20
-15
-10
-5
0
5
10
15
20
25
1 2 3 4 5 6 7 8 9 101112131415161718191 EA2 LU3 MT4 DE
13 EE14 SI15 IE16 PT
17 LV18 ES19 GR
5 AT6 BE7 FR8 CY
9 NL10 FI11 SK12 IT
highmedium
lowtotal
Sources: Eurostat and ESCB staff calculations.
55ECB
Monthly BulletinOctober 2014
The impact of the economic crisis on
euro area labour markets
articles
countries to pension entitlements and retirement ages.7However,itisalsolikelytoreflectthestronginstitutionaldisparities insomeeuroareaeconomies, inparticularstrongemploymentprotectionlegislation forpermanentworkers,whichdiscourages the selective retentionofpotentiallymoreflexibleanddynamicworkersandpromotesdismissalsalong“last in,firstout” lines.Bydintofboth lower tenureandahigherpropensity tobeemployedon temporarycontracts,youngerandprime age workers are likely to have been less costly to dismiss than older workers (see panel (ii) of Chart 78 and the discussion in Box 2).
7 Itmayalsoreflectthegreaterexperienceandsectororfirm-specifichumancapitalembodiedinolderworkers,whichmakethemmorevaluable than younger workers to firms faced with lay-off decisions.
8 Disaggregating employment reactions to the two phases of the crisis by contract type (see panel (iii) in Chart 7) reveals the disproportionate impact of job losses on temporary workers in both phases of the crisis.
box 1
youth employment and unemployment durinG the crisis
Youth unemployment (among the under-25s) has risen substantially over the course of the crisis–fromaround15.4%in2007toaround24%bythemiddleof2013.Insomeeuroareacountries,theincreasehasbeenmoresubstantialstill,withyouthunemploymentrisingtoover45% in the stressed economies as a whole(seeChartA) and to 56% inSpain and 59%Greece by the middle of 2013. There are many reasons why youth unemployment rates are typically higher than aggregate unemployment rates,notleastthelowerexperienceandfirm-specific human capital of young workers and their lower participation rates (discussed below). However, the very rapid rise ofyouth unemployment over the crisis can also be partly explained by the typically higher representation of the under-25s among temporary workers, who are generally morevulnerable to cyclicality than permanent workers and who were disproportionately displaced from employment during the crisis.1 The rise in youth unemployment poses a particular challenge for euro area policy-makers, not only because of the possiblelong-term “scarring” effects2 of protracted unemployment spells at the beginning of young people’sworking lives on later career
1 See“Comparisonsandcontrastsoftheimpactofthecrisisoneuroarealabourmarkets”,Occasional Paper Series,Box2,Section2.1.2,ECB,FrankfurtamMain,forthcomingandtheboxentitled“Developmentsinyouthunemploymentineuroareacountriessincetheonsetofthecrisis”,MonthlyBulletin,ECB,FrankfurtamMain,February2014.
2 SeeArulampalam,W.,“IsUnemploymentReallyScarring?EffectsofUnemploymentExperienceonWages,”The Economic Journal,111(475),2011,pp.585-686,whichfinds“permanentscars”intermsofbothwagepenaltiesandre-employmentprobabilitiesfromprotractedunemploymentspellsearlyinyoungpeople’sworkinglives.
chart a unemployment rates and youth unemployment rates across the euro area over the course of the crisis(percentages of the respective labour forces)
0
5
10
15
20
25
30
35
40
45
50
0
5
10
15
20
25
30
35
40
45
50
euro area – total euro area – youth
other countries – youthstressed countries – youth
2007 2008 2009 2010 2011 2012 2013
Source:Eurostat(EULabourForceSurvey).
56ECBMonthly BulletinOctober 2014
and earnings development, but also because evidence from earlier recessions suggests thatthese protracted unemployment spells may lead to a higher propensity for discouragement and inactivityamongyoungpeople,thushavinganadverseimpactonlonger-termdevelopmentsinthe potential labour supply.
An alternative measure of youth unemployment – unemployment ratios
To some extent, simple comparisons of youth unemployment rates somewhat exaggerate theimpact of the crisis on youth labourmarkets, since the cohort typically includes twodistinctgroups with very different characteristics: the first group consisting of 15-19 year olds includes alargenumberofyoungpeoplewhoarestillatschoolorintraining;thesecondgroup,whichismadeupof20-24yearolds,maybelesslikelytobestill ineducationortraining,butmayhaveyettofindtheirfirstjob.Consequently,thefirstgrouptypicallyhasasignificantlylowerparticipationratethanboththelattergroupandthepopulationof(25-54year-old)“primeage”workers.3
Analternative–andpotentiallymoremeaningful–measureistheyouthunemploymentratio,which is computed as the ratio of young unemployed to the total cohort. Chart B shows that on thismetric,youthunemploymentseemstobesomewhatlesspronouncedthanissuggestedbyheadlinerates,butthatsubstantialdifferencesneverthelessremainacrosscountries,withyouth
3 In2012participationrates(i.e.theproportionofeachcohortactivelyengagedintheeuroarealabourforce)rangedfrom19.9%fortheunder-20sto64.2%forthe20-24year-olds,comparedwith78.1%for“primeage”workers.
chart b unemployment rates and unemployment ratios for young persons (aged 15-24) in 2013(unemployment rates as a percentage of youth labour force; unemployment ratios as a percentage of the total population aged 15-24)
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
unemployment rateunemployment ratio
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
1 EA2 GR3 ES4 IT5 CY
11 LV12 SI13 FI14 EE15 LU
16 MT17 NL18 AT19 DE
6 PT7 SK8 IE9 FR
10 BE
Source:Eurostat(EULabourForceSurvey).
chart c proportion of young people not in employment, education or training, by country, in 2007 and 2013(percentages of the total population aged 15-24)
0
5
10
15
20
25
0
5
10
15
20
25
unemployedinactive
1 EA2 LU3 NL4 DE5 AT
6 SI7 FI8 MT9 FR
11 BE12 LV13 SK14 PT15 IE
16 ES17 CY18 GR19 IT
10 EE
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Sources: Eurostat and ECB calculations.Notes: Left-hand bars represent 2007 averages and right-handbars 2013 averages.
57ECB
Monthly BulletinOctober 2014
The impact of the economic crisis on
euro area labour markets
articles
Low-skilledworkershavebeendisproportionatelydisplaced fromemploymentoverbothphasesofthecrisis,whereastheemploymentofhigh-skilledworkershaskeptonincreasinginallbuttheworst affected economies. While medium-skilled workers (those with secondary level education or equivalent trade certification) saw something of a reprieve in the rate of job losses during the recoveryineuroareaGDPbetweenmid-2009andlate2011,low-skilledworkersenduredongoingemployment losses. This divergent evolution of employment by skill level appears to have been particularlyacuteinthestressedeconomies,wherejoblossesamongthelow-skilledaccountforasubstantial part of the decrease in employment.
3 structural mismatch and structural unemployment
Attheonsetofthecrisis,theinitialstrong(3percentagepoint)riseintheeuroareaunemploymentratewasdrivenlargelybyincreasesinshort-termunemployment(seeChart8),asistypicalduringthe
unemployment ratios in the stressed economies standing at around four to five times higher than thoseofGermany,Luxembourg,theNetherlandsandAustria.
Young people not in employment, education or training
Lowerunemployment ratios to someextent reflect the fact that it is easier foryoungercohortsthanolderworkerstoexercisethe“outsideoption”ofstayingonineducationandtrainingduringperiods of economic downturn. That said, the numbers exercising this option over the courseof the crisis appear to have been lowest in those countries characterised by the highest youth unemployment ratios. Chart C combines the proportion of under-25 year olds who are unemployed with that of thosewho are not inmore productive activities (not in employment, education ortraining).Together, thesegroupsformacategory that is typicallyreferred to in the literatureasNEETs.Itshowsthat,asidefromthestrongdivergenceinyouthunemploymentratiosacrosstheeuroareacountries,youthinactivityratesarealsohighestinthosecountrieswhereunemploymentratesaretypicallyhigher.Asaconsequence,NEETratesincreasedamongthe15-24year-oldagegroupinalleuroareacountriesoverthecourseofthecrisis,withtheexceptionofGermany,MaltaandAustria.However,inmostcountries,thelargeincreaseintheNEETrateismainlyexplainedby a rise in the number of unemployed rather than by an increase in inactivity.
Concluding remarks
Despite diminished labour market prospects, young people who are not yet in education,employment or training nevertheless remain attached to the labour market and are looking forwork.While, in time, theEU“youthguarantee” initiative4 may help to provide access to work experience and productivity-enhancing training for young people who are currently only marginallyattachedtothelabourmarket,itisnosubstituteforwidereffortstoencouragemoreflexible labour markets. These will require a dismantling of harmful labour market dualities – including overly rigid employment protection legislation, which effectively reserves jobopportunitiesforincumbent“insiders”andthussignificantlyreducesyoungpeople’saccesstocompete in the labour market.
4 SeetheCouncilRecommendationof22April2013onestablishingaYouthGuarantee(OJC120,26.4.2013,p.1),whichrecommendsensuringthatallunder-25sareofferedsomeformofemployment,traineeshiporcontinuededucationwithinfourmonthsofleavingeducation or becoming unemployed.
58ECBMonthly BulletinOctober 2014
initialjob-sheddingphasesofrecessions.However,asthecrisistookhold,job-findingratesdeclinedmarkedly(seeBox2), leadingtolongerunemploymentspells.Thisraisedboththeunemploymentrate and the share of long-term unemployment (defined here as persons unemployed for 12 months or more). Chart 9 summarises the contemporaneous evolutions of the euro area unemployment rate and the share of long-termunemployment.With the onset of the secondphase of the crisis, bothmetrics deteriorated further, the unemployment rate rising by a further 2 percentage points,whilelong-termunemploymentrosefromaround45%(inlinewithitspre-crisisaverage)toaround52%oftotalunemployment.Bytheendof2013thestockoflong-termunemployedaccountedforover6%ofthetotaleuroarealabourforce,morethandoubleitspre-crisislevel,sothatmuchoftheprogressmade in reducing average unemployment spells from the mid-2000s had been reversed. From a policy perspective, themarked rise in long-termunemploymenthasbeenoneof themost serious labourmarket consequences of the crisis, since long unemployment spells may translate into structuralunemployment and thus a marked reduction in potential output in the longer term.
box 2
labour marKet floWs over the course of the crisis
ThisboxusesquarterlyLabourForceSurvey(LFS)data toanalyse labourmarketflowsacrosseuroareacountriesoverthecourseofthecrisis.Reflectingdataavailability,theanalysiscoverstwelve euro area countries (EE, IE, GR, ES, FR, IT, NL, AT, PT, SI, SK and FI) over theperiod up to (at least) the end of 2012. These data track changes in the labour market status of
chart 8 euro area unemployment rate by duration
(percentages of the labour force; shares of total unemployment)
2
4
6
8
10
12
2005 2006 2007 2008 2009 2010 2011 2012 2013
2
4
6
8
10
12
0 0
6 months or less euro area unemployment rate
2+ years 12-23 months 6-11 months
Sources: Eurostat and ESCB calculations.Notes: Long-term unemployment is defined as persons out ofwork for 12 months or more.
chart 9 evolution of the euro area unemployment rate and long-term unemployment share(percentages of the labour force; shares of total unemployment)
30
35
40
45
50
55
30
35
40
45
50
55
Q2 00
Q1 05Q1 06Q1 07
Q1 08
Q1 09
Q1 10
Q1 11Q1 12Q1 13
Q1 14
6 7 8 9 10 11 12 13
y-axis: long-term unemployment as a share of total unemployment
x-axis: unemployment rate
Sources: Eurostat and ESCB calculations.Notes: Long-term unemployment is defined as persons out ofwork for 12 months or more. Blue lines show the period from Q22000toQ12008(pre-crisis),redlinescoverthefirstpartofthecrisis,fromQ22008toQ120011,whilethelineswithcirclesrepresent the second part of the crisis and the subsequent recovery.
59ECB
Monthly BulletinOctober 2014
The impact of the economic crisis on
euro area labour markets
articles
individuals over the consecutive quarters they remain in the survey.1 To assess the impact of thedifferentphasesofthecrisis,developmentsin labour market flows are compared over three distinct periods: the pre-crisis period (from the first quarter of 2005 to the first quarter of 2008), the Great Recession and itsaftermath (from the second quarter of 2008 to the second quarter of 2011) and the sovereign debt crisis (from the third quarter of 2011 to the first quarter of 2013). TheLFSmicrodatainclude detailed information on worker and job characteristics, which permit analysis ofthe main determinants of worker flows. This analysis focuses on movements between employment and unemployment (job separation rates) and unemployment to employment (job-finding rates).
Chart A shows that, over the course of thecrisis,jobseparationrates2 – due to job losses and voluntary quits – increased for the euro area 113 from around 4.3% to 4.7% duringtheGreatRecession,withafurthermarginalincreaseinthesecondphaseofthecrisis.Atthecountry level, job separation rates rose sharply inEstonia, Ireland andSpain, and to a lesserextentinGreece,theNetherlands,SloveniaandSlovakiawiththeonsetoftheGreatRecession.Bycontrast,FranceandItalyshowamarkedlylowercyclicalsensitivity,withjobdestructionrateshardlychangingoverthewholeperiod.Forthemostpart,jobdestructionratesrosefurtheroverthesecondphaseofthecrisis.However,severaleconomies–Estonia,IrelandandAustria–appeartoshowasubsequentdeclineinseparationratesinthesecondphaseofthecrisis,albeittostill elevated rates compared with the pre-crisis period (with the exception of Austria). Analysis of worker characteristics shows that much of the sharp rise in job destruction rates in the first phaseofthecrisis–particularlyinIreland,SpainandEstonia,andtoalesserextentinGreece,Slovenia and Slovakia – can be attributed directly to the strong downsizing in the construction sector. 4
Differences by contract type
At thestartof thecrisis in2008, jobdestruction rates for temporaryworkers rosesharply, toreach almost10%of total temporary employment (on amoving averagebasis; seeChartB),andhaveremainedatsimilarorevenslightlyhigher levelseversince.Bycontrast,while job
1 TheselinkedLFSmicrodataareavailableonlyatcountrylevel.Flowsserieshavebeenprovidedbytherespectivenationalcentralbanks.2 Defined as the ratio of newly unemployed (who were employed one quarter earlier) to total employment.3 Portugal is not included in these aggregates since data are available only from the second quarter of 2011.4 Therearevarious reasons for themarkedcross-countrydifferences in the starting levelsof the flowdata,not least, labourmarket
institutions (including employment protection legislation), which can slow both the outflows from and inflows into employment. InGreece,arelativelylowratiooftemporaryemployeesalsoappearstoplayaroleinexplainingthelowjobseparationratesthere,sinceratesamongpermanentworkersaresimilartothoseofothereuroareaeconomies.(See,also,Section1.1.2ofthe2012StructuralIssuesReport,entitled“Euroarealabourmarketsandthecrisis”,Occasional Paper Series, No138,ECB,October2012.)
chart a flows out of employment into unemployment over the crisis
(job separation rates; percentages)
0
1
2
3
4
5
6
7
8
0
1
2
3
4
5
6
7
8
EA11 EE IE GR ES FR IT NL AT PT SI SK FI
Q1 2005-Q1 2008 Q2 2008-Q2 2011 Q3 2011-Q1 2013
Sources: Eurostat (EU Labour Force Survey microdata) andESCB calculations.Notes: Separation rates are computed as percentages of those employed in the previous quarter. Separations include voluntary quits.
60ECBMonthly BulletinOctober 2014
destruction rates among permanent employees also rose markedly at the onset of the crisis – from less than0.9% inadvanceof thecrisis to1.4% in2009,before settlingataround1.2%since then – they remain far lower than the destruction rates seen for temporary employees. Furthermore, jobseparationrates foreuroareaworkersofbothcontract typesappear tohaveremainedatelevatedlevelssincetheonsetofthecrisis,particularlyamongtemporaryworkers,despite the typically more limited fall in GDP over the second phase of the crisis.
Flows out of unemployment
Turningtothedataonflowsoutofunemploymentandfocusingonmovementsintoemployment,5 ChartCshows that inadvanceof thecrisis, roughly25%of theunemployedacross theeuroarea11foundajobineachquarter,butthatthisprobabilitydeclinednotablywiththeonsetofthecrisisandhasfallenfurtherstill–toaround15%–sincethesecondphaseofthecrisis.Atthecountrylevel,thisdownwardtrendhasoccurredacrossvirtuallyalleuroarealabourmarketsin the sample, with the exception of Estonia, where a cyclical recovery is evident. Amongthe countries most affected by the crisis, the probability of exiting from unemployment toemploymenthasdeclinedparticularlysharply,fallingfromalmost35%to15%inSpain,to10%inIrelandandtolessthan5%inGreece.Job-findingratesinItaly,PortugalandSlovakiahavealso shown notable declines.
Chart D shows that job-finding rates among the unemployed differ considerably according to unemployment duration. While the duration dependence of unemployment was already clearly
5 For theeuroarea11, flowsfromunemployment to inactivityappear tohaveshownamoderatedeclinesince thestartof thecrisis(although to a much lower degree than the decline in flows from unemployment to employment described in the text).
chart c flows from unemployment to employment over the crisis
(job-finding rates; percentages)
0
5
10
15
20
25
30
35
0
5
10
15
20
25
30
35
EA11 EE IE GR ES FR IT NL AT PT SI SK FI
Q1 2005-Q1 2008 Q2 2008-Q2 2011 Q3 2011-Q1 2013
Sources: Eurostat (EU Labour Force Survey microdata) andESCB calculations.Note: Job-finding rates are computed as percentages of those unemployed in the previous quarter.
chart b Job destruction rates by contract type, euro area 11
(job destruction rates; percentages)
0
2
4
6
8
10
12
0
2
4
6
8
10
12
2006 2007 2008 2009 2010 2011 2012
euro area recessions exits from permanent employmentexits from temporary employment
Sources: Eurostat (EU Labour Force Survey microdata) andESCB calculations.Notes:Theeuroarea11comprisesAT,EE,ES,FI,FR,GR,IE,IT,NL,SIandSK.Thedataarefourquartermovingaverages.
61ECB
Monthly BulletinOctober 2014
The impact of the economic crisis on
euro area labour markets
articles
visible in advance of the crisis (with job-finding rates typically considerably higher among those with lower unemployment spells than for those unemployed formore than a year), job-finding rates have fallen substantially for both groups over the course of the crisis. Country-levelanalysesrevealsimilarpatterns,albeit with some improvements visible in job-finding rates among the shorter-term unemployed in Estonia, Ireland and Finland.The significant downward trend in job-finding rates among those unemployed for 12 months or more warrants particular attention from policy-makers as it points to an elevated risk of a marked increase in structural unemployment across the euro area and potential hysteresis effects.
While many euro area economies have seen marked rises in long-term unemployment over the courseof thecrisis, thestressedeconomieshave,on thewhole, sufferedmuchsteeper increases(see Chart 10). Part of the explanation for this undoubtedly lies in the subdued labour demand conditions still prevalent in many of the stressed economies, but it may also partly reflect adivergence between the labour market characteristics of the unemployed and the skill needs of potentialemployers.Toillustratemoreclearlythedegreeofcross-countryheterogeneity,Chart11compares the contemporaneous evolutions of unemployment and the long-term unemployment share in Germany and Spain.
chart d Job-finding rates by unemployment duration
(job-finding rates; percentages)
0
5
10
15
20
25
30
35
40
0
5
10
15
20
25
30
35
40
unemployed for less than one year unemployed for one year or more
2006 2007 2008 2009 2010 2011 2012 2013
Sources: Eurostat (EU Labour Force Survey microdata) andESCB calculations.
chart 10 long-term unemployment in stressed economies and other economies
(percentages of total unemployment)
0.00.10.20.30.40.50.60.70.80.91.0
20
25
30
35
40
45
50
55
60
euro areastressed economiesothers
2007 2008 2009 2010 2011 2012 2013
Sources: Eurostat and ESCB calculations.
chart 11 evolution of the unemployment rate and the share of long-term unemployment: Germany and spain(percentages)
70
60
50
40
30
20
10
70
60
50
40
30
20
10
Q1 08
Q2 00
Q1 09
Q1 14 Q1 00
Q1 08 Q1 06Q1 09
Q1 10
Q1 12
Q1 14
y-axis: long-term unemployment (as a share of total unemployment)x-axis: unemployment rate (as a share of the labour force)
GermanySpain
0 5 10 20 25 3015
Sources: Eurostat and ESCB calculations.Notes: Long-term unemployment is defined as persons withoutjobs for 12 months or more. The lighter shaded parts of the lines show the pre-crisis period.
62ECBMonthly BulletinOctober 2014
Bothcountriesbeganthecrisiswithunemploymentratesofaround8%.However,inGermany,theadvent of the crisis led to little disruption in the downward trends seen in both the unemployment rate and the long-term unemployment share since the mid-2000s, in part as a consequence ofstructural reforms introduced at that time. Meanwhile, in Spain, the unemployment rate hasincreasedmorethanfourfold,whiletheshareoflong-termunemploymenthasrisenfromlessthanone-fifthtooverone-halfoftotalunemployment.Similar,albeitlesspronounced,patternsareseeninallofthestressedeconomies,suggestingthatthereareconsiderablebarrierstore-employmentinthese economies.
an outWard shift in the euro area beveridGe curve Beveridge curve analysis provides a simple and well-established approach to investigating the extent to which developments in unemployment may be the result of a transitory downturn in labour demand or a structural mismatch. Chart 12 shows the euro area Beveridge curve according to two measuresoflabourdemand:(i)euroareavacancyrates;and(ii)employers’perceptionsoflabourshortages.Priortothecrisis,thecounter-clockwisemovementsobservedintheaggregateeuroareaBeveridgecurvefromthemid-2000sreflectedatypicalbusinesscyclepattern,withunemploymentfallingasvacanciesincreased.However,astheGreatRecessiontookhold,strongdeclinesinlabourdemand resulted in a strong increase in euro area unemployment,with the euro areaBeveridgecurvemovingoutwards,reflectinglowvacancyratesandhighunemployment.
Duringtheinitialstagesofthecrisis,itwasnotclearwhetherthissimplyreflectedtypicalcyclicalmovements along a pre-existing Beveridge curve (and thus the transitory effects of low demand) or thefirstsignsofanoutwardshiftoftheBeveridgecurve,markingthestartofastructuralchangeintheunderlyingunemployment-vacancyrelationship.However,thepick-upinlabourdemandseenover the course of 2010 only generated a very small decrease in the euro area unemployment rate.
chart 12 evolution of the euro area beveridge curve over the crisis
(i) Based on Eurostat vacancy rates (ii) Based on DG-ECFIN labour shortages
06Q1
07Q1
08Q1
09Q1
10Q1
11Q1 12Q1
13Q1 14Q1
1
2
3
1
2
3
7 9 11
y-axis: vacancy ratex-axis: unemployment rate
99Q1
00Q1
01Q1
02Q1
06Q1
07Q1
08Q1
09Q1 10Q1
11Q1
12Q1
13Q1
14Q1
0
2
4
6
8
10
12
0
2
4
6
8
10
12
7 9 11 13
y-axis: labour shortagesx-axis: unemployment rate
Sources: Eurostat (harmonised euro area unemployment rate, job vacancy rate andmanufacturing employers’ perceptions of labourshortages) and ESCB calculations.Notes: Blue lines show the pre-crisis period from Q1 2006 in panel (i) and from Q1 1999 in panel (ii) to Q1 2008 in both panels; red lines representtheGreatRecessionperiodfromQ22008toQ22009;greenlinesdepictthesubsequentrecoveryfromQ32009toQ32011,while the black lines trace the evolution of the Beveridge curve since the onset of the second recession and during the subsequent recovery (i.e. from Q4 2011 to the latest observation).
63ECB
Monthly BulletinOctober 2014
The impact of the economic crisis on
euro area labour markets
articles
Moreover, thesecondrecessionaryepisode,starting in the finalquarterof2011, led toa furtherstrong increase in the unemployment rate even though aggregate vacancy rates remained elevated. Accordingly,Beveridgecurveanalysisshowsmountingsignsofentrenchedmismatchacrosseuroarealabourmarkets.VisualinspectionandeconometricanalysissuggestconsiderablediversityinBeveridgecurvemovementsat thecountry level,withstrongevidenceofnotableoutwardshiftsalso having taken place in Spain and France by a variety of metrics.9
evidence of sKill mismatchAn obvious factor that may help to explain the Beveridge curve movements observed over the course ofthecrisiswouldbeanincreaseinskillmismatch(thatis,thediscrepancybetweentheskillsoflabour force participants and the skill needs of employers) across the euro area. Analysis of the evolution of skill mismatch10 across 16 of the euro area economies (subject to data availability) suggestsanotable increase inskillmismatch in the initialphaseof thecrisisat regional,countryandeuroarealevel,irrespectiveofwhethermismatchismeasuredrelativetothelabourforceasawhole or simply by comparing the skills of those in work to those of the unemployed (see Chart 13). In both cases, the gap appears to be higher at the regional level than at the intra-country level,suggesting that at least part of the strong skill mismatch evident at euro area level could be significantly alleviated through higher inter-regional labour mobility.
9 Visual inspection suggested that several other candidates (e.g. Greece and Slovenia) showed clear outward shifts, while results forother stressed economies were often inconclusive owing to data limitations or the lag structure of the adjustment process. (See also Bonthuis,B.,Jarvis,V.andVanhala,J.,“What’sgoingonbehindtheeuroareaBeveridgecurve(s)?”Working Paper Series,No1586,ECB,September2013.)
10 Skillmismatch indices are computedas thedifferencebetween skill demandand supply at country and regional level,whereby skillsupply is approximated by the share of the labour force (or unemployed, respectively)with a given level of educational attainments(disaggregated according to the six discrete levels of the International Standard Classification of Education) and skill demand is proxied bytheeducationalattainmentsofthoseemployed.See,also,Section3.4of“Comparisonsandcontrastsoftheimpactofthecrisisoneuroarealabourmarkets”,Occasional Paper Series,ECB,forthcoming,andSection2.2ofthe2012StructuralIssuesReport,entitled“Euroarealabourmarketsandthecrisis”,Occasional Paper Series, No138,ECB,October2012.
chart 13 skill mismatch indicators for the euro area
(percentage differences)
euro area country region
(i) Employed compared with total labour force (ii) Employed compared with unemployed
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
2001 2003 2005 2007 2009 20110.00
0.01
0.02
0.03
0.04
0.05
0.06
0.00
0.01
0.02
0.03
0.04
0.05
0.06
2001 2003 2005 2007 2009 2011
Sources: Eurostat and ESCB calculations.Notes: The skill mismatch index (SMI) is computed as the difference between skill demand (proxied by the educational attainments of theemployed)andskillsupply(proxiedbytheeducationalattainmentsof the labourforceorunemployed,respectively).Thecountryindex aggregates 16 SMIs computed at country level across six skill levels. The region index aggregates SMIs computed at regional level.
64ECBMonthly BulletinOctober 2014
Country-based results suggest particularly marked and immediate increases in skill mismatch at thestartofthecrisisinIreland,Greece,SpainandPortugal,probablyreflectingasharpreversalofearlierconstructionbooms,withmismatchemergingsomewhat later(albeit toasimilardegree) inthe remaining stressed economies (Italy, Cyprus and Slovenia).While skill mismatch appears tohaveremainedsubduedoverthecrisisinsomeeuroareaeconomies(Belgium,GermanyandAustria,where it has even shown a marked decline beginning in the mid-2000s) or at least to have remained containedwithin its normal historical limits in others (France, Luxembourg, theNetherlands andSlovakia),Estoniaappearstobearemarkablecase,inwhichalltheadverseeffectsofthecrisisonskillmismatchwerereversedwithinjustafewyears,followingawaveoffar-reachinglabourmarketreformsadoptedfrom2009.Theseincludedamarkedeasingofemploymentprotectionlegislation,combined with a trebling of spending on active labour market programmes to retrain the unemployed.11
estimates of structural unemploymentThe strong increases in both long-term unemployment and measured skill mismatch give rise to important concerns related to structural unemployment. Estimates provided by international organisations–inparticular,theEuropeanCommission,theOECDandtheIMF–suggestthatthecrisishasresultedinanincreaseinstructuralunemploymentacrosstheeuroarea,fromanaverage(acrossinstitutions)of8.8%in2008,inadvanceoftheonsetofthecrisis,to9.4%in2010,followingtheGreatRecession,and,further, to10.3%by2013,followingtheemergenceofsovereigndebtconcerns(seeChart14).Overall,however,theseestimatessuggestthattheaverage1.6percentagepoint increase in structural unemployment represents around one-third of the almost 5 percentage point increase seen in theheadlineunemployment rate,while cyclicalunemployment representsaround two-thirds of the increase. In addition to the strong upward revisions to estimates for the euro
11 SeeBrixiova,Z.andEgert,B.,“LabourMarketReformsandOutcomesinEstonia”,IZA Discussion Paper series, No6336,IZA,Bonn,February 2012.
chart 14 structural unemployment estimates for the euro area
(percentages of the active labour force)
6
7
8
9
10
11
12
6
7
8
9
10
11
12
2000 2002 2004 2006 2008 2010 2012
European Commission IMF OECD actual unemployment rate
Sources: European Commission, Eurostat, IMF, OECD andESCB calculations.
chart 15 dispersion of structural unemployment estimates across countries
(standard deviations)
0
1
2
3
4
5
6
7
0
1
2
3
4
5
6
7
2000 2002 2004 2006 2008 2010 2012
European CommissionIMF OECD unemployment rate
Sources: European Commission, Eurostat, IMF, OECD andESCB calculations.Note: Dispersion is calculated as the standard deviation of differences between country-level structural unemployment estimates and the euro area average.
65ECB
Monthly BulletinOctober 2014
The impact of the economic crisis on
euro area labour markets
articles
area,thecrisishasalsoledtoaconsiderableincreaseincross-countrydispersion(seeChart15),reflectingmarkedincreasesinstructuralunemploymentestimatesforIreland,Greece,Spain,andPortugal–particularly since theadventof the sovereigndebtcrisis–whileBelgium,Germany,Austria and Finland show stable or slightly declining estimates.
Real-timeestimatesofstructuralunemploymentaresurroundedbyconsiderableuncertainty,varyingby institution, by methodology and over time, so that there are sizeable differences in estimatesfor some economies – in particular, for Greece, Cyprus and Portugal. However, themarked and consistent upward revisions of each subsequent release have been a feature common to all institutions.
4 WaGe adJustment over the crisis
Against the background of heavy employment losses, sharp rises in unemployment in somecountriesandlengtheningunemploymentspells,Chart16suggeststhatintheinitialphaseofthecrisis, euro areawage responses to labourmarket conditionswere rather limited,with all fourmain wage indicators continuing to grow strongly into 2009. For compensation per employee and negotiatedwages,thisongoinggrowthlargelyreflectedstipulationsinwagecontractsconcludedbefore the crisis, that is to say it was a consequence of the longer-run nature of collectivebargainingandindexationagreements.Forhourlywageseries,italsoreflectedthelargedownwardadjustment in hours worked that was observed in some euro area countries combined with the less-than-proportional reduction in compensation.
Unit labour costs rose sharply in 2008-09 (see Chart 17) on the back of robust wage growth and a strong decline in labour productivity. While some deceleration in the rate of wage growth was apparentbythestartof2009,itremainedinsufficienttopreventthelossofalmostfourmillionjobsacross the euro area over this period.
chart 16 euro area wage indicators
(annual percentage changes)
0
1
2
3
4
5
0
1
2
3
4
5
2002 2004 2006 2008 2010 2012
compensation per employee negotiated wages hourly labour cost compensation per hour
Sources: Eurostat and ESCB calculations.
chart 17 euro area labour cost indicators
(annual percentage changes)
-6
-4
-2
0
2
4
6
8
-6
-4
-2
0
2
4
6
8
2001 2003 2005 2007 2009 2011 2013
compensation per employee labour productivity unit labour cost
Sources: Eurostat and ESCB calculations.
66ECBMonthly BulletinOctober 2014
Onthewhole,thegrowthofcompensationperemployeeremainedrobust,averagingover2%peryearwellinto2012.Nevertheless,somesignsofgreaterwageresponsivenessbecameapparentassovereigndebtconcernsemerged,leadingtofurtherjoblossesinthestressedeconomies,withthegrowth in negotiated wages and compensation per employee declining markedly below its long-term average by 2010-11.
Chart 18 depicts a traditional Phillips curve relationship between annual changes in compensation per employee and unemployment rates at the country level in the pre-crisis period and for the twophasesofthecrisis.Itshowsthat,duringthefirstphaseofthecrisis(surroundingtheGreatRecessionof2008-09),theestimatedresponseofwagestochangesintheunemploymentratewaslowerthaninthepre-crisisperiod,butappearstohaveincreased(witha“steeper”Phillipscurve)in the second phase.12
Turningtowagedevelopmentsintheprivateandpublicsectorsacrosstheeuroarea,Chart19showsthat, while the growth rate of private sector hourly compensation rebounded in line with therecoveryaftertheGreatRecession,compensationgrowthremainedsubduedinthepublicsector,
12 These basic Phillips curves charts do not take account of other factors affecting wage developments (such as productivity or inflation). Nevertheless,econometricanalysisonthebasisofpanelestimatesforawageequationcoveringthemajorityoftheeuroareacountriesshowsthattheeuroareaischaracterisedbydownwardwagerigidities,withwagestypicallyshowingamoremutedresponsetochangesin unemployment during downturns (See Section 4.2 of “Comparisons and contrasts of the impact of the crisis on euro area labourmarkets”, Occasional Paper Series, ECB, forthcoming). However, the analysis finds that wages were increasingly responsive torising unemployment as the crisis continued, possibly as a result of “threshold effects” reflecting the largemagnitude of the rise inunemploymentinsomeeuroareacountries,theprotractednatureofthecrisis,incipientdownwardpressuresonwagesfollowingthewaveof labour market reforms introduced in some countries and ongoing public sector pay restraint resulting from fiscal consolidation.
chart 19 euro area compensation per hour – private and public sectors
(annual percentage changes)
-1
0
1
2
3
4
5
6
-1
0
1
2
3
4
5
6
2002 2004 2006 2008 2010 2012
private sector public sector
Sources: Eurostat and ESCB calculations.
chart 18 phillips curves for the euro area
(annual percentage point changes; annual percentage changes)
-8
-4
0
4
8
12
14
-8
-4
0
4
8
12
14
-5 0 5
y-axis: change in compensation per employee x-axis: change in unemployment rate
pre-crisis (2005-07) Great Recession and aftermath (2008-11) second recession (2012-13) pre-crisis Great Recession phase second recession phase
Sources: Eurostat and ESCB calculations.Notes: Data points are country-based contemporaneous relationships over the period shown.Latvia is excluded owingto data limitations.
67ECB
Monthly BulletinOctober 2014
The impact of the economic crisis on
euro area labour markets
articles
inpart reflecting fiscal consolidationefforts acrossmanyeuroareaeconomies, includingpublicsectorwage freezesandcuts insomeeuroareacountries.Moreover,as thesecondphaseof thedownturn continued, the growth in private sector compensation deceleratedmarkedly, bringinggrowth rates down towards the lower levels seen in the public sector over the trough of the Great Recession.However,stylisedfactsbasedonaggregatedataobscureanimportantelementofwagegrowthoverthecrisis–namely,theupwardimpactonaggregatewagesofemploymentcompositioneffects,reflectingtheheavyconcentrationofjoblossesamonglowerpaidworkers(includingthelowskilledandtheyoung).Therefore,comparisonsbasedentirelyonaggregatetrendsmaytosomeextent underestimate the increase in wage flexibility in the euro area in recent years.13
5 concludinG remarKs
The considerable increase in unemployment observed over the course of the crisis has been heavily concentrated in those euro area economies particularly affected by the financial market stress,wherethecrisisexposedsectoraloverheating,structuralimbalancesandlabourmarketrigidities.Thiscontributedtosharpfallsinoutputandemployment.Somegroups(theyoung,theunskilled,those on temporary contracts and those displaced from earlier overheated construction sectors) were particularlyhardhit.Moreover,thecrisishasledtoastrongincreaseinlong-termandstructuralunemployment in some countries.
Duringthefirstphaseofthecrisis,theestimatedresponseofwagesintheeuroareatochangesintheunemploymentratewaslowerthaninthepre-crisisperiod,butappearstohaveincreased(witha“steeper”Phillipscurve)asthecrisispersisted.Inthepresenceofhighunemployment,amorerapidand flexible response of wages to labour market conditions should help to restore competitiveness and encourage job creation. Further reforms to collective bargaining – which enable firm-level wageagreements tobetter reflect local labourmarketconditionsandproductivitydevelopments,and which allow for greater wage differentiation – would improve signalling mechanisms regarding demand for different types of worker.14
Labourmarketreformshavebeenparticularlyintenseinthosecountriesinreceiptofinternationalfinancialassistance(Ireland,Greece,Spain,CyprusandPortugal).Theseeffortsnotwithstanding,progress in labour market reform remains partial and uneven across the euro area. While the impact of reforms that have already been undertaken may take some time to produce their full effects,moremayberequiredtoachievethedegreeoflabourmarketflexibilitycompatiblewithmembership of a monetary union. Enhanced efforts to increase inter-regional and inter-country labour mobility across the euro area economies would help tackle high localised unemployment levels, thus reducing the risk that current high levels of unemployment translate into furtherincreasesinstructuralunemployment,andhelpalleviateemergingbottlenecksinstrongergrowingeuro area economies. Further reductions in employment adjustment rigidities and labour market dualities would also help to speed up the reallocation of employment to more productive sectors.
Countering the strong rise in long-term unemployment will require greater emphasis on (re-)activationpolicies,viaareprioritisationofactivelabourmarketpolicies–includingtargetedretraining measures – so as to enhance the employability of those displaced from permanently
13 Foramoredetailedanalysisofthisaspectofwageadjustment,see“Comparisonsandcontrastsoftheimpactofthecrisisoneuroarealabourmarkets”,Occasional Paper Series,ECB,forthcoming.
14 Some labour market reform recommendations were outlined in Mr Draghi’s speech at the 2014 Economic Policy Symposium atJacksonHole,entitled“Unemploymentintheeuroarea”.
68ECBMonthly BulletinOctober 2014
downsized sectors. Measures should focus in particular on the young and the less skilled in order to prepare these groups for new employment opportunities, help to alleviate the skillmismatchobservedandtargethigher-productivityactivities,allofwhichwillhelpspeeduptherestructuringprocess. However, while active labour market policies can help reintegrate young people andthe unemployed into employment and provide access to productivity-enhancing training and experience,theyarenosubstituteforthenecessarywidereffortstoencouragemoreflexiblelabourmarkets.
Finally, in order to reap the full benefits of labour market reforms, further reforms to productmarkets will be required in order to increase competition and the resilience of the euro area to futureshocks, thusavoidingthehighercostsof lostoutputandhigherunemploymentassociatedwith slower and more protracted adjustments.