American Economic Association From Sick Man of Europe to Economic Superstar: Germany's Resurgent Economy Author(s): Christian Dustmann, Bernd Fitzenberger, Uta Schönberg and Alexandra Spitz-Oener Source: The Journal of Economic Perspectives, Vol. 28, No. 1 (Winter 2014), pp. 167-188 Published by: American Economic Association Stable URL: http://www.jstor.org/stable/43193721 Accessed: 16-02-2016 18:48 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

American Economic Association is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Economic Perspectives.

http://www.jstor.org

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

- Volume 2014- Pages167-188 28, Number1- Winter JournalofEconomic Perspectives

From

Sick

Superstar:

Christian Uta

Man

of

Europe

Germany's

Dustmann,

Schönberg,

and

to

Economic

Resurgent

Bernd Alexandra

Economy

Fitzenberger, Spitz-Oener

the late 1990s and into the early2000s, Germanywas oftencalled "the sick man of Europe" (for example, Economist2004), a phrase usuallyattributed In to commentsby Czar Nicholas I of Russia about the troublesfaced by the Ottoman Empire in the mid-19th century.Indeed, Germany'seconomic growth averagedonlyabout 1.2 percentper yearfrom1998 to 2005, includinga recession in 2003, and unemploymentratesrose from9.2 percentin 1998 to 11.1 percentin 2005 (accordingto WorldBank data) . Today,afterthe GreatRecession,Germanyis describedas an "economicsuperstar"(forexample,in themovie"Made in Germany: Europe's Economic Superstar,"http://films.com/ItemDetails.aspx?TitleId=29218). Germany'snumber of total unemployed fell from 5 million in 2005 to about 3 millionin 2008, and itsunemploymentrate had declined to 7.7 percentin 2010 (accordingtodata fromGermany'sFederalEmploymentAgency,theBundesagentur fürArbeit). In contrastto most of its European neighborsand the United States, Germanyexperienced almost no increase in unemploymentduring the Great Recession,despitea sharp decline in GDP in 2008 and 2009 (an episode discussed ■ ChristianDustmannis Professor , University ofEconomics CollegeLondon, London, UnitedKingdom.BerndFitzenberger is Professor , UniverofStatisticsand Econometrics Uta is Associate sityofFreiburg, Economics, Freiburg,Germany. Schönberg Professor of is Professor University CollegeLondon,London,UnitedKingdom.AlexandraSpitz-Oener Humboldt is also Research Berlin,Berlin,Germany. ofEconomics, University Fitzenberger Associateat ZEW Mannheim,and Schönberg and Spitz-Oener are also ResearchAssociatesat theIAB, Nuremberg, Germany.Theiremailaddressesare [email protected], ac.uh,andalexandra.spitz-oener de,u.schoenberg@ucl. bernd.fitzenberger@vwl. uni-freiburg. @wiwi.hu-berlin. de. +ToaccesstheAppendix, visit http://dx.doi.org/10.1257/jep.28.L167 doi=10.1257/jep.28.1.167

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

168

JournalofEconomic Perspectives

in Möller 2010; Burda and Hunt 2011). Germany'sexports reached an all-time recordof $1.738 trillionin 2011, whichis roughlyequal to halfof Germany'sGDP, or 7.7 percentofworldexports.Even the euro crisisseems not to have been able to stop Germany'sstrengthening economyand employment. How did Germany,withthe fourth-largest GDP in the world (afterthe United and transform itself from "the sick man of Europe" to an States,China, Japan) "economic superstar"in less thana decade? One common answerpointsto a series oflegislativelabor marketreformsthatstartedin themid 2000s,theso-called"Hartz reforms." Anotherexplanationfocuseson theevolutionofGermany'seconomyand trade balance in the contextof the eurozone. However,we will argue that these factorsdid not playa decisiverole forthe transformation of the Germaneconomy, of labor in competitiveness the its market and the increase namely restructuring that has helped German exports.We instead present evidence that the specific governance structureof the German labor marketinstitutionsallowed them to react flexiblyin a time of extraordinaryeconomic circumstances,and that this distinctivecharacteristicof its labor marketinstitutionshas been the main reason forGermany'seconomic successoverthe lastdecade.1 We begin byarguingthatthe evolutionof Germany'sper unitlabor costs- that - in both the manufacturingsectorand the is, labor costs relativeto productivity othersectorsin theeconomyhas playedan importantrole in thefavorableevolution We then investigatethe mechanisms of German tradablemanufacturingindustry. and the dramaticdecrease in real wagesat thelower thatallowedforwage restraints end of thewage distribution. The specificfeatureof the Germansystemof industrialrelationsthatwe stress is thatit is not rooted in legislation,but insteadis laid out in contractsand mutual agreementsbetweenthe threemain actorsin Germany:employerassociations,trade whichis dominated unions,and workscouncils.The institutionell setupofthissystem, byindustry-wide wagebargaining,remainedbasicallyunchanged.However,manyindicatorsdemonstratethatitdid change in thewayitoperates.For example,theshareof Germanworkerscoveredbyanykindof union agreementhas sharplydeclined,and union agreementshas sharply deviationsfromindustry-wide thenumberoffirm-level increasedsince the mid 1990s. Overall,thesegradual changeswithinthe systemled to an unprecedenteddecentralizationof the wage-setting processfromthe industry one mayreferto thisprocess as an increasing level to the firmlevel.Alternatively, localizationof Germany'sindustrialrelations. The decentralizationin wage settingin Germanyis in contrastto manyof its neighborswhere the statutoryminimumwage is often high (relativeto productivity),where union wages and work hour regulationsapply to all firmsin the

1Our andSoskice tothatofCarlin inspirit issimilar (2008,2009),whoarguethatitisrestrucargument basedon employer-worker institutions German sector, usingtraditional private byGermany's turing forthe aretobecredited state reforms that and welfare labor market not and government cooperation, German recovery.

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

Christian Dustmann UtaSchönberg, andAlexandra , BerndFitzenberger, Spitz-Oener169

and whereinstitutionalchange thereforerequiresbroad consensusalong industry, the politicalspectrum. We then turnto a discussionof whyGermany'slabor marketexperience has been so distinctive withincontinentalEurope. On the one side, thefallof theBerlin Wallin 1989 and the dramaticcostof reunificationburdened the Germaneconomy in an unprecedentedway,leading to a prolonged period of dismalmacroeconomic performance.On the otherside, it gave German employersaccess to neighboring East European countriesthatwere formerly locked awaybehind the Iron Curtain, and thatwere characterizedby low labor cost,yetstable institutionsand political structures.These factorschanged the power equilibriumbetween employerand employeeassociationsand forcedthe latterto respond in a farmore flexibleway than manywould ever have expected. Finally,we discussthe relationshipbetween our analysisof the flexibility of Germany'slabor marketinstitutions and twoother events:Germany'sHartz reformsof 2003 and the arrivalof the euro in 1999. How Did Germany Improve its Competitiveness? RelativeUnitLabor Cost In Figure1,we plotthe"relativeunitlabor costs"fora country'soveralleconomy adjusted forthe changingcompositionof the marketsin whichit competes,fora selectionof countries,in dollar terms.This index is computed bythe OECD based on year-to-year changesof unitlabor costsand showsthe relativechange in the unit labor costsovertime (normalizedto 1995) translatedinto US dollarsat the current exchange ratecompared to a weightedaverageof a country'stradingpartners.The weightsof the tradingpartnersadjust annuallyto changes in tradingpatterns.An increasein thisindex indicatesa deteriorationof the competitiveposition.A drop - is caused by some in this index- that is, an improvementin competitiveness combinationof threefactors:1) a decrease in the wage per worker(or per hour); 2) an increasein productivity (per workeror per hour); and 3) a nominaldepreciationof a country'sforeignexchange rate. Since 1995, Germany'scompetitiveposition has persistently improved,while thecompetitiveness ofsome ofitsmain European tradingpartnershas deteriorated (Spain and Italy) or remained close to the 1995 position (France). The competitivenessof the United Kingdom has likewisedeteriorated,although it improved dramaticallybetween 2007 and 2009 due to the sharp depreciationof the British pound against other currencies.The US economy also lost competitivenessrelative to Germanyin the late 1990s as the US dollar appreciated in value relative to European currencies,but improvedconsistently afterthe 2001 recession,pardy achievedthrougha dollardepreciation(forinstance,whiletheeuro/dollarexchange ratewas around 1 in 2001, it had depreciated to 0.8 in 2009). However,Germany's withregard to France, Italy,and Spain cannot be due to gains in competitiveness currencydepreciation(and in factthe euro appreciatedrelativeto the currencyof most tradingpartners),because these countriesall share the euro, and so it must

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

1 70 JournalofEconomic Perspectives

Figure1 Evolutionof Competition-Weighted RelativeUnitLabor Costs, Selected Countries, 1994-2012

Indicators. Source: OECDEconomic intherelative i compared theannualchange unitlaborcostsofcountry Notes: Thisindexaccumulates where laborcostsaretranslated intodollars andthe ofitsmaintrading toa weighted average partners tothechangeintradepattern. Theannualchangeinlogsiscalculated areadjusted annually weights - Y. g¡jlAlog(ULC]t = (witLit)/Yit = Alog(ULClt whereULCit istheunitlabor as Alog(RULCit) etl) e]t) divided i inperiodt,computed as thetotalwagebillwitLit costforcountry bythevalueaddedofthe rateeit. intoUS dollars Theunitlaborcostsaretranslated Yit. usingtheexchange industry country's The ratesaredefined as indexrelative tosomebaseyear. Boththeunitlaborcostsandtheexchange markets ofcompetition inbothexport andimport account ofthestructure scheme g ¡Jltakes weighting ona year-by-year basis.SeeOECDEconomic Outlook anditadjusts ofthegoodssector ofthosecountries, andMethods Outlook Sources (2012,Issue2,No.92) andOECDEconomic (http://www.oecd.org/eco onthemethod ofcalculation. fordetails /sources-and-methods) relative have arisen because Germanwages grewat a slowerpace than productivity to theseothereurozone countries. Wage Trendsand Wage Inequality Figure 2 shows the evolutionof real wages in WestGermanysince 1990. The figureillustratesthe dramaticdevelopmentin wage inequalityin West Germany over the past 15 yearsor so (Dustmann,Ludsteck,and Schönberg 2009; see also and Sommerfeld2010; Card, Heining,and Kline 2013). 2 Antonczyk,Fitzenberger, Real wages at the 15th percentilefell dramaticallyfromthe mid 1990s onwards. From the early2000s onwards,median real wages startedto fall,and onlywages at the top of the distributioncontinued to rise. Notice that all wage figuresthat we report stand for West Germany (although, henceforth,we referto them as "Germany"),because developmentsin East Germanyare stronglyaffectedby the transitionafterGermanunification.

2Details with this Aavailable online onthewagedataareinAppendix journalathttp://e-jep.org.

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

s Resurgent FromSickMan ofEuropetoEconomic Economy 171 Germany Superstar:

Figure2 Indexed Wage Growthof the 15th,50th,85thPercentiles,WestGermany,1990-2008

Full-Time Workers between forWest German 20and60years Notes: Calculations basedonSIABSample ofthe15th, and85thpercentiles of ofage.Thefigure shows theindexed 50th, (log)realwagegrowth the consumer Nominal are deflated thewagedistribution, with1990as thebaseyear. wages using price Office. index(1995= 100)provided Federal Statistical bytheGerman Ifthe increasein wage inequalityand the modestgrowthin wagesoverall- and in particularthe dramaticdecline in real wages at the bottomof the wage distribution- has contributedto the favorableevolution of unit labor costs in Germany relativeto the United Statesand othereurozone countries,thenone should expect this development to have been particularlypronounced in the tradable manufacturingsector- the backbone of the German exportingindustriesaccounting for80 percent of German exports.This insightturnsout to hold true,but in an unexpectedway. To furtherexplore the increase in wage inequality,we classifysectorswith exportvolumesbelow the 25th percentileof the distributionof exportvolumesin 1995 as "nontradablesectors,"and thosewithexportvolumesabove thisthreshold as "tradablesectors.""Tradable manufacturing" are all those tradablesectorsthat belong to the manufacturingsector,and "tradable services"are all other tradable sectors.3Figure 3 breaks down the evolution of real wages along the wage 3Details ontheconstruction ofthesecategories canbefound inAppendix with thispaper A,available athttp://e-jep.org.

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

172 JournalofEconomic Perspectives

Figure3 Indexed Wage Growthof the 15th,50th,85thPercentiles,WestGermany,by Sectors,1990-2008

Notes: Calculations basedonSIABSample forWest German Full-Time Workers between 20and60years ofage.Thefigures showtheindexed ofthe15th, and85thpercentiles ofthe 50th, (log)realwagegrowth with 1990 as the base Nominal are deflated the consumer distribution, wage year. using wages priceindex Federal Statistical Office. PanelA shows theevolution ofthese (1995= 100)provided bytheGerman forthenontradable andpanelC fortradable services. sectors, figures panelBfortradable manufacturing, ofthedistribution Weclassify sectors with volumes below the25thpercentile ofexport volumes in export with 1995as"nontradable andthose volumes abovethisthreshold andthat tothe sectors", export belong as"tradable sector Thesectors abovethisthreshold that donotbelong to manufacturing manufacturing." as"tradable themanufacturing sector areclassified services." distributionseparatelyfor the nontradablesector,tradable manufacturingsector (henceforthdenoted as manufacturing),and tradable services sector. By this sectorrose at all percentilesof the wage measure,real wages in the manufacturing distributionuntil the mid 2000s and afterwardscontinued to rise at the median and the 85th percentile.Germany'sreal wages in the nontradablesector hardly increasedat all at anypartof the wage distributionduringthe 1990s and startedto so decline fromtheearly2000s onwardseven at the 85thpercentile,but particularly at the 15thpercentile.The sharpestincreasein inequalityoccurredin the tradable servicesector,wherebetween1990 and 2008 real wagesdid not showan increaseat the median,increasedby 12 percentat the 85th percentileand declined byalmost

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

Christian UtaSchönberg, andAlexandra BerndFitzenberger, Dustmann, Spitz-Oener1 73

15 percentat the 15thpercentile.At firstglance, thesefiguresdo not seem to lend supportto the hypothesisthatwage restraintin the manufacturingsectorwas an in thatsector. importantfactorin improvingcompetitiveness and Domestic Inputs Exports,Tradable Manufacturing, The end productin manufacturing, however,containsa large share of inputs in in the produced other sectors: Germany, value added in manufacturingis only roughlyone-thirdof the value of the end product,with the remainderof value added being contributedthroughinputs from other industries,either domesticallyor fromabroad (the literatureso far has focused on Germany'simportsof intermediateproductsfromabroad, see Geishecker2006; Sinn 2006; OECD 2007, sectormayhave benefited chap. 3; OECD 2012, chap. 3). Hence, themanufacturing fromlow wages in other domesticsectorsand fromcheap importsfromabroad. In addition, Germany'smanufacturingsector may have experienced increases in whichexceeded the increasesin wages in the manufacturing sector. productivity More detailed evidence suggeststhatboth of these factorsmaybe at play.4In sectorcomprised21.6 percentof alljobs in 1995, but Germany,the manufacturing 17.7 percent of all jobs in 2007, while the value added of thissector (in current prices) remainedessentiallyunchanged at 22.8 percentof all value added in 1995 compared with22.7 percent of value added in 2007. This patternsuggestslarger increasesin the manufacturing sectorthan in the othersectors,where productivity employmentshares increased over the same period, withvalue added remaining roughlyconstant.This patternis not uncommon across high-incomecountries.5 However,the share of manufacturingin output value (value of final products), as opposed to value added, rose steadilyfrom35 percentof all output in 1995 to 39.3 percent of all output in 2007. This patternreflectsthat the manufacturing sectorindeed relies to an increasingextenton inputsfromotherdomesticsectors and on importedinputs(because thesharein finalproductshas increasedwhilethe share in value added has remained the same), and may thus have benefitedfrom the lowwage growthin otherdomesticsectorsand fromcheaper imports. Diggingdownintothemore detaileddata,shownin Table 1, thevalue ofinputs over the value of output is nearlytwiceas high in manufacturingas in the other twosectors(66.1 percentin 1995 versus37.8 percentin the tradableservicesector) and this share increased by about 7 percentage points to 72.9 percent in 2007. The share of domestic inputs remained constantover the same period at about 51 percent. Thus, the increase in the share of inputs used by Germany'smanufacturingsector,relativeto the outputvalue in thatsector,is drivenby increased 4SeeTable1andTableAl in C available online with thispaperathttp://ejep.org fordetails Appendix anddatasources. 5Pilat, andWebb(2006)point outthattherelatively fast inmanufacOlsen, Cimper, productivity growth isassociated with relative declines oftheprices formanufacturing cost turing (thisisBaumol's products invalueaddedatcurrent understate theshareofvalueaddedatconstant Thus,shares disease). prices inmanufacturing tototalvalueaddedatconstant which makes itremarkable thatmanuprices prices, inGermany hasretained itsshareinvalueaddedatcurrent facturing prices.

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

1 74 JournalofEconomic Perspectives

Table1 Evolutionof the Share of Value of Total Inputsand Domestic Inputsover the Value of Output,Overall and bySector,1995-2007 Overall

Nontradable sectors Tradable Tradable services manufacturing

PanelA:ValueofTotalInputs/ Value Output 1995 48.2% 39.9% 2000 51.0% 37.9% 2007 53.2% 38.2%

66.1% 70.1% 72.9%

37.9% 41.4% 41.6%

PanelB:ValueofDomestic Value Inputs/Output 1995 39.8% 35.3% 2000 40.3% 32.2% 2007 40.5% 32.1%

51.7% 51.7% 51.2%

32.4% 34.8% 34.2%

PanelC:ValueofDomestic ofTotalInputs Inputs/Value 1995 82.6% 88.3% 2000 79.0% 84.9% 2007 76.1% 83.9%

78.1% 73.7% 70.3%

85.6% 84.0% 82.2%

Notes: Calculations basedon input-output statistics from theGerman Statistical Office 18, (Fachserie Reihe2,Years:1995-2007). Weclassify sectors withexport volumes belowthe25thpercentile ofthe distribution ofexport in 1995as "nontradable volumes sectors" andthosewithexport volumes above thisthreshold andthatbelongtothemanufacturing as "tradable sector Thesectors manufacturing." abovethisthreshold thatdonotbelong tothemanufacturing sector areclassified as"tradable services." use of inputsfromabroad relativeto inputs fromdomestic industries.However, even in 2007, 70 percentof overallinputsin Germany'smanufacturing sectorwere the that sector Thus, domesticallyproduced. argument Germany'smanufacturing has become nothingmore thanan assemblyplace forforeignproduced inputs(for has made example, Sinn 2006) is unjustified.In fact,whileGermanmanufacturing use of the share of domestic in increasing importedinputs, inputs manufacturing finaloutputvalue had remainedhigh and relatively stablebetween1995 and 2007. To whatextenthave Germany'sdomesticinputscontributedto competitiveness in its export-oriented sectorand the twoothersectors?In Figure4, manufacturing we plot the evolutionof unit labor costsin the threesectors,where industriesare When computing weightedwithrespectto theirexportsforthe twotradablesectors.6 added in the sector,as denoted by unit labor costs,we firstconsideronlythe value solid linesin Figure4. We thenconsiderfinaloutputvalue in thesector,whichis the sumofvalue added in thesectorand all inputsintothesectordenotedbydottedlines in Figure 4. This index (Unit Labor Costs: "End Products")incorporatesgains in in a sectordue to theusage ofinputsfromotherdomesticsectors.We competitiveness also plotmedianrealwages,adjustedusingGermany'sConsumerPriceIndex,forthe sectoris relatively threesectors.Whilerealwagegrowthin themanufacturing modest, 6Details A andinAppendix inAppendix canbefound onhowunitlaborcostsarecalculated C,which online with thisarticle athttp://e-jep.org. areavailable

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

'sResurgent FromSickMan ofEuropetoEconomic Economy 1 75 Germany Superstar:

Figure4 Evolutionof Real DailyWages and Unit Labor Costs bySector,1995-2007

Notes: Thefigures showindexed realmeandaily (baseyear1995= 100).Nominal wages bysector wages aredeflated Federal Statistical usingtheconsumer bytheGerman priceindex(1995= 100)provided Office. Thedataunderlying theseindices areincolumns also (1), (4),and(7) ofTableA2.Thefigure shows indexed unitlaborcostsboth"Valueadded"and"Endproduct" Thedataunderlying bysector. these indices areincolumns TableA2 (2),(3), (5),(6), (8),and(9) ofTableA2oftheonline Appendix. includes a detailed ofdataandmethods ofcalculations. description at about 8.2 percentover the 11-yearperiod,in the othertwosectorsaveragewages fellin real termsby1.2 and 4.1 percent,respectively, overthistimeperiod. As visiblein the figure,domesticunit labor cost fortotalproductionin manufacturing,takingaccount of inputs produced in other sectors ("end products"), declinedfarmorerapidlythanunitlabor costsin value added- a decline thatcannot be explained by the increase in the share of importedinputsin totaloutputvalue. Moreover,unitlabor costsin end productsstartto decline at thestartof theobservationperiod in 1995,whileunitlabor costsin value added decrease rapidlyonlyfrom 2003 onwardswhen mean wages,and in particularwages at the 15thpercentileof thewage distributionstartto decrease in real terms(as shownearlierin Figure3). Thus, Germany'smanufacturingsector improvedcompetitivenessin several drewon inputsfromdomesticallyprovidednontradable ways.First,manufacturing and especiallytradableservices,whererealwagesfellbetween1995and 2007. Second, the decline in unit labor costs,coupled withthe increase in mean real wages in increasesin the manufacturing sectorhave manufacturing, impliesthatproductivity

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

1 76 JournalofEconomic Perspectives

outpacedwageincreasesin thatsector.In comparison,totalunitlaborcostsfelllessin the nontradablesectors(minus22.2 percent)and much less in the tradableservices (minus9.7 percent),even thoughnominalwagesgrewmuch less in thesetwosectors Note also thatproductivity increasesin the compared to tradablemanufacturing. sector exceeded in have the increases the other two sectors. manufacturing Finally,to increasethe competitiveness of itsown finalproducts,the manufacturing sectorhas made increaseduse of trade integrationwithEasternEuropean countriesthrough inputs importedfromabroad, and far more so than other European countries. These inputsmade up 14.5percentoftotaloutputin themanufacturing sectorin 1995 and 21.5 percentin 2007.CalculatingtheoutsourcingindicatorsuggestedbyEggerand Egger (2003, p. 642) for Germany,France, and Italy regardingimportedinputs fromPoland, Hungary,and the Czech and the Slovak Republics,using data from the OECD Input-Output-Tables (at http://www.oecd.org/trade/input-outputtables .htm)and OECD InternationalTrade and Balance of PaymentsStatistics(at http:// shows that in the year 2000, importedinputsfromthese www.oecd.org/std/its/), fourcountriesamounted to about 8.5 percentof inputsin Germany,compared to 2.5 percentin Italyand 1.9 percentin France (relativeto GDP). The Increase in Competitiveness Institutions

and Germany's Labor Market

The movementsin Germanwages,withinand acrosssectors,belie thecommon are overlyrigid.Instead,we arguethat beliefthatGermany'slabormarketinstitutions thespecificgovernancestructureof theGermansystemofindustrialrelationsoffers In theearlyto mid 1990s,theseinstitutions allowedfor variousmarginsofflexibility. in of the an unprecedentedincrease thedecentralization(localization) processthat and setswages,hours,and otheraspectsof workingconditions,fromthe industryin firm or the which of the even level to the level worker, single single region-wide of distribution. at the lower end the to down wage wages particularhelped bring This decentralizationtook place even though the institutionalsetup of the domiwage bargainingbasicallyremainedunchanged. natingsystemof industry-wide The specificfeaturewhichwe stresshere is that the governancestructureof the German systemof industrialrelationsis not rooted in legislationand is not governedby the political process,but instead is laid out in contractsand mutual agreementsbetweenthe threemain labor marketparties:trade unions, employer associations,and works councils (the worker representativeswho are typically presentin medium-sizedand large firms).7For this reason, Germanywas in the positionto reactin an unprecedentedwayto the challengesof the early1990s. 7Works so whendemanded morethanfive with havetobesetupinestablishments councils employees inestablishments thathavemorethan50employees ofemployees About 92percent bytheemployees. inestablishments thatare ofemployees butonly18percent a works with inestablishments work council, andKrakel andWagner smaller Föhr, 1997;Beckmann, 2010). Schnabel, (Addison.

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

Christian Dustmann UtaSchönberg, andAlexandra , BerndFitzenberger, Spitz-Oener177

The principle of autonomyof wage bargainingis laid down in the German constitutionand implies that negotiations take place without the government minimumwage directlyexertinginfluence.As such, Germanyhas had no statutory imposed by the political process over the period we study.Rather,an elaborate systemofwage floorsis negotiatedperiodicallybetweentradeunions and employer at the industryand regionallevel. associations,typically This model of industrialrelationshas been verysuccessfulin Germany,where negotiationwith unions and participationof work councils in decision-making processes are widely regarded as an important cornerstone in furthering common interestsand even improvingproductivity. As a consequence, negotiationsare usuallyfarmore consensus-basedand less confrontationalthan in other countries.For example, Germanylost on average 11 days of workeach year per 1,000 employeesbystrikesand lock-outsbetween 1991 and 1999, but onlyfivedays per 1,000 employeesbetween2000 and 2007. These figuresforthe earlierand later timeperiod compare to 40 and 32 daysper 1,000 employeesin the United States, 30 and 30 daysin the United Kingdom,73 and 103 daysin France, 158 and 93 days in Italy,and 220 and 164 daysin Canada (Lesch 2009). Germany'scultureof common interestis dissimilarto the viewabout worker representations commonlyheld in the United States.A recentUS example is the of the attempt managementof the German companyVolkswagento introduce a workscouncil at its Chattanoogaplant in Tennessee. While the participationof workscouncils in managementdecisionsis consideredbyVolkswagenas a cornerstone of successfulfirmpolicythathelps furthering common interests,Tennessee GovernorBill Haslam has been outspokenin opposing anyunion formationat the plant,fearingthatit endangersthe state'seffortto attractinvestment(Greenhouse betweenUS and Germanlabor marketinstitutions lies in the 2013). A keydifference factthata workscouncilin Germanyelected bythe employeesdoes not have to be a union representative (althoughin practicethe majorityof workscouncilsare union whilethe installationof a workscouncil in a US firmautomatically representatives), involvesthe firmbecomingunionized.Thus, workscouncilsin Germanymayact in greaterindependencefroma union ifthesurvivalof theirfirmis at stake. Unions and EmployerAssociations In Germany,contractualagreementsbetweenunionsand employerassociations are negotiatedeitheron theregion-industry levelor on thefirmlevel.In additionto wages,workingtimeregulationsare an importantcomponentof the negotiations. A distinguishing featurefromUS labor marketinstitutions is thatthe recognitionof tradeunions in Germanyis at the discretionof thefirm,and union contracts coveronlytheworkersin firmsthatrecognizethe relevantsectoralwage bargaining (union) contract- regardlessofwhethertheworkeris a union member(fordiscussion, see OECD 2004; Dustmann and Schoenberg 2009; Fitzenberger,Kohn, and Lembcke 2013). Also, Germanfirmsthatonce recognizedthe union contractscan lateropt out at theirown discretion.Even withinunion wage contractsnegotiated at the industrylevel, there is scope for wage flexibility at the firmlevel through

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

1 78 JournalofEconomic Perspectives

so-called "opening" or "hardship"clauses, provided thatworkers'representatives agree (forexample,Hassel 1999; Hassel and Rehder 2001; Carlinand Soskice2009; Brändle, Heinbach, and Meier 2011; Bispinck,Dribbusch, and Schulten 2010). Afteropting out of a collectiveagreement,firmsstill have to pay wages for the incumbentemployeesaccordingto the collectiveagreementuntila newagreement at the firmlevel has been reached, but theydo not have to honor new negotiated wage increasesand the firmneed not followthe old collectiveagreementsfornew hires.Thus, overtimea firmmaybe able to lowerwage costsconsiderablybyopting out of the union contract- provideditsemployeesaccepted this. After1995,therewas indeed a dramaticdecline in union coveragein Germany. This decline is almost entirelydrivenby a decline in industry-wide agreements.8 From 1995 to 2008, the share of employeescovered by industry-wide agreements fell from75 to 56 percent,while the share covered by firm-level agreementsfell from10.5 to 9 percent.The percentageof Germanworkersthatwere not covered byan agreementin 1995-1997 was highestin the tradableservices(22 percent),as compared to tradablemanufacturing(9.8 percent) and nontradables(12 percent). By2006-2007, noncoveragehad sharplyincreasedin all threesectorsto 40, 27, and and nontradablesrespectively, 32 percentin the tradableservices,manufacturing, and thisshare continued to rise. By 2010, according to the German Structureof EarningsSurvey,41 percentof all employeesin firmswithat least 10 employeesin the sectorsManufacturing, Mining,and Servicesare not covered by any collective (StaBu 2013). wage agreement Has thisdecrease in union coverage ratescontributedto a reductionin wage growthand to an increase in inequality?We investigatethisquestion in Figure 5, wherewe plot the observedchanges in log real wagesbetween1995 and 2008 along the wage distribution.We also plot the counterfactualchanges that would have occurredifunionizationrateshad remainedat the same level as in 1995, using the reweightingapproach developed in DiNardo, Fortin,and Lemieux (1996), which essentiallyreweighswages observedin 2008 withthe odds-ratiothata workerwith in has been observed in the 2008-coverage-status specificobserved characteristics in 2008. Notice that this 1995 versusbeing observed in the 2008-coverage-status constructedcounterfactualexerciseis by no means "causal,"among otherreasons because it ignores general equilibrium effectsof de-unionization.The figure suggeststhatGermany'swages in 2008 would have been higherifunion coverage but the had remainedthe same as in 1995 throughoutthe entirewage distribution, differenceis particularly large at the lowerend of thewage distribution. WorksCouncils and Opening Clauses among employeescoveredbyunion Wage inequalityhas also increasedstrongly contracts,thussuggestingthattheGermansystemofindustrialrelationshas allowed forwage adjustmentseven withinthe unionized sector.This patternis illustratedin 8SeeData AandTableA3inAppendix online with thepaperathttp://e-jep.org. C,available Appendix

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

FromSickMan ofEuropetoEconomic 9sResurgent Germany Economy 1 19 Superstar:

Figure5 Observed versusCounterfactualWage Growthbetween1995 and 2008 along the Wage Distribution:The Role of De-unionization

Notes: Thefigure showstheobserved between 1995and2008,as wellas wagegrowth bypercentile thecounterfactual which wouldhaveprevailed iftheshareofworkers covered either wagegrowth or firm-wide hadremained at its1995level.The counterfactual byindustry-wide agreements wage distribution iscomputed andLemieux Fortin, usingthereweighting approach developed byDiNardo, arebasedintheLIAB. (1996).Calculations

Figure6, wherewe showthe evolutionof the 15th,50th,and 85th percentileof the indexed at 0 in 1995,forthosecoveredbya union (panel A) and wage distributions, uncoveredbya union (panel B) sectors.The figureshowsthatwage inequalityrose stronglyin the coveredsectorboth at the bottomand the top of the wage distribution,while in the uncovered sector it remained basicallyconstantat the bottom of the wage distributionand only increased at the top of the wage distribution. However,notice thatdue to the indexationthe figureshide the largerdifferentials in wage levelsat specificpercentilesin the uncoveredsectorrelativeto the covered sector:While the 85-50th and 50-15th differentials were on average 0.4 and 0.34 in the coveredsectorbetween1995 and 2008, theywereabout 0.5 in the uncovered sector.Thus, threefactorscontributedto the rise in overall inequalityduringthe timeperiod under consideration,namely,the shiftof workersfromthe covered to the uncoveredsector (whichled, due to the largerdifferencesin wage levelsin the uncoveredsector,to an increasein lowertailinequality), the increase in inequality in the coveredsector,and the increasein inequalityat the top of thewage distributionin the uncoveredsector.

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

180

JournalofEconomic Perspectives

Figure6 Indexed Wage Growthof the 15th,50th,85thPercentiles,WestGermany, byUnion Coverage Status,1995-2008

between Full-Time Workers 20and60years forWest German basedonLIABSample Calculations Notes: of and85thpercentiles ofthe15th, theindexed shows ofage.Thefigure 50th, (log)realwagegrowth aredeflated Nominal with 1995asthebaseyear. thewagedistribution, price usingtheconsumer wages Statistical Office. Federal index(1995= 100)provided bytheGerman

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

andAlexandra Christian BerndFitzenberger, UtaSchönberg, Dustmann, Spitz-Oener181

We think that the change in wage inequalityin the covered sector is due to the decentralizationof wage settingsince the beginning of the 1990s, when collectivebargainingcame under increasingpressurefromemployers industry-level and differentiated who demanded more firm-specific regulations.Workscouncils whichalso strengthaccommodatedthisdecentralizationto securejobs in Germany, ened theirrole in the industrialrelations.As another response,the trade unions and employers'associationsagreed on an increasingnumberof "opening clauses" in industry-level collectiveagreements.Opening clauses allow firmsto deviatefrom standards.At first,these opening clauses focused collectivelyagreed industry-wide on hoursofwork,but latertheyalso affectedwages.Also, the opening clauses were but latertheywere also implemented initiallyonlytemporaryto avoid bankruptcy, to ensure competitivenessin more general terms.Firmsthatuse opening clauses negotiatethe detailsconcerningpay and workingtimeagreementswiththe works council. Under German law, firmswithouta workscouncil cannot use opening clauses, but such firmsmay instead decide to stop recognizinga union contract. Firmswitha workscouncil not coveredbya union contractmayreach an agreement on wageswiththeworkscouncil. Brändle,Heinbach, and Meier (2011, Figure1) reportthatopening clauses for wagesonlystartedto gain importancein 1995 (opening clauses regardinghoursof workhad existedbefore 1995). Among industry-wide collectivecontractsin manuless than 5 involved clauses forwages in 1995, but this facturing, percent opening had risen to about 60 percentin 2004. Accordingto a surveyof workscouncils in 2005, about 75 percentof all firmswithcollectiveagreementsuse opening clauses (Bispinck2007; Bispinck,Dribbusch,and Schulten2010). To summarize,the specific governance structureof the German systemof industrialrelationsallowed foran unprecedented increase in the decentralization of thewage settingprocess,leading to a decrease in real wages,in particularat the lower end of the wage distribution.This was drivenby two main developments: 1) a sharp decline in the share of workerscovered by union agreements; and 2) an increase in opening clauses thatstrengthenedthe role of firm-basedworks councils in wage determinationrelativeto trade unions. This argumentis consistentwith the findingthat the rise in firm-leveldifferencesin wages contributes stronglyto the rise in wage inequalityin Germany(Antonczyk,Fitzenberger,and Sommerfeld2010; Card, Heining, and Kline 2013). What Led to Greater Flexibility in the German Labor Market? Whywere wage restraintsand decreasing real wages at the lower end of the wage distributionin Germanypossible afterthe mid 1990s but not before?After all, German firmshave alwayshad the option not to recognize a union contract and to pay wages below the union wage, provided theiremployeesaccepted this. Opening clauses had been possible before the mid 1990s. Our answer traces to the major changes in Germany'seconomy in the early 1990s related to the

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

182

JournalofEconomic Perspectives

reunificationof Germanyand the opening up of the nearbycentraland eastern European economies. On one hand, the extraordinarycost of German unificationburdened the Germaneconomyin an unprecedentedway,whichis partlyresponsibleforGermany's dismalperformancethroughoutthe 1990s and early2000s. The GermanCouncil of of about Economic Experts (SVR 2004, table 100, p. 644) estimatesnet transfers 900 billioneurosfromformerWestGermanyto East Germanyduringthetimeperiod 1991 to 2003. The totalsum of net transfers correspondsto about halfof one year's GDP in Germanyduringthattimeperiod.On theotherhand,theopeningofcentral and eastern European countries constituteda unique opportunityfor German industryto move productionabroad. They offereda stable investmentclimate,as well as (despite being locked awayforseveraldecades behind the Iron Curtain) a long historyof trade and interactionwithGermany.The structureof industryand whichsurvivedthe Soviet forinstance,shared manysimilarities, education systems, era. Vocationaltrainingplaysa keyrole in the education system,in a waysimilarto Germany,in countrieslike Hungaryor Poland. Germanwas also widelyspoken in partsofCentraland EasternEurope. Atthesame time,wagesin thesecountrieswere far lower than in Germany,and workingregulationsmore flexible (for example, Geishecker2006; Marin 2006). Movingproductionabroad to these countriestook place at a moderatepace: forexample, the stockof German foreigndirectinvestmentto Poland, Hungary,as wellas theCzech and theSlovakRepublicsamountedto about 1 percentofGermanGDP in 2000 and about 2.3 percentin 2010 (accordingto our calculationsand data fromhttp://stats.oecd.org/Index.aspx?DatasetCode=FDI thatGermanfirmsmightrelocate _POSITION_PARTNER).However,the possibility productionto these low-wagecountrieswas verycredible,and widelydiscussedin public (among German media outlets,see the articlesin DIHK 2003; Mihm and Knop 2004; Hawranek,Hornig,and Jung2004). The fiscal burden of German reunification,coupled with an immediately more competitiveglobal environment,made it increasinglycostlyfor German firmsto pay highunion wages.The new opportunitiesto move productionabroad, whileremainingstillnearby,changed the powerequilibriumbetweentradeunions and employerfederations,and forced unions and/or workscouncils to accept deviationsfrom industry-wide agreementswhich often resulted in lower wages for workers.In a similarvein, Burda (2000) predicted that the EU-accession of EasternEuropean countrieswould fostera reductionof labor marketrigiditiesin the old EU membercountries(includingGermany).Germany'sunions and works councils realized that theyhad to make concessions in order not to be further marginalized,and the specificcharacteristicsof the German systemof industrial allowed the tradeunions to adapt to the new economic realitiesand to institutions make these concessions.As a result,the German labor marketappeared to be far more flexiblethan manywould everhave expected. Whydid other continentalEuropean countriesnot react in the same wayas Germany?One importantreason is that the particularlydifficulteconomic situation in whichGermanyfounditselfin the early1990s was to a large partspecificto

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

FromSickMan ofEuropetoEconomic : Germany 'sResurgent Economy 183 Superstar

ofGermany,whichwas notfeltin otherEuropean Germany,due to thereunification countries.This was reinforcedbyGermany'sgeographicvicinityto the countriesof centraland easternEurope, which gave Germanyan earlytasteof the challenges of globalization.This decade of economic stagnationand hardship,when Germany was the "sickman of Europe," prepared the population foracceptingagreements forthe sake of economic growth,whichsaw inequalityrisedramaticallyforthe first timein the after-war period. In addition,the systemof industrialrelationsin other continentalEuropean countriesdoes not allowforthe same inherentopportunitiesof flexibleadaptation as the Germansystem.For example,in countrieslike France and Italy,union wages are oftenbargained at the national level and apply to all firmsin the economy, regardlessof whetherthe firmexplicitlyrecognizesthe union contract.Coverage stableat veryhighlevelsat about byunion wage contractshas remainedremarkably 90 percentin France and 80 percentin Italyduringthe 1990s and the 2000s (OECD in contrastto Germany,unionwagecontracts 2004,2012;Visser2013) . Furthermore, are typically extended to all workersin an industry(OECD 2004, table 3.4, p. 148; Visser2013, table4, pp. 96-98). In theseand othercontinentalEuropean countries, into collectiveagreementswould require politicalreformsat the adding flexibility national level. More generally,manyof the regulationswhich are determinedby labor contractsin Germanyare either legallyenforced in other countries (such as the minimumwage in France) or nationallyimplemented(forexample, union agreementsextend to all firmsin the economy), and thereforerequire consent on a much higherlevel (nationally,or even on the politicallevel) to be modified and changed. There is much less scope in these countriesfor a decentralization of wage setting(and other aspects of workingconditions) withintheirsystemof industrialrelations. In general, the decentralizationof union agreements is certainlybeing discussed more widelyacross Europe, but whetheror when such changes might occur more widelyremainsuncertain. Discussion

and Outlook

We have argued thatthe remarkabletransformation of the German economy fromthe "sickman of Europe" to a lean and highlycompetitiveeconomywithin littlemore thana decade is rooted in the inherentflexibility of the Germansystem of industrialrelations.This systemallowed German industryto reactappropriately and flexiblyover timeto the demands of German unification,and the global chalbecame only lenges of a new world economy. However,this intrinsicflexibility evidentunder the extraordinary difficult economic circumstancesand the extreme duressin whichGermanyfound itselfin the decade afterreunification.How does our thesisfitwithtwootherpossibleexplanationsforGermany'sincreasedcompetitiveness:Germany'sHartz labor marketreformsof 2003, or the changes brought about bythe adoption of the euro?

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

184

JournalofEconomic Perspectives

Germany'sgovernmentunder Gerhard Schröder implementedthe so-called "Hartz Reforms"to the labor marketsin 2003, which are oftencreditedfor spurring Germany'seconomy (for example, Rinne and Zimmermann2012, 2013; see Fitzenberger2009 for a criticalassessmentof the Hartz Reforms).These reforms controversial at thetime.Theyreducedand limitedthebenefitswhile wereextremely liberalized agencywork,reformed"active"labor marketpolicies,and unemployed, Labor the Federal Agency,but did not make anyinstitutional changesin reorganized thewage settingprocess. The Hartz reformswere implementedstartingin 2003, hence nearlya decade afterthe process of wage decentralizationand the improvementin competitiveness had begun in Germany.It seems plausible thatthe changes alreadyunderway in Germany'slabor marketshelped in preparingthe politicalgroundforthe Hartz reforms.In addition,as the enumerationof the main componentsof the reforms makes clear,the scale of the reformsis modest enough thattheyseem unlikelyto have triggeredthe dramaticincrease in competitivenessor the enormous drop in German unemploymentor to have led Germany'slabor marketthroughthe deep recession in 2008-2009. Further,while the focus of the reformswas on creating incentivesforseekingemployment,theydid littleto supportthe remarkablewage restraintwitnessedsince the mid 1990s,which is the keyfactorin explaining the gain in competitiveness. We thereforebelieve that while the Hartz reformshave contributedto the recentdecline in long-termunemploymentand to the continued increase in wage inequality at the lower end of the wage distribution,theywere not central or essential in the process of improvingthe competitivenessof German industry. Moreover,although one sometimes hears the argument that other continental European countries should musterthe political will to adopt theirown version of the Hartz reforms,we believe thatsuch a recommendationmaybe misleading. In our view,the specific governance structureof the German systemof industrial relations- activatedunder extreme duress- is what paved the way for the remarkable decentralizationof wage determinationfrom the industrylevel to the level of the single firmor singleworker,and which togetherwitha significant increase in productivity ultimatelyimprovedGermany'scompetitiveness.Whether the political process would have been able to achieve a similardegree of wage decentralization,had the autonomyof wage bargainingnot existed in Germany, is doubtful.In our view,the policy recommendationfromGermanyfor the rest of continentalEurope should not be the Hartz reforms(the advice givenoftenby policymakers,as in a February2013 speech byGerman Chancellor Angela Merkel reportedin de Weck 2013), but reformsthatwould targetthe systemof industrial relations by decentralizingbargaining to the firmlevel while keeping workers' representativesinvolvedto secure that employees benefitagain when economic conditionsimprove. Some argue that the adoption of the common European currencyis a main factorthathas helped Germanyto improvecompetitiveness. Again,we believe that but been a the arrivalof the euro mayhave contributingfactor, not the main one.

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

andAlexandra Christian Dustmann UtaSchönberg, , BerndFitzenberger, Spifa-Oener185

First,recall thatGermanywas shiftingitslabor marketinstitutionsand improving its competitivenessduring the mid 1990s, and the euro did not startuntil 1999. Second, withinthe common currencyarea, and after2001, Germanycontinued to gain competitivenesswith respect to its main trading partnerssuch as Italy and Spain. Third, the euro has persistentlyappreciated against the US dollar, leading to the increase in competitivenessof the United States as we illustrated in Figure 1. It seems unlikelythat Germany'sdeutschemark(if the euro had not been introduced) would have appreciated much more againstthe dollar than the euro has, at least not beforethe startof the global financialcrisisaround 2008 and the ongoing European debt crisis.Finally,it is not clear whetheran appreciation of a German currency(which probablywould not have taken place before 2008) would have had a dramaticimpact on Germany'soverall competitivenessat least in the medium-term, because it would also have made importedinputsless costly and it would possiblyhave fosteredeven strongerlabor marketadjustmentsof the typewe have described above. For example, the depreciation of the British pound by nearly30 percent in 2008-2009 has done littleto help UK manufacturingexports. Of course, the existenceof the common euro currencyarea raises a number of issues forcountrieswithinthe eurozone. Withoutthe possibilityto depreciate national currencies,the only way for countries like France, Italy,and Spain to gain competitivenessrelativeto other countriesof the eurozone is to reduce unit labor costs- that is, by increasing productivityrelative to real wages. Whether these countries will succeed in this endeavor remains an open question. The more centralizedand legallyanchored nature of theirlabor marketinstitutions, in comparison to Germany,does put them at a disadvantagein making such an adaptation. Boeri (2011) providesan assessmentof the politicaleconomyof labor market reformswith a particular focus on countries of southern Europe. He argues that the political process oftenallows only for two-tierreforms(affecting onlya subsetof all employees) instead of complete reforms,whichmaynot result in an increase of competitiveness The risein inequalityin Germanyhas led to an intensivedebate about itssocial consequences,and itseffecton povertyand socialjustice. For example,recentnegotiationsbetween employersand employee associations in Germanysuggest that futurewage settlementswilltryto make up forthe loss in real wages manyworkers experiencedin recentdecades. It is also likelythatcertainaspectsoflabor and wage regulationswillin the futurebe "put in legislativestone."As one example, the new coalition governmentin Germanywill introducea nationallylegislatedminimum forGermanyto relyon itssystemofindustrialrelationsto wage.Thus, thepossibility its improve competitivepositionbyhavinga decentralizeddecision makingprocess maybe cut back, and thismayrestrictGermany'sabilityto react in similarwaysto futureeconomic challenges.If thatoccurs,then futuregains in German competitivenesswillneed to be accomplishedratherthroughincreasesin productivity that increases. This to in the outstripwage patternmayhelp bringconvergence competitivenessof the countriesin the eurozone.

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

186

JournalofEconomic Perspectives

■ ChristianDustmannis grateful and stimulating discussionswithWendy forinsightful on earlier . We also comments Carlin. WethankWendy drafts benefited fromexcellent for and Marina research assistance Furdas,Stefanie Licklederer, OlgaOrlanski, FlorianWeiss. by BerndFitzenberger dedicateshis contribution forthispapertoJohnPencavel,hisfabulous teacher in laboreconomics.

References auf die europäischen andJoaquim der EU-Osterweiterung T.,ClausSchnabel, Addison, John imOsten undWesten unemofMandatory Arbeitsmärkte 1997. "OntheDeterminants [More Wagner. - theprice The ofeastwards inGermany." Industrial Relations ployment Councils Works expansion? oftheeastwards oftheEU on 36(4):419-45. expansion impact in theEastandWest]." labormarkets andKatrin European Bernd Dirk, Fitzenberger, Antonczyk, desVereins Duncker & Humblot, Serientitel: the Sommerfeld. 2010."Rising Schriften WageInequality, Neue andtheGender fürSocialpolitik Decline ofCollective 274,79-102. Folge Bargaining, andJennifer Hunt2011."What Economics Michael, Burda, 17(5):835-47. WageGap."Labour Miracle inthe LaborMarket theGerman andMatthias Explains Föhr, Beckmann, Michael, Sylvia onEconomic KräkeL 2010."Rent Papers Brookings Security GreatRecession?" Seeking, Employment for and Evidence Councils: andWorks 42(1):273-335. Activity Theory Kline. andPatrick Review Business Schmalenbach Card,David, 62(1):2-40. Joerg Heining, Germany." and theRise Reinhard. 2007."Löhne,Tarifver- 2013."Workplace Heterogeneity Bispinck, in Deutschland ofGerman und Tarifsystem of Quarterly Journal WageInequality." handlungen Economics WSI Discussion 1995-2005." 128(3):967-1015. PaperNo. 150, Rate 2012."RealExchange undSozialwissenschaftliches WirtschaftsInstitut, Carlin, Wendy. andFiscal Düsseldorf. Institutions, Adjustment, Wage-setting 'sFirst Lessons oftheEurozone and Stabilization HeinerDribbusch, Reinhard, Policy: Bispinck, Studies oEconomic Collective Decade." Schulten. Thorsten 2010."German 59(3):489-519. CESif and David Soskice.2008. Continuous ina European Carlin,Wendy, Perspective: Bargaining andEconomic Macroeconomic of Multi-Employer"Reforms, Erosionor Re-Stabilisation Policy In in Economic Performance No. Discussion WSI 171, Policy Germany." Paper Agreements?" editedby and Europe, undSozialwissenschaftliches WirtschaftsInstitut, Proposals forGermany andJochem 72-118. Ronald Düsseldorf. Schettkat, Langkau, and Reforms Boeri,Uto.2011."Institutional Routledge. and David Soskice.2009. LaborMarkets." inEuropean Dualism Carlin,Wendy, Chap.13in Performance: Economic andDavidCard, eds.,Handbook "German Ashenfelter, Disentangling Orley, Macroeconomic Role of the North and 4B. 4A Volume Labor Economics Reforms, , Supply-side of Institutions." Elsevier. Economy Holland, Policyand Coordinated Review Socio-Economic andM.F.Meier. 7(1):67-99. T.,W.D. Heinbach, Brandie, andJens Sebastian in Deutschland: Findeisen, 2011. "Tarifliche Dauth, Wolfgang, Öffnung " theFar East and Rise of the "The Suedekum. 2012. . Determinanten, Ausmaß, Auswirkungen Journal andTradeIntegraLaborMarkets East* German Research Labour Market 44(1-2):163-72. for No.6685. IZADiscussion tion." der ArbeitsloseMichael. 2000."Mehr Burda, Paper CitesEast 2013."Merkel ZurAuswirkung de Weck, Preisfürdie Osterweiterung? Joseph.

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

: Germany 'sResurgent FromSickMan ofEuropetoEconomic Economy 187 Superstar

inGermany?" Workers' World Lessons for EU'sProblem States. German Jobs Economy Bloomberg, Manual com.February 19. http://www.bloomberg.com 29(5):559-83. Steven.2013. "VW and Its 13-02-1 /news/20 8/merkel-cites-east-german Greenhouse, .htmi . Workers a Unionata Tennessee Plant." -lessons-for-crisis^vvracked-euro-states Explore 6.http://www.nytimes DIHK [Deutscher Industrieund Handels- NewYork Times, September 2003. "Produktionsverlagerung .com/2013/09/07/business/vw-and-auto-workers kammertag]. von als Elementder Globalisierungsstrategie -explore-union-at-tennessee-plant.html. Unternehmen: einerUnternehHassel,Anke.1999."The Erosionof the Ergebnisse Mai 2003." ["Offshoring of German of Industrial Relations." British System mensbefragung as PartoftheGlobalization Industrial Relations 37 : 483-505. Production (3) Strategy Journal of ofFirms: Results ofa FirmSurvey inMay2003, andBritta Rehder. 2001."InstiHassel, Anke, tutional intheGerman Berlin."] http://www.dihk.de/ressourcen Change WageBargaining - The Roleof Big Companies." MPIfG System /downloads/produktionsverlagerung.pdf. Nicole and Thomas 01/9. DiNardo, Fortin, John, Working Paper Lemieux.1996. "LaborMarketInstitutions Frank andAlexHawranek, Dietmar, Hornig, and theDistribution of Wages,1973-1992: A ander 'MadeinGermany.'" 2004."Bye-bye Jung. Econometrica DerSpiegel 44:94-109. 64(5): Semiparametric Approach." 1001-1044. Peter,and StephanSeth. Jacobebbinghaus, Datafrom the 2010."Linked Dorner,Matthias, Jörg Heining,Peter Employer-Employee and StephanSeth. 2010. IAB:LIABCross-sectional (Model2,1993-2008)." Jacobebbinghaus, ofIntegrated Labour Market 05/2010. "Sample Biographies FDZ-Datenreport undEntwick01/2010. Lesch,Hägen.2009."Erfassung (SIAB)1975-2008." FDZ-Datenreport andUtaSchönberg. in OECD-Ländern." IW Trends, 2009. Dustmann, Christian, lungvonStreiks andUnionWages." Review derdeutschen Institut Köln. Wirtschaft, "Training ofEconomics 1/2009, andStatistics 91(2):363-76. Marin,Dalia. 2006."A New International and Division ofLaborin Europe:Outsourcing and Dustmann, Christian, Ludsteck, Johannes Uta Schönberg. theGerman Offshoring toEastern Euro2009."Revisiting Europe." Journal ofthe Association 4(2/3):612-22. WageStructure." Quarterly Journal ofEconomics peanEconomic and Carsten 124(2):843-81. Mihm,Andreas, Knop.2004. Economist on theMend." Bis ich mitder Gewerkschaft 2004."Germany "Outsourcing: November 17. http://www.economist.com/node klarbin,hat die Konkurrenz schongeliefert UntilI havereachedan agree/3352024. [Outsourcing: ment with theunion, thecompetitor hasalready Egger,Harmut,and PeterEgger.2003. andSkillSpecific ina FAZ[Frankfurter delivered]." "Outsourcing Employment Allgemeine Zeitung], SmallEconomy: Austria after theFalloftheIron March21. http://www.faz.net/aktuell/wirtschaft Curtain." Economic 55(4):625-43. Oxford Papers /outsourcing-bis-ich-mit-der-gewerkschaft-klarBernd. 2009."NachderReform -hat-die-konkurrenz-schon-geliefert-1 144367.html. Fitzenbeiger, ist vor der Reform? Eine arbeitsökonomische Möller, Labor 2010."TheGerman Joachim. derHartz-Reformen." MarketResponsein the WorldRecession Analyse ausgewählter Aspekte - Die wichtigstenDe-mystifying In Reformen a Miracle." Market für Deutschland Journal forLabour aus ökonomischer editedby Research 42(4):325-36. Sicht," Handlungsfelder Günther Schulze Schäffer-Poeschel OECD.2004, Outlook. 2012.Employment 2007, pp.21-48. Verlag. Karsten Bernd,KarstenKohn,and Pilat,Dirk,AgnèsCimper, Olsen, Fltzenberger, Alexander C. Lembcke. 2013."UnionDensity andColinWebb.2006."TheChanging Nature andVarieties ofCoverage: TheAnatomy ofUnion ofManufacturing inOECDEconomies." OECD, in Germany." Industrial Labor RelaSTIWorking WageEffects Paper2006/9. tions Review 66(1): 169-97. 2012. Rinne, Ulf,andKlausF. Zimmermann. Karsten andQingwei "Another Economic Miracle? TheGerman Labor Bernd, Kohn, Fltzenberger, ofUnionMembership Market 2011."TheErosion andtheGreat Recession." IZAJournal Wang. of inGermany: Determinants, Densities, 1/3. Decomposi- Labor Policy tions, Economics F.Zimmermann. 2013."Is 24(1):141-65. Rinne, Ulf,andKlaus "/owrmz/ ofPopulation Hermann.2005. "The Imputa- Germany theNorth StarofLaborMarket Gartner, Policy?" tionof Wagesabovethe Contribution Limit IMFEconomic Review 61(4): 702-729. withthe GermanIAB Employment 2006."ThePathological Sinn,Hans-Werner. Sample." 02/2005. Howto FDZ-Methodenreport ExportBoomand theBazaarEffect: Puzzle." World Geishecker, Ingo.2006."DoesOutsourcing SolvetheGerman 29(9): Economy to Central and Eastern threaten 1157-75. Europereally

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions

188

JournalofEconomic Perspectives

Alexander. 2012."TheNewRole 1997:"German Streeck, Spermann, Wolfgang. Capitalism: ofTemporary Work inGermany." In Die DoesItExist? CanItSurvive?" New Political Agency Economy neue Rolle derZeitarbeit inDeutschland zur 2(2):237-56. , Beiträge editedbyRicharda B. Bouncken, SVR[German Council ofEconomic Flexibilisierung Experts]. - Herausforderungen Lutz Bellmann, and ManfredBornewasser, 2004."Erfolge imAusland Volume München andMering. imInland." Annual to theGovernment, 3,203-224. Report StaBu [German Federal Statis- Wiesbaden. http://www.sachverstaendigenrat tical Office]. 2013. Verdiensteund -wirtschaft.de/frueherejahresgutachten.html. inDeutschland 2010. InstituVisser, Arbeitskosten-Tarifbindung Jelle.2013."WageBargaining Statistisches Wiesbaden, from Crisis toCrisis." Economic Bundesamt, 488, https:// tions Papers Commission. www.destatis.de/DE/Publikationen/Thematisch European April.http://ec.europa /VerdiensteArbeitskosten/Tarifverdienste .eu/economy_finance/publications/economic 3/ecp488_en.htm. /Tarifbindung.html. _paper/201

This content downloaded from 193.54.110.35 on Tue, 16 Feb 2016 18:48:54 UTC All use subject to JSTOR Terms and Conditions