C O N S I D E R I N G T H E F U T U R E O F L A B O R

March 2000

COMPACT

Ten Years after Unification – Lessons for theEast-German Labor MarketA decade after unification, Germany isstill struggling with the integration of theformer eastern command economy intothe social market-led economy. Hopesthat the economic transformation couldbe completed fast and smoothly weresoon disproved as overly optimistic. Oneof the most striking features of the east-ern economy is the persistence of veryhigh levels of unemployment. Since theeconomic catch-up process has sloweddown in recent years, an employmentupswing seems unlikely in the medium-term future. A current IZA study surveysthe emergence and persistence of unem-ployment in East Germany and discusseswhether there are promising policies toreduce what appears to many as an intol-erably high number of displaced workers.For a detailed account see: Holger Bonin/Klaus F. Zimmermann, The Post-UnificationGerman Labor Market, IZA Discussion Pa-per No. 185 (forthcoming in: R. T. Riphahn/D. J. Snower/K. F. Zimmermann, Employ-ment Policy in Transition: The Lessons ofGerman Integration for the Labour Market,Heidelberg 2000).

The transformation of the eastern com-mand economy has by necessity affectedthe level of employment negatively. Theartificial full employment policy of theGDR made state-owned companies andpublic administration overstaffed. Themonetary union at parity forced the east-ern economy into competition with theworld market at a high real exchange rate,leaving many East Germans in economi-cally non-viable jobs. Immediately afterunification, a complete collapse of em-ployment was avoided only with substan-tial financial transfers from the west. Assoon as the eastern economy entered intothe recovery phase of the output j-curve,the contraction of employment sloweddown. However, when Germany as awhole entered into a recession, easternunemployment went up again. The recenteconomic upswing has not relieved theeastern labor market. The rectification ofstructural distortions has rather short-ened employment opportunities further.By the end of 1999, registered unemploy-ment in the east exceeded 19 percent ofthe labor force.

The formidable level of unemploymentwas a traumatic experience for many EastGermans. To avoid social tensions, exten-sive work creation measures were intro-duced. In the first period after unifica-tion, when creation of fresh employmentwas expected to start before long, policiesaimed at stretching out the reduction ofobsolete jobs. The Treuhandanstalt, thecentral privatization agency, often negoti-ated job guarantees with private inves-tors, which ensured at least temporaryjob stability. In the first half of 1991,government authorities subsidized up to2 million workers on short-time work.When most of the short-term workersbecame unemployed soon after, the em-phasis of labor market policy shifted tojob creation programs (ABM) and train-ing schemes. Although the eastern laborforce was seen to exhibit a high level offormal qualification, western-style re-training was expected to boost labormarket opportunities. When the demandfor specific qualification declined, theshare of those unemployed involved intraining programs in the east graduallyconverged to the western level. Publiclyfinanced work programs, by contrast, re-main a popular policy to cut registered

unemployment. In 1999, 18.9 percent ofthe unemployed in the east benefitedfrom ABM, compared to only 2.7 percentin the west.

Empirical studies have not reached aunanimous answer to the questionwhether active labor market policy hasbeen effective in promoting regular em-ployment in the east. In general, school-ing programs seem to have supportedreemployment more successfully than jobcreation programs. One standard expla-nation is that participants search lessintensively for regular employment. In ad-dition, ABM measures, which often sup-port labor-intensive types of work, do notalways meet with the demand on thelabor market. Therefore, participation inABM may give a negative signal to poten-tial employers. The weak impact of publicjob creation programs on individual em-ployment performance casts doubt ontheir role as the central tool of activelabor market policy in the eastern states.This does not say that ABM is not justi-fied for social reasons. In fact, it appearsthat in the eastern states active labormarket programs are used as a socialpolicy instrument to cushion the burdens

September 2000

2 IZA COMPACT · September 2000

of long-term unemployment. In Ger-many, the long-term unemployed typi-cally descend on the transfer income lad-der, from unemployment support to un-employment assistance to social welfare.This process is slowed down in East Ger-many by using active labor market policyto perforate the unemployment period.Since the interruption of unemploymentestablishes new claims on unemploymentbenefits, the perforation strategy post-pones the gradual decline in unemploy-ment transfer levels for the long-term un-employed.

In view of the importance of short-timework and training and employment cre-ation programs in the east, the officiallypublished unemployment figures coverthe actual size of the employment prob-lem only vaguely. In 1999, consideringunemployment hidden by active labormarket policies, actual joblessness in theeast was 40 percent higher than regis-tered unemployment. Besides, commut-ers, out-migrants, and early pensionersdriven out of the labor force have relievedthe eastern working-age labor force by1.3 million persons. Thus, shortage ofemployment in the east has affected morethan 35 percent of the working-age popu-lation.

While the level of unemployment seemsto be stabilizing on a stubbornly highlevel, the structure of employment in theeast continues to differ significantly fromthat in the west. The economically back-ward GDR had maintained the structure

in Germany as a whole over the lastdecade, is rather uncommon on the east-ern labor market. In 1998, for each threepart-time workers in the west there wasonly one in the east. This hints atunexploited opportunities for increasedflexibility of labor.

In order to explain what causes the labormarket crisis in East Germany, many ana-lysts have argued that high wages are amajor source of unemployment and in-sufficient investment. After unification,trade unions successfully pressed forrapid convergence of wages. Given thelow labor productivity after the break-down of the socialist commandeconomy, wage setting certainly did notfollow the logic of neoclassical equilib-rium theory. Nevertheless, the relevanceof wage policy has probably been over-rated. Since the wage drift has been sub-stantial, the focus on union wages exag-gerates the differential between wagecosts and labor productivity. Effectivewages have been approaching westernstandards more slowly than union wagesfor several reasons. First, payment abovethe union rate, frequent in the west, isvery rare practice in the eastern states.Secondly, eastern workers often do notreceive certain extra payments(Urlaubsgeld, Weihnachtsgeld) that arequite standard in West Germany. Finally,standard wages frequently did not prevailon the labor market. In 1998, 79 percentof the companies in the producing sectordid not participate in collective wage bar-gaining. Many of them paid wages belowunion rates. Furthermore, in many firmsthat take part in the collective bargainingprocess formally, internal agreementshave been reached to pay less than stan-dard wages. The adaptation of the wagesetting process to the transformation cri-sis has been preventing wage conver-gence. Effective wages in the east hadreached about three quarters of the west-ern level in 1997 when eastern output perworker was close to 60 percent of that inthe west. The relative wedge between ef-fective wages and labor productivity hasdecreased in the course of the transfor-mation process.

The argument that inadequate wagepolicy is fundamental to the high level ofeastern unemployment can be challengedalso for theoretical reasons. Certainly,wages should follow productivity in anequilibrium state. However, it is less cer-tain that the equilibrium concept wouldprovide a valuable answer to the prob-lems of the East German economy duringtransformation. When the capital stockof the GDR became obsolete, most EastGerman goods would not have sold atany level of socially acceptable realwages. Higher wages, however, should beno problem for a capital stock that isnewly built up. Modern technology canpay high wages. In this perspective, theinitial rapid wage increase in the east set

of an advanced industrial society, so thatthe employment share of agriculture andthe government and producing sectorswas larger, that of trade, transportation,and services smaller than that in the west.After unification, the share of agricultureand the producing sector declined rapidly(the latter below the West German level),while the service sector started to catchup to western standards. Still, conver-gence with western sector patterns is farfrom complete. The eastern governmentsector, which is not exposed to competi-tion, remains overstaffed. What seemsmore worrying with regard to the futureemployment outlook is the fact that boththe trade and transportation sector andthe service sector have not caught up tothe west. Thus there appears to be littleevidence that the eastern states will beable to leave West Germany behind as amodern service economy after comple-tion of the transformation process.

With regard to the occupational positionof the gainfully employed, there is littleevidence as well that employment pat-terns are going to converge soon. Twotrends may be particularly relevant for theeastern employment prospect. First, self-employment, which started from a verylow level when the GDR commandeconomy collapsed, is substantially lessfrequent than in the west. It seems thatthe transformation process did not openup sufficient opportunities for private en-terprise that could promote dependentemployment. Secondly, part-timeemployment, which has been increasing

IZA COMPACT · September 2000 3

incentives for creative destruction. A lowwage policy in East Germany would haveresulted in a low productivity capitalstock associated with very slow conver-gence to the west, which was not a politi-cally feasible strategy.

Shortage of employment is a relative phe-nomenon. The high unemployment ratesreported for East Germany also reflectspecific patterns of labor market partici-pation and working hours. Labor forceparticipation in the former GDR, wherepolitics set strong incentives to work, wastraditionally higher than in the West Ger-man states. In contrast to what manyobservers of the transformation processexpected, marked differences in laborsupply have been persistent. In 1998,labor force presence of East Germans wassome 8 percent higher than that ofwesterners. In particular, the propensityof East German women to take up a jobhas not converged to western standards,despite rising obstacles to find employ-ment. In view of the comparatively loweastern wage levels, one reason for thisbehavior appears to be the necessity tomaintain family income. This may explainwhy labor force participation of marriedfemales is particularly high.

With strong disparities in labor supplypersisting, a reduction of eastern unem-ployment to the western level cannot beexpected. Furthermore, the focus on re-gional disparities of unemployment ratesprobably exaggerates the severity of theemployment crisis. To assess the actualsize of the employment problem, it is alsoimportant to what extent potential laboris being used. In 1998, the easternemployment rate was 56.4 percent of theworking-age population, some 5 percent-age points below the rate for West Ger-many. In this perspective, although theeffective employment levels in the easthave stabilized below the western stan-

dard, shortage of employment opportu-nities is clearly less substantial. This iseven more evident considering the effec-tive volume of eastern employment. Ifmeasured in terms of yearly workinghours per gainfully employed, labor vol-

ume per worker in the east has consis-tently surpassed that in the west and hasbeen converging only slowly to westernstandards. In 1998, eastern workers wereoccupied 6 percent longer than westernworkers. The regional differences of indi-vidual employment volume are only par-tially due to the less generous wage agree-ments. They also reflect the considerableshortage of part-time occupations on theEast German labor market. Had workinghours in the east been distributed accord-ing to western standards, the size of gain-ful employment in 1998 could have beensome 390,000 persons higher. Then theeffective employment quota would havebeen as high as 60 percent, compared to61.7 percent in the west. Given the vol-ume of working hours relative to the sizeof the population at employable age,there also appears to be no substantialshortage of employment opportunities inthe east. In fact, the effective easternemployment volume even exceeded thatin the west during parts of the transfor-mation process. In 1998, it was less thanthree percentage points below the west-ern rate. To avoid misunderstandings, itis important to note that this type ofcalculation does not imply that the largesize of unemployment in East Germany isnot a real social problem. Nevertheless, itmust be emphasized that a remarkableadjustment of the overall employmentvolume has already taken place in theeast. Given that wages and labor produc-tivity are converging to western stan-dards, it is unlikely that the overall em-ployment volume in the east will stabilizeabove the western level and absorb themore ample eastern labor supply.

Ten years after unification, there is littledoubt that the big bang approach of fasteconomic integration was without alter-native. Still, it seems that some opportu-nities were missed. A major political mis-take was to underestimate the economiccosts associated with the transition. Dueto the political myth of smooth economicintegration, the challenge of transformingthe eastern socialist economy was notperceived as an opportunity for re-think-ing the political and economic system inGermany as a whole. A frequent attitudewas that the reunited Germany shouldbecome an enlarged Federal Republic.However, many of the institutions ex-panded to the eastern states had beenimpeding economic progress also in thewest. Not surprisingly, the systemic prob-lems aggravated when the transition crisiscalled for greater flexibility. Lack of insti-tutional innovation during the transfor-mation process imposed serious ob-stacles to economic recovery in the east.Public administration, for example, waslargely rebuilt after western examples.Hence the opportunity was missed toinstall flexible techniques of governmen-tal management that could have pro-vided the necessary flexibility for copingwith the problems during the transforma-tion process. Ineffective governmentslowed down the reconstruction of theoutdated eastern infrastructure. Regionaldisparities in public infrastructure remainsubstantial. The difficult conditions oflocal administration and the lack of pub-lic infrastructure may explain why thegenerous investment incentives that wereintroduced to promote capital formationwere only a partial success.

A second example of inadequate conver-gence is the initially unqualified adapta-tion to pre-existing labor market institu-tions. Universal collective bargaining ex-erted decisive influence on the rapid in-crease of eastern labor costs during thefirst period after unification. In addition,the acceptance of the highly regulatedworking conditions in the west was in-compatible with the need for more flex-ible labor organization acute on the east-ern labor market in transition. The massunemployment in East Germany detectedthe inflexibilities of traditional Germanlabor market institutions all too cruelly.Fortunately, there are some signs that thedeep transformation crisis in East Ger-many also fosters institutional renewal.The progressive loss of membership inunions and employers’ associations andthe establishment of bargaining practicesat plant level ignoring industry guidelinesdemonstrate that the eastern society iscapable to independent initiative.

A historically unique experiment is aboutto enter its second decade – Germanunification. Early hopes for a rapid andsmooth economic transformation soonturned out to be overly optimistic. De-spite massive financial transfers, the po-litical promise of a “blooming land-scape” remains a vision. Actual develop-ments have left deep scars on the labormarket, and the effects will be felt fordecades to come.

Was this outcome to be expected, per-haps even inevitable? What went wrong,and what were the available options? Oris the current state of the eastern Germanlabor market in fact better than is com-monly assumed? To answer these ques-tions, the internationally renowned au-thors of this volume examine differentstages of the labor market developmentin eastern Germany. The relevant factsare certainly diverse: While the rate ofunemployment in the “new Laender” issubstantially higher than in the west, theactual amount of work done per labor

force participant is about the same inboth parts of the country. Moreover, realwage convergence has been much slowerthan originally predicted.

After an introduction to the subject andan overview of labor market develop-ments in eastern Germany, part 1 of thisvolume will turn to specific issues ofadjustment: wage trends, business start-ups, and early retirement. Part 2 reviewsthe policy options available in the trans-formation process and their labor marketimplications: wage subsidies, profit shar-ing, invested pay, further education, andvarious forms of active labor marketpolicy. The final chapter draws parallelswith the experience of other transitioneconomies.

This book will be released in October, 2000.

R. T. Riphahn/D. J. Snower/K. F. Zimmermann,Employment Policy in Transition: The Lessonsof German Integration for the Labour Market,Heidelberg 2000; ISBN 3-540-41166-6.

New book release: Lessons learned from ten years of market economyin former East Germany - Renowned economists analyze the eastern

Introduction

1. IntroductionRegina T. Riphahn (University of Munich, IZA/Bonn, andCEPR/London)Dennis J. Snower (Birkbeck College/London, IZA/Bonn,and CEPR/London)Klaus F. Zimmermann (Bonn University, IZA/Bonn, DIW/Berlin, and CEPR/London)

2. The Post-Unification German Labour MarketHolger Bonin (IZA/Bonn)Klaus F. Zimmermann (Bonn University, IZA/Bonn,DIW/Berlin, and CEPR/London)

Part I: Analysis of the German Labour MarketProblem

1. Wages and Structural Adjustment in the New GermanStatesMichael Burda (Humboldt University Berlin, IZA/Bonn,and CEPR/London)Michael Funke (University of Hamburg)

2. East/West-Wage Rigidity in United GermanyKarl-Heinz Paqué (University of Magdeburg)

3. Early Retirement in East and West GermanyAxel H. Börsch-Supan (University of Mannheim and NBER/Cambridge)Peter Schmidt (University of Bremen)

4. Employment Effects of Newly Founded Businesses inEast GermanyThomas Hinz (University of Munich)Rolf Ziegler (University of Munich)

5. Earning Dynamics in the East German Transition ProcessJohannes Schwarze (University of Bamberg and IZA/Bonn)Gert G. Wagner (University of Frankfurt/Oder, DIW/Berlin,and IZA/Bonn)

Part II: Policy Options

1. Eastern Germany Since Unification: Wage SubsidiesRemain a Better WayDavid Begg (Birkbeck College/London and CEPR/London)Richard Portes (London Business School and CEPR/London)

2. Economic Efficiency and Social Acceptance of WageSubsidiesHenning Klodt (Kiel Institute of World Economics)

3. Revenue-Sharing Subsidies as Employment Policy:Reducing the Cost of Stimulating East GermanEmploymentDennis J. Snower (Birkbeck College/London, IZA/Bonn,and CEPR/London)

4. Investment Wages and Capital Market ImperfectionsGerhard Illing (University of Frankfurt/Main)

5. Public Sector Sponsored Continuous Vocational Trainingin East Germany: Institutional Arrangements,Participants, and Results of Empirical EvaluationsMartin Eichler (University of Mannheim)Michael Lechner (University of St. Gallen and IZA/Bonn)

6. Active Labour Market Policies in Central Europe: FirstLessonsHartmut Lehmann (University of Dublin, Trinity College,and IZA)

Employment Policy in Transition:The Lessons of German Integration for the Labour Market

edited by R. T. Riphahn, D. J. Snower, and K. F. Zimmermann

4 IZA COMPACT · September 2000

Shaping Working Hours – The Case of the Netherlands

IZA COMPACT · September 2000 5

Last year the Dutch parliament acceptedlegislation on the right of employees toadjust their contractual working hours up-ward or downward under certain condi-tions. By July 1, 2000, this legislation en-tered into force. Its goal is to increase thepossibilities to combine work and privatelife for both men and women, particularlyto stimulate the economic independence ofwomen. The legislation is part of a muchmore comprehensive legislation on workand family care that is still in development.

Only in the last two decades has the Dutchgovernment been quite active in shaping theconditions of combining work and care.Before the 1980s the need was not preva-lent. At that time, female labor force par-ticipation was low compared to othercountries. But since the early 1980s, femalelabor participation has been increasingstrongly and has almost caught up with thesituation in Germany. Accordingly, the needfor shaping the conditions of combiningwork and care increased. The solution ofthe Dutch government to handle this prob-lem was to stimulate part-time work. In thebeginning of the 1990s the Dutch parlia-ment passed several acts that forbid em-ployers to treat employees unequally withregard to the conditions under which anemployee contract is initiated, prolonged orended, solely based on a difference in thenumber of working hours. Thus the actawards part-time employees an explicitright to equal treatment concerning wages,paid leave and holiday entitlements, over-time payment, bonuses and training. It isunclear to what extent this really contrib-uted to the success of part-time work in theNetherlands, but one result is that the pro-portion of both men and women workingpart-time is high in the Netherlands.

Although the new law gives a lot of rights toemployees, there are some clauses to pro-tect the employers. First of all, the law doesnot hold for firms with less then 10 employ-ees. Secondly, the legislation gives firms thepossibility to reject the wishes of employeesif they can show that otherwise they wouldface considerable problems. Examples forthat are a lack work or the reverse, a lack ofpersonnel. The employees do not need togive arguments for their wishes, but thefirms have to come up with arguments incase they do not want to fulfill the employ-ees’ wishes.

It is an open question why such a rigorouslegislation would be needed in the Nether-lands. Part-time employment is quite com-mon, and the Dutch labor market seemsflexible in terms of working hours. Obvi-ously, even in such a labor market theactual number of working hours does notseem to fit the preferences, as previous IZAresearch has shown (see Rob Euwals/Arthurvan Soest, Desired and Actual Labor Supply ofUnmarried Men and Women in the Nether-lands, Labour Economics 6 (1999), pp. 95-118

and Rob Euwals, Female Labor Supply, Flexibil-ity of Working Hours, and Job Mobility in theNetherlands, IZA Discussion Paper No. 83).Studying labor supply and employment be-havior of unmarried men and women in the

Netherlands, it has become evident thateven in the Dutch labor market it is substan-tially more difficult to find a part-time jobthan a fulltime job. For an individual with anaverage education living in the west of thecountry (where the largest cities are), theprobability to get a job-offer for a fulltimejob is about 90 percent, while the probabil-

ity to get a job-offer for a part-time job isabout 45 percent.

The restrictions work in both directions:many individuals who work fulltime want towork part-time but also some would preferto work fulltime instead of part-time. Fur-thermore, studying labor supply and em-ployment behavior of Dutch women, it can

be shown that within a job (with the sameemployer) working hours on average adjustabout 15 percent into the preferred direc-tion. Thus, if a woman changes her prefer-ences on working hours because of chang-

ing family circumstances (if, for instance,children leave the household, parents be-come sick, or her partner’s incomechanges), it is actually difficult to changeher working hours within the job accord-ingly. In other words: even in a labor marketwhere part-time employment is quite com-mon it is not easy to match the actual

working hours with the preferred workinghours and to combine work and care. It isdoubtful whether the restrictive legislationof the Dutch parliament will be able toinfluence the situation. IZA continues tofollow developments in the Netherlandsand to analyze implications for workingtime policy in other European countries.

6 IZA COMPACT · September 2000

IZA hosts meeting of top experts

IZA organized and hosted this year’s an-nual conference of the European Societyfor Population Economics (ESPE). Themeeting, which took place at theWissenschaftszentrum (Research Center)in Bonn, June 15-17, 2000, was an op-portunity for more than 200 populationeconomists to discuss questions of greatrelevance to the present situation.

Mothers, Fathers and Children afterDivorce

In her presidential address, ESPE Presi-dent Daniela Del Boca (University ofTurin, Italy) lectured on the impact ofdivorces on the children’s welfare and theeconomic consequences for both par-ents. Until recently this issue has beenwidely neglected. But since the number ofdivorces, especially in the United Statesand Britain, has risen significantly andcaused substantial welfare effects, thetopic has finally received increased atten-tion.

Del Boca pointed out the serious conse-quences of a divorce for the future devel-opment of the affected children. Theseconsequences extend beyond the socialimplications and also affect human capi-tal accumulation and labor market pros-

pects. Close cooperation between bothparents, Del Boca argued, is therefore anessential – albeit only moderately realistic– precondition for the children’s welfareand job prospects.

Del Boca further elaborated on the ques-tion why a divorce often has an asymmet-ric effect on the parents. This is due to thefact that fathers usually earn a higher

ESPE annual conference gives new impetus to research inpopulation and labor economics

income, while in most cases mothers aregranted custody of the child, which limitstheir employment possibilities. After adivorce, families no longer benefit fromthe economies of scale inherent in sharedhouseholds. Therefore, it is often vitalthat the financially stronger partner re-main employed in order to support thedivorced spouse and the children in ac-cordance with the law. With increasedfemale participation in the workforceand better representation of women inhigh-salary jobs, however, this situationmay change. It is likely that in the futuremore and more fathers will receive cus-tody of their children.

Notwithstanding this development, thecentral goal from an economic point ofview should be a fair balance of interests.Benefits from regular child visitation, forinstance, are not just confined to theinterpersonal aspect. Visitation has posi-tive effects on the child and possibly onthe payment morale of the supportingparent, which would ultimately contrib-ute to the child’s welfare.

Evaluation of Official Labor MarketPrograms

Gerard J. van den Berg (Free University ofAmsterdam, Netherlands) gave a presen-tation on the scientific evaluation of theeffectiveness and efficiency of official la-bor market programs. The purpose ofthis assessment is to measure the ben-efits, for instance, from participation infurther training, and to weigh these ben-efits against the economic costs incurredin the process. The question of potentialnegative incentives in the form of highertaxes that are necessary to finance suchprograms is not answered in this context.

The fundamental problem of this type ofevaluation is the scarcity of informationabout participants of government pro-grams. An unfavorable evaluation couldresult from the fact that participantsshowed little motivation because unem-ployment benefits were conditional onparticipation in the program. The oppo-site case – participants with exceptionallystrong motivation – would also be con-ceivable and could distort the evaluation.

It is therefore necessary, van den Bergexplained, to give close attention to theamount of time each participant spendsin the program. Highly motivated partici-pants can generally be expected to finishthe program earlier than those who par-ticipate involuntarily. Van den Berg’s re-search of labor market programs in theNetherlands showed that they had astrongly positive effect. He pointed out,however, that further research is neededto weigh these benefits against the costsborne by the economy.

The Timing of Work and Leisure

Daniel S. Hamermesh (University of Texasat Austin, USA) analyzed the timing ofwork and leisure among family house-holds and unmarried couples. Contraryto the predominant approach, his analy-sis concentrates on the leisure side. Basedon U.S. data, Hamermesh showed thatpartners indeed manage to coordinatetheir labor market activities. It is neces-sary, however, to differentiate betweencouples with and without children, be-cause parents of young children also haveto coordinate their working hours withthe schedules for child care and baby-sitting. In this case, mothers usually carrythe burden of adjustment.

Da

nie

la D

el B

oca

Da

nie

l S.

Ha

mer

mes

h

Ger

ard

J. v

an

den

Ber

g

IZA COMPACT ·September 2000 7

According to Hamermesh, the changes inworking hours during the past decadeshave substantially improved the possibili-ties of timing work and leisure. As re-quested by employees, working hours arenow increasingly concentrated, creatingnew opportunities for leisure activities.With better education and higher salary,selection between work and leisure gener-ally becomes easier. This leads to theconclusion that preferred working hoursapparently have to be bought with for-gone wage payments. Consequently, thewelfare gap between highly trained andless trained workers must be wider than isapparent from the existing wage differen-tials alone.

European Integration and the Wel-fare State

Torben M. Andersen (University of Aarhus,Denmark) talked about the conse-quences of European integration on thewelfare state in current and future EUmember countries. For quantitative rea-sons alone, this is a highly relevant topicbecause European welfare budgets haveplayed an ever more important role sincethe 1950s.

Financing the welfare state while integrat-ing the economies poses a great challengeto the members of the European Union.Inasmuch as international integration en-courages spatial mobility, the tax basewill become more mobile and migrate tothe countries with the best tax condi-tions. This could lead to an inefficientcompetition in tax policy. As a conse-quence, immobile factors, which alreadycarry most of the burden of financing thewelfare state, would have to bear an evenheavier weight.

But, according to Andersen, even if laborremains relatively immobile and contin-ues to serve as the main source for thefunding of welfare payments, the com-modity markets will become increasinglycompetitive as a result of integration. On

the one hand, this would weaken theposition of labor unions because withdiminishing monopolist’s profits the op-portunities of distribution will shrink. Onthe other hand, higher wages would in-duce firms to move their operationsabroad. Unless wages and social securitycontributions are adjusted in each coun-try, higher unemployment would be theinevitable consequence.

Future social security requirements,Andersen explained, would depend oncollective risks for which insurance re-mains necessary. The impact of integra-tion on insurable individual risks wouldtherefore play a major role. Andersenassumes that a larger economic area isbetter equipped to cover risks than asmall area. Consequently, the futureneed for collective social security insur-ance would in all likelihood be lowerthan today.

Pension Reform and IntergenerationalRedistribution

Another highly relevant issue discussed atESPE 2000 is the problem of pensionfunding. The experts stressed the fact

that the German system ofintergenerational redistribution dependslargely on the size of the workforce andthus on the number of contributors.Since families play a major role in sustain-ing the current pension system, it seemsreasonable to compensate them for partof the costs, possibly in the form ofpension entitlements. Participants of theconference also underscored the rel-evance of redistributional effects. Theyounger generations should not – as thewidely discussed models in the Germandebate about pension reform suggest –carry a disproportionately high costs.Today’s observations already show thatolder generations can still increase theirper capita consumption, while youngerand future generations will face substan-tially reduced consumption opportuni-ties. Prudent reforms are necessary tocounteract this economically unhealthytrend.

Outlook

In summary, ESPE 2000 was conductedon an exceptionally high academic level.It demonstrated once again the greatimportance of population economics infinding answers to compelling questionsof demography, social security, and labormarket policy. IZA Director Klaus F.Zimmermann remarked at the end of theconference: “Population economics ismore than ever at the center of publicattention. Key issues that will shape ourfuture, such as the aging of western soci-eties, coping with immigration, reform ofthe welfare states, and creating the rightincentives for more employment are call-ing for convincing political concepts.Policymakers should embrace the an-swers that population economists offer:We need a systematic control of immigra-tion, incentives for individual initiative,and a social safety net that concentrateson need rather than redistribution. ESPE2000 has given science and politics animportant impetus.“

The following contribution is to a largeextent based on a publication by LilianaWinkelmann and IZA program directorRainer Winkelmann, “Why are the unem-ployed so unhappy? Evidence from paneldata”, published in Economica 65 (1998),pp. 1-15. This publication received in 2000the “Friends of the DIW Prize for the BestPaper 1984-1998/99 using the GermanSocio-Economic Panel (Second Prize)”.

The achievement of full employment isone of the undisputed goals of economicpolicy, because unemployment always

Measuring the Cost of Unemployment

implies the reduction of a country’s well-being. Attempts to quantify the cost ofunemployment for the economy as awhole remain controversial, however. Thetraditional economist’s view holds thatthe main cost of unemployment is theloss of production, which is possibly cor-rected by an estimate of the value theunemployed receive from time spent onalternative non-market activities. The em-pirical cornerstone of this approach isOkun’s law, according to which a 3 per-cent increase in GDP is associated with a1 percent reduction of the unemployment

rate. In principle, the shortfall in GDPassociated with a given unemploymentrate can thus be computed.

This approach has come under criticismfrom various sides. First, the stability andvalidity of the empirical law has beenquestioned in subsequent research. Re-sults differ depending on country andtime period under consideration, but thegeneral impression is that Okun’s lawsubstantially overestimates the size of thetrade-off between unemployment andoutput. The issue is complicated further

To

rben

M. A

nd

erse

n

8 IZA COMPACT · September 2000

by the fact that, strictly speaking, Okun’slaw applies only to cyclical unemploy-ment, while a large part of current unem-ployment in Western Europe and else-where is considered to be structural.

More importantly, lost production – evenif measured properly – is too narrow ameasure of the cost of unemployment tobe meaningful and applicable. The un-employed are not just an “idle factor ofproduction”. They are deprived of animportant source of social relationships,identity and individual self-esteem. If em-ployment was just a source of income,the social safety net with its – in the caseof Germany - relatively high replacementrates would be more successful in miti-gating the adverse effect of unemploy-ment on the individual. However, there iswidespread evidence that this is not thecase in general. Individuals are negativelyaffected by unemployment in ways thatgo well beyond the pure pecuniary effect.

This finding is not entirely new, of course.Psychologists have developed modelsand provided explanations for the harm-ful effect of unemployment on the psy-chological well-being of the individual.For instance, they point to a lack ofcontrol, security, and sense of belongingthat is observed among the unemployed.What is new is that there is now a sub-stantial body of evidence based on large-scale survey data that provides a firmempirical basis for the previous conjec-tures and produces some startling re-sults.

Why are the unemployed so un-happy?

The most direct evidence on the non-pecuniary cost of unemployment is ob-tained by asking people how satisfiedthey are with their present life in general.Then the responses of employed and un-employed persons can be compared. TheGerman Socio-Economic Panel (SOEP),a household survey conducted annuallyby the German Institute for EconomicResearch (DIW), regularly includes aquestion on general satisfaction with lifeon a scale from 0 (= completely dissatis-fied) to 10 (= completely satisfied). In1984, the first year of the survey, theaverage response of employed persons tothis question was 7.6 on the scale from 0to 10, whereas the average response ofunemployed persons was 5.6. This is aconsiderable gap in general satisfactionwith life. Similarly, only 5 percent of theemployed were dissatisfied whereas 27percent of the unemployed were dissatis-fied. In other words, unemployed per-sons were more than five times as likely toreport general dissatisfaction with theirlife than employed persons. Qualitativelysimilar patterns are found throughoutthe 1980’s and the 1990’s. There is sys-tematic evidence that the unemployedare more dissatisfied with their lives com-pared to people who have a job.

In a different type of analysis, the longitu-dinal aspect of the SOEP is used to followa person who was employed in one yearbut unemployed in the next. Then thechange in the response to the life satis-faction question between the two years iscomputed for each person. On average,satisfaction goes down by one unit (onthe 0-10 scale) in this case. Conversely,satisfaction goes up by one unit forpeople who become reemployed. More-over, the strong negative association be-tween unemployment and life satisfac-tion remains intact also when a wholearray of other potential influences is ac-counted for. This includes a person’shealth, marital status, and income. Thedata from the German Socio-EconomicPanel shows that particularly for menemployment is the most important con-tributor to general life satisfaction. It is alot more important than income, whichhas a small positive effect on satisfaction.

The transition from employment to un-employment has two aspects. First, theperson loses income depending on theapplicable replacement ratio, and hencebecomes less satisfied. Secondly, the per-son becomes less satisfied for non-pecu-niary reasons. These two factors com-bined constitute the total cost. Based onestimates from the German-Socio Eco-nomic Panel for men between 1984 and1989, some rough calculations showthat only 10 percent of the total costs ofunemployment are pecuniary, whereas90 percent are non-pecuniary. While thisestimate may be at the upper end of thespectrum, other data sources show quali-tatively similar results, among them theBritish Household Panel Survey and the

International Social Survey Program dataset, which covers twelve different coun-tries and other time periods.

Externalities

So far only the effect of individual unem-ployment on the life satisfaction of theunemployed person was considered. Tocalculate the total social cost, however, itmust be taken into account that indi-vidual unemployment can affect otherpeople’s life satisfaction as well, which isdescribed by the term “externality”. Forexample, data from the SOEP has beenused to show that the life satisfaction of awoman drops significantly when her hus-band is unemployed. This is a negativeexternality. But positive externalities canexist as well: part of the psychologicalcost of unemployment arise since it ex-cludes people from society and deprivesthem of contact possibilities. This effectwill be weaker the more people in aperson’s social reference group are unem-ployed. Indeed, there is strong empiricalevidence that high unemployment in thereference group, for example a neighbor-hood or a profession, reduces the adversepsychological effect of unemployment atthe individual level. In contrast to thelarge negative “internal” effect of unem-ployment, it is still uncertain whether thepositive or the negative effects dominateon the external side.

Implications for welfare policy

The existence of negative effects of unem-ployment on life satisfaction has a num-ber of implications. It shows that unem-ployment is hardly “voluntary”. This isnot surprising to the layperson, but it is a

IZA COMPACT ·September 2000 9

contentious issue within the economicsdiscipline. Indeed, the finding shows thaton average the unemployed put a remark-ably low value on their leisure. For in-stance, when offered the choice betweenworking and being unemployed, a personwould presumably chose to work in orderto increase her life satisfaction, even inthe hypothetical case when income is leftunchanged. This is due to the pure non-pecuniary or psychological benefits ofemployment. Based on this evidence, theoften-heard argument that unemploy-ment compensation ought to be reducedin order to increase work incentives andthereby employment is inappropriate.The non-pecuniary benefits of employ-ment already provide a massive incentivefor taking up work, and the fact thatpeople are still unemployed shows thatunemployment results from an insuffi-cient demand for certain types of work-ers, rather than being supply-driven.

Reduced life satisfaction and persis-tent unemployment

Research on the connection between un-employment and life satisfaction hasbeen refined in a number of ways. Amongthe hypotheses tested are those of scar-ring and habituation. For instance, it has

SHORT REPORTS

been shown that past unemployment hasa permanent negative effect on psycho-logical well-being. At the same time, itseems that people get used to being un-employed: The negative effect of unem-ployment on satisfaction is weaker forthose who have been unemployed moreoften in the past. Furthermore, these sub-jective well-being measures can be usedto explain economic behavior. For ex-ample, the habituation effect could meanthat the long-term unemployed “dislike”their situation less and less, give up, andabandon search. Empirical researchshows that those who experience greaterreductions in well-being at the beginningof their being unemployed are more likelyto become reemployed soon. An alterna-tive view of the unemployment problem isbased on the notion of psychologicalcapital. Psychological capital, as well ashuman capital, is reduced during unem-ployment episodes, making reemploy-ment less likely and causing self-perpetu-ation and hysteresis.

Conclusion

The use of life satisfaction data for study-ing the non-pecuniary cost of unemploy-ment has generally been a successful re-search program that has produced a

number of interesting and robust find-ings. Still, some people may not feel atease with the subjective nature of lifesatisfaction data and with the whole con-cept of measuring utility and preferencesdirectly. Economists in particular tend todefend their position vigorously: Theystudy what people do and infer prefer-ences indirectly, rather than studyingwhat people think – or say they think –their preferences are. However, parallelresearch has studied the association be-tween unemployment and a number ofdirectly observable “objective” outcomemeasures such as mental illness, divorce,crime, suicide and mortality. These out-comes are all indicators of various as-pects of the social or non-pecuniary costof unemployment. The general evidencecorroborates the results of the life satis-faction approach: Unemployment hasnegative effects on observed outcomes,and it is safe to conclude that the socialcosts of unemployment are far greaterthan the cost of lost production.

+++ Third IZA European SummerSchool a Grand Success+++

The third annual IZA European SummerSchool in Labor Economics was heldfrom May 29 to June 4, 2000. In therelaxed atmosphere at the conferencecenter of the Deutsche Post World Net(German Postal Service) in Buch at the

lake of Ammersee, Bavaria, Germany, thesummer school once again brought to-gether a group of highly talented Ph.D.students with distinguished scholars tostudy new areas in labor economics. Theparticipants, selected by a committeeout of 130 applicants, came from a widerange of European countries, creating avery productive international environ-

ment. During the course of the week, thestudents were given the opportunity topresent their research ideas eitherthrough giving a formal presentation orthrough explaining their work in postersessions. These presentations framed thesetting for many open and lively discus-sions, leaving students with numeroushelpful comments from both their col-leagues and the established researchers.In addition, the students had the oppor-tunity to attend lectures given by DanielS. Hamermesh (University of Texas atAustin) on “Labor Demand” and GerardJ. van den Berg (Free University ofAmsterdam) on “Duration Models andSearch Models in Labor Economics.” Thefourth IZA Summer School in Labor Eco-nomics will take place from September 10to 16, 2001. Details on the applicationprocedure and the program of next year´sSummer School will soon be available onthe IZA homepage.

+++ IZA Workshop “Welfare Transi-tions”, October 7-8, 2000 +++

IZA is organizing an international work-shop on welfare transitions. The aim ofthe workshop is to learn more about the

Conference Center of the Deutsche Post World Net, Buch am Ammersee

10 IZA COMPACT · September 2000

IZA VISITORS PROGRAM

From March until July 2000 a number of renowned economists stayed at IZA to discuss issues of labor market research and policy:

determinants of the flows in and out-ofwelfare, in particular from a public policyperspective. The impact of labor marketconditions and the generosity of socialassistance on the transitions from welfareto work will be discussed. Special atten-tion will be paid to immigrants and mi-norities. The workshop will bring togetheracademics from a variety of countries tolearn more about the different institu-tional settings of the countries consid-ered. This will also allow assessments ofthe effects of welfare reforms. More de-tails on this forthcoming event are avail-able on the IZA homepage (www.iza.org).

+++ IZA Program Director RainerWinkelmann Co-Winner of SOEPAward 2000 +++

Since 1984, data from the German Socio-economic Panel (SOEP) has appeared inover 2,000 academic publications. At thefourth international SOEP user confer-ence, which took place on June 7, 2000,at the Social Science Research Center inBerlin, the best publications on the basis

of SOEP data for the first time received anaward. Among the prize-winning workswas the essay entitled “Why are the un-employed so unhappy? Evidence frompanel data” by Liliana Winkelmann andIZA program director Rainer Winkel-mann. This article, published in Eco-nomica 65 (1998), received the “Friendsof the DIW Prize for the Best Paper 1984-1998/99 using the German Socio-Eco-nomic Panel (Second Prize)”. For a sum-mary of the key research results, see pages7-9 of this issue of IZA compact.

+++ IZA makes JSTOR available +++

Since last April, IZA researchers have beenable to access the electronic databaseJSTOR. This makes IZA the fourth Ger-man institution to grant its employeesaccess to one of the most efficient Ameri-can journal databases. The three otherinstitutions are the Bavarian State Li-brary, Munich, Georg August University,Göttingen, and Johann Wolfgang GoetheUniversity, Franfurt/Main. JSTOR cur-rently offers access to full-text articles

from more than 100 academic journals insuch disciplines as economics, demogra-phy, political science, mathematics, andstatistics. With JSTOR, IZA researchersbenefit from a modern, efficient researchinstrument, which facilitates the retrievalof academic information and substan-tially accelerates progress in research.

+++ IZA moves into second building inBonn +++

IZA is expanding. The growing number ofemployees reflects this development. InJune 2000, with support of the DeutschePost Foundation, IZA was able to moveinto a second office building in the imme-diate vicinity of the IZA “headquarters”.The institute now enjoys a perfect workenvironment in the heart of Bonn, a prof-itable research location. Since severalguests can be accommodated in ad-equate offices at the same time, IZA visit-ing scholars certainly benefit from thenew situation as well.

Lennart Flood(Gothenberg University, Sweden)Christian Dustmann(University College London, UK)Christoph M. Schmidt(University of Heidelberg, Germany)John Haisken-DeNew(DIW, Berlin, Germany)Michael Shields(University of Leicester, UK)Robert Wright(University of Stirling, Scotland))Scott M. Fuess(University of Nebraska, USA)Adriaan Kalwij(University of Oxford, UK)Kai Konrad(Free University of Berlin, Germany)Horst Entorf(University of Würzburg, Germany)Ekkehart Schlicht(University of Munich, Germany)Dennis Snower(Birkbeck College, London, UK )Maria Karanassou(Birkbeck College, London, UK)

Hessel Oosterbeek(University of Amsterdam, Niederlande)Rebecca Blank(University of Michigan, USA)Christian Belzil(Concordia University, Montreal, Canada)Peder Pedersen(University of Aarhus, Denmark)Andrew Newell(University of Sussex, UK)Uwe Blien(IAB Nürnberg, Germany)T. Paul Schultz(Yale University, USA)Marc Regets(National Science Foundation, USA)Alessandro Cigno(University of Florence, Italy)Etienne Wasmer(Université Libre de Bruxelles, Belgium)Yves Zenou(University of Southampton, UK)Wayne A. Cornelius(University of California, San Diego, USA)Junsen Zhang(Chinese University of Hong Kong)

Dan Hamermesh(University of Texas at Austin, USA)Barry R. Chiswick(University of Illinois, Chicago, USA)David Jaeger(Princeton University, USA)Anders Klevmarken(Uppsala University, Sweden)Jörn-Steffen Pischke(MIT, USA)Erik Plug(University of Amsterdam, Niederlande)Joop Hartog(University of Amsterdam, Niederlande)Wim Vijverberg(University of Texas at Dallas, USA)Stefan Bender(IAB, Nürnberg, Germany)Ira N. Gang(Rutgers University, New Brunswick, USA)Olaf Hübler(University of Hannover, Germany)Gil Epstein(Bar-Ilan University, Ramat-Gan, Israel)

IZA COMPACT ·September 2000 11

NEW IZA DISCUSSION PAPERS

A complete list of IZA Discussion Papers is available on our homepage at www.iza.orgMost papers are downloadable.

97 A. BarrettIrish Migration: Characteristics, Causes andCon-sequences98 J.P. Haisken-DeNew,

C. M. SchmidtIndustry Wage Differentials Revisited: A Longi-tudinal Comparison of Germany and USA99 R. T. RiphahnResidential Location and Youth Unemployment:The Economic Geography of School-to-Work-Transitions100 J. Hansen, M. LofstromImmigrant Assimilation and Welfare Partici-pation: Do Immigrants Assimilate Into or Out-ofWelfare? 101 L. Husted, H. S. Nielsen,

M. Rosholm, N. SmithEmployment and Wage Assimilation of MaleFirst Generation Immigrants in Denmark102 B. van der Klaauw, J. C. van OursLabor Supply and Matching Rates for WelfareRecipients: An Analysis Using NeighborhoodCharacteristics103 K. BrännäsEstimation in a Duration Model for EvaluatingEducational Programs104 S. KohnsDifferent Skill Levels and Firing Costs in aMatching Model with Uncertainty – An Exten-sion of Mortensen and Pissarides105 G. Brunello, C. Graziano, B. ParigiOwnership or Performance: What DeterminesBoard of Directors’ Turnover in Italy?106 L. Bellmann, S. Bender,

U. HornsteinerJob Tenure of Two Cohorts of Young GermanMen 1979 – 1990: An analysis of the(West)German Employment Statistic RegisterSample concerning multivariate failure timesand unobserved heterogeneity107 J. C. van Ours, G. RidderFast Track or Failure: A Study of the CompletionRates of Graduate Students in Economics108 J. Boone, J. C. van OursModeling Financial Incentives to Get Unem-ployed Back to Work109 G. J. van den Berg,

B. van der KlaauwCombining Micro and Macro UnemploymentDuration Data110 D. DeVoretz, C. WernerA Theory of Social Forces and Immigrant SecondLanguage Acquisition111 V. Sorm, K. TerrellSectoral Restructuring and Labor Mobility: AComparative Look at the Czech Republic112 L. Bellmann, T. SchankInnovations, Wages and Demand for Hetero-geneous Labour: New Evidence from a MatchedEmployer-Employee Data-Set113 R. EuwalsDo Mandatory Pensions Decrease HouseholdSavings? Evidence for the Netherlands114 G. Brunello, A. MedioAn Explanation of International Differences inEducation and Workplace Training115 A. Cigno, F. C. RosatiWhy do Indian Children Work, and is it Bad forThem?116 C. BelzilUnemployment Insurance and Subsequent JobDuration: Job Matching vs. Unobserved Hetero-geneity

117 S. Bender, A. Haas, C. KloseIAB Employment Subsample 1975-1995.Opportunities for Analysis Provided by theAnonymised Subsample118 M. A. Shields, M. E. WardImproving Nurse Retention in the BritishNational Health Service: The Impact of JobSatisfaction on Intentions to Quit119 A. Lindbeck, D. J. SnowerThe Division of Labor and the Market forOrganizations120 P. T. Pereira, P. S. MartinsDoes Education Reduce Wage Inequality?Quantile Regressions Evidence from FifteenEuropean Countries121 J. C. van OursDo Active Labor Market Policies HelpUnemployed Workers to Find and Keep RegularJobs?122 D. Munich, J. Svejnar, K. TerrellReturns to Human Capital under the Commu-nist Wage Grid and During the Transition to aMarket Economy123 J. HuntWhy Do People Still Live in East Germany?124 R. T. RiphahnRational Poverty or Poor Rationality? The Take-up of Social Assistance Benefits125 F. Büchel, J. R. FrickThe Income Portfolio of Immigrants in Germany- Effects of Ethnic Origin and Assimilation. Or:Who Gains from Income Re-Distribution?126 J. Fersterer, R. Winter-EbmerSmoking, Discount Rates, and Returns toEducation127 M. Karanassou, D. J. SnowerCharacteristics of Unemployment Dynamics:The Chain Reaction Approach128 O. Ashenfelter, D. Ashmore,

O. DeschênesDo Unemployment Insurance Recipients ActivelySeek Work? Evidence From Randomized Trials inFour U.S. States129 B. R. Chiswick, M. E. HurstThe Employment, Unemployment and Unem-ployment Compensation Benefits of Immigrants

130 G. Brunello, S. Comi, C. LuciforaThe Returns to Education in Italy: A New Look atthe Evidence131 B. R. ChiswickAre Immigrants Favorably Self-Selected? AnEconomic Analysis132 R. A. HartHours and Wages in the Depression: BritishEngineering, 1926-1938133 D. N. F. Bell, R. A. Hart,

O. Hübler, W. SchwerdtPaid and Unpaid Overtime Working in Germanyand the UK134 A. D. Kugler, G. Saint-PaulHiring and Firing Costs, Adverse Selection andLong-term Unemployment135 A. Barrett, P. J. O’ConnellIs There a Wage Premium for Returning IrishMigrants?136 M. Bräuninger, M. PannenbergUnemployment and Productivity Growth: AnEmpirical Analysis within the Augmented SolowModel137 J.-St. PischkeContinuous Training in Germany138 J. Zweimüller, R. Winter-EbmerFirm-specific Training: Consequences for JobMobility

139 R. A. Hart, Y. MaWages, Hours and Human Capital over the LifeCycle140 G. Brunello, S. ComiEducation and Earnings Growth: Evidence from11 European Countries141 R. Hujer, M. WellnerThe Effects of Public Sector Sponsored Trainingon Individual Employment Performance in EastGermany142 J. J. Dolado, F. Felgueroso,

J. F. JimenoExplaining Youth Labor Market Problems inSpain: Crowding-Out, Institutions, orTechnology Shifts?143 P. J. Luke, M. E. SchafferWage Determination in Russia: An EconometricInvestigation144 G. Saint-PaulFlexibility vs. Rigidity: Does Spain have the worstof both Worlds?145 M.-S. YunDecomposition Analysis for a Binary ChoiceModel146 T. K. Bauer, J. P. Haisken-DeNewEmployer Learning and the Returns to Schooling147 M. Belot, J. C. van OursDoes the Recent Success of Some OECD Coun-tries in Lowering their Unemployment Rates Liein the Clever Design of their Labour MarketReforms?148 L. GoerkeEmployment Effects of Labour Taxation in anEfficiency Wage Model with Alternative BudgetConstraints and Time Horizons149 R. Lalive J. C. van Ours

J. ZweimüllerThe Impact of Active Labor Market Programsand Benefit Entitlement Rules on the Durationof Unemployment150 J. DiNardo, K. F. Hallock,

J.-St. PischkeUnions and the Labor Market for Managers151 M. WardGender, Salary and Promotion in the AcademicProfession152 J. J. Dolado, F. Felgueroso,

J. F. JimenoThe Role of the Minimum Wage in the WelfareState: An Appraisal153 A. S. Kalwij, M. GregoryOvertime Hours in Great Britain over the Period1975-1999: A Panel Data Analysis154 Michael Gerfin, Michael LechnerMicroeconometric Evaluation of the ActiveLabour Market Policy in Switzerland155 J. HansenThe Duration of Immigrants’ UnemploymentSpells: Evidence from Sweden156 C. Dustmann, F. FabbriLanguage Proficiency and Labour Market Per-formance of Immigrants in the UK157 P. Apps, R. ReesHousehold Production, Full Consumption andthe Costs of Children158 A. Björklund, T. Eriksson,

M. Jäntti, O. Raaum, E. ÖsterbackaBrother Correlations in Earnings in Denmark,Finland, Norway and Sweden Compared to theUnited States159 P.-J. Jost, M. KräkelPreemptive Behavior in Sequential Tournaments

12 IZA COMPACT · September 2000

O P I N I O NThe Labor Market and Ten Years of German Unification

Institute forthe Studyof Labor

IMPRESSUMEditor: Prof. Dr. Klaus F. Zimmermann

Managing Editor: Holger HinteIZA, P.O.Box 7240, D-53072 BonnTel.: +49 (228) 38 94 222, Fax +49 (228) 38 94 180e-mail: compact@iza.orgGraphics/Photographs: IZALayout/Printing: Verlag Erik Dynowski,Cologne, Germany

German reunification, originally hailed asa once-in-a-lifetime gift, now looks morelike a Danaens’ present to many observ-ers. It was supposed to pull easternGermany’s economic wagon on the jour-ney toward a “blooming landscape” –after an unprepared cold or “jump start”.But the wagon got stuck in the mudshortly after the journey began. Even ifthe level of transfer payments to easternGermany remains as high as it is today, itwould take at least another thirty years tocomplete the east-west convergence pro-cess in such areas as public infrastruc-ture.

It is often concluded that eastern Ger-many is slow to catch up because it washeaded in the wrong economic directionfrom the outset of unification. But thereis no point in blaming politicians alonefor all the undesirable developments.Economists must just as well admit totheir own errors. The metaphor of a coldor “jump start”, alluding to the extempo-rary transposition of the West Germansocial market economy into the “newLaender”, is misleading. After all, awarm-up lap certainly wouldn’t havehelped the East German “Trabbi” auto-mobile to take the pole position. What

the car really needed was a new engine –and we might as well have given it a wholenew body, too.

At the heart of this misunderstanding liesa flawed interpretation of the role ofwage costs in explaining the desolate em-ployment situation in eastern Germany. Itis simply wrong to attribute the unem-ployment problem mainly to the high andrapidly rising wages after the new cur-rency was introduced at a one-to-oneexchange rate. We must keep in mindthat the entire product range of formerEast Germany had become hard to sell –even in Eastern Europe.

As a consequence, the effective rate ofunemployment, which includes both reg-istered and hidden unemployment, stillremains at over 30 percent in easternGermany today. According to a back ofthe envelope calculation, neglecting allthe relevant structural aspects, wageswould have to fall by more than 80 per-cent in order to achieve full employmentin the “new Laender”. The suggestion toconcentrate on the wage factor is there-fore far from reality, just like the assump-tion that the eastern German labor mar-ket functions according to neo-liberaltextbook economic principles.

The extent to which wages in east andwest have converged is often overesti-mated since a comparison of unionwages is easily misleading. In the west,union wages are often much lower thanactual wages. In the east, most compa-nies do not bother to participate in col-lective wage bargaining. Therefore, col-lectively agreed wages often do not apply,or they are simply ignored. Fringe ben-efits, such as vacation or Christmas pre-miums, are still substantially lower in theeastern part of the country. From thisperspective, the eastern German labormarket is much better than its reputationwith regard to creativity and flexibility.

Therefore, the effective wages of east Ger-mans, compared to their compatriots inthe west, are not as far apart from pro-ductivity as is commonly alleged in the

public debate. But a general comparisonof productivity always tends to be decep-tive: It only works for sectors with interre-gional trade of goods, not for areas thatare dependent on the public sector or onlocal purchasing power.

It is safe to assume that the labor marketin the east will continue for a long time toplay by different rules than the west.Despite high unemployment, the averagenumber of hours worked per potentialworker in the “new Laender” is almost atthe same level today as in the rest of thecountry. Another example is the above-average education level, which – in con-trast to the west – does not effectivelyprotect workers in the east from losingtheir jobs.

Nonetheless, active labor market policy inits current form has no future. Regardlessof its function as a social stabilizer, it hasfailed economically (with the exception ofa few individual programs for further edu-cation). The political quarrel about thiskind of policy will undoubtedly remainuntil we manage to establish an indepen-dent scientific evaluation of labor marketpolicies as it has long been commonpractice in other countries.

Notwithstanding the progress made,overemployment in the public sector andan underdeveloped service sector still re-flect the structural deficits of the easternGerman economy. Rapid economictransformation has unleashed the poten-tial for a modern service economy meet-ing the challenges of the information age.But this potential has not been fully real-ized. The lack of export orientation is yetanother enormous structural deficit. EUeastward enlargement, which is alreadylooming on the horizon, may be a prom-ising opportunity to do some catchingup. Nothing will be gained, however,without a consistent improvement ofpublic infrastructure. This will remain animportant task for support from the fed-eral government.