Hartz 4 und 450€ job

Die Entwicklung von Beschäftigungsstabilität und Löhnen von Einführung der Hartz-Reformen

Journal zum Labour sector Research volume 49, pages 269–294 (2016)Cite this article

We address the concerns about rising inequality an the German labour industry after the implementation des the Hartz reforms bolzen 2003 and 2005. We focus on the quality von new jobs started bolzen 1998 and 2010 in West Germany bei terms of job stability and level of earnings. Making use of social security data drawn from the Integrated employed staff Biographies, we analysieren the distributions of job durations and wages and model their determinants at die worker level. Ours results nur a high level of job stability during und after the revolutionieren years, decreasing fairy levels and increasing wage dispersion.

Du schaust: Hartz 4 und 450€ job


Seit den Hartz-Reformen in den jahren 2003 bis 2005 gibt es Hinweise an eine gestiegene Ungleichheit im deutschen Arbeitsmarkt. Basierend der Indikatoren siedeln und Entlohnung untersuchen wir die Qualität von Beschäftigungsverhältnissen, die im auch weiterhin von 1998 bis zu 2010 begann haben. Mit administrativen scham aus ns Integrierten Erwerbsbiographien analysieren wir ns Verteilungen und modellieren individuelle Determinanten von Beschäftigungsdauern und Löhnen. Ns Ergebnisse weisen an ein hohes Maß an Stabilität in den Beschäftigungsdauern, ein Rückgang ns Lohnniveaus und ein Anstieg der Lohnungleichheit inzwischen und nach das Periode der Hartz-Reformen hin.


The Hartz revolutionary were implemented from 2003 zu 2005 after ns so-called “placement scandal” von the commonwealth Employment company (Fleckenstein 2008). The aims des the Hartz revolutionary were kommen sie improve public employment services, enhance die efficiency von active labour industry policies und decrease ns number von unemployed persons bei Germany. Among the reforms were ns restructuring von the commonwealth Employment Agency, a reorganisation von the regional employment agencies, und several young legislative changes related zu dismissal protection, fixed-term contracts und temporary agency work (see, e. g., Jacobi und Kluve 2007). Ns most vital reform, Hartz IV, abolished the then-existing joblessness assistance for long-term unemployment workers und consolidated this routine with society assistance zum households in need, thereby worsening die conditions zum a vast majority des longer-term unemployment advantage recipients.Footnote 1

In die years following the reforms, die labour market bei Germany performed surprisingly well: from 2005 zu 2008, joblessness decreased von one-third, if employment liable zu social insurance money increased von approximately the same lot (1.5 million persons, see Fig. 1). During die recession years des 2008/2009, both the decrease in employment and the increase an unemployment were modest, back GDP reduced dramatically (see Fig. 2). This phenomenon was noticed worldwide, und Krugman even called it “Germany’s job miracle”. Nevertheless, there are concerns around rising fairy inequality (Card et al. 2013) and the raising prevalence von atypical and low-wage arbeit (Eichhorst und Tobsch 2015), tendencies the may schutz laid the groundwork zum the introduction of a general minimales wage an Germany in 2015.

Fig. 1



This file takes the positive all at once performance von the German labour market following die Hartz reforms und during die great recession together a starting point for our analysis von whether over there is bei unsavoury side to this optimistic trend bei the form of lower job quality. Us use job duration und wages as signs of arbeit quality und look weist new arbeit started betwee 1998 and 2010 (the terms “job duration” and “job stability” are supplied as synonyms throughout die paper). We room interested in whether changes in job quality occurred in the “middle” von the work market and therefore analyse the stability and wages von regular arbeit covered über social security.

The quality des a specific arbeit has assorted dimensions and many des them space difficult kommen sie measure. In addition zu wages and stability, elements such together mental and physical stress, autonomy und self-responsibility, short-term versus long-term contracts, working time, and the reconciliation von family und working life schutz been die subjects of research.Footnote 2 The taste reasons zum our choice des indicators of job quality are that job stability und earnings are extremely necessary economically and that our dünn allow us to observe this indicators an extremely precisely (limitations are debated later in the paper).

The expectation that the Hartz reforms schutz increased inequality und created a generation of job that are lower paid and less steady than jobs in the past zu sein supported über search-theoretical arguments. Extensive monitoring and stricter use von sanctions will increase the job search intensity of unemployed workers. Ns worsening of conditions for unemployment advantage recipients wollen lead kommen sie lower preventive wages und increase the willingness von unemployed workers zu accept jobs of a given quality. Rebien und Kettner (2011) report the after the reforms, job applicants more often accepted arbeit with worse working conditions, e. g., longer commutes und even reduced wages. In Sweden, Van den Berg and Vikström (2014) demonstrate that die use von sanctions induced unemployment job-seekers to take arbeit with lower wages, fewer working hours und fewer qualification requirements.

The Hartz reforms deshalb contained certain deregulative elements, such as a reduction von dismissal protection bei small firms und a relaxation of the legislation on temporary company work. Die growing use des working contracts the offer less employment protection might be another source of the increasing inequality in the labour market.

There are so good reasons to expect confident effects von labour industry reforms on arbeit quality. Die matching von unemployed workers to jobs may have been positive influenced von the reorganisation des local employed agencies an unified with die tightening des active labour sector policies. Even if increased pressure on unemployment job-seekers will induce them kommen sie accept worse jobs in the first round, accepting these job might result an shorter joblessness durations, much less depreciation von human capital, much less stigmatisation and better signals zu firms an the second round.Footnote 3 Similarly, atypical job may become stepping stones right into permanent und better-paid jobs bei the medium run.Footnote 4 All an all, that is in empirical concern whether die Hartz reforms schutz had a positive or a negative influence on arbeit quality.

It need to be detailed that the extent to which ns Hartz reforms schutz influenced die German labour market zu sein controversial. In a recent und prominent paper, Dustmann et al. (2014) controversy that die scale of the Hartz revolutionary was not sufficiently large kommen sie substantially contribute to die positive changes bei competitiveness and unemployment it was observed after the Änderung period. These authors stress die importance des factors such together flexibility in the German wage-setting institutions and the growing importance des firm-level negotiation of wages since ns restructuring von the German economic climate after reunification. However, another view zu sein that die Hartz reforms and “especially ns merger between unemployment assistance und welfare, deeply adjusted the basic labor sector institutions” (Möller 2015, p. 164). Similarly, ns reduced generosity von the joblessness benefit system as a result of the Hartz reforms is considered a key determinant des the wage mäßigung during the years preceding die great recession (Gartner and Merkl 2011).

Although us share ns view that die Hartz reforms had an important affect on the German labour market, we do not aim kommen sie estimate ns size of this impact or kommen sie disentangle that from other components that schutz contributed to ns positive labour market outcomes observed in recent years. Our outcomes on job durations and wages room descriptive in character; nevertheless, us will also perform econometric model analyses that account weil das worker und job heterogeneity.

We contribute neu evidence zu previous research in several ways. First, vice versa, existing studies schutz focussed primarily on wages, we analyse both wages and job duration due to the fact that both aspects are radikale dimensions of job quality. Second, a distinctive feature of our analysis ist that we choose cohorts of newly began jobs. Us expect the employment contracts des workers acquisition new jobs to respond immediately zu changes an both legislation and market conditions. Third, us update die evidence through 2010, providing us a period des analysis that enables us kommen sie consider a sufficiently lang time expectancy (five years) after the Hartz reforms.

The paper is organized as follows. In Sect. 2, we review the literature on job stability and wages in Germany. An Sect. 3, we describe our empirical strategy, and bei Sect. 4, we present ns results. Die paper ends with concluding remarks.

There are several macro-level lernen analysing the influence des the Hartz reforms on accumulation matching efficiency. Fahr und Sunde (2009) estimate optimistic effects of the zuerst two revolutionieren packages (which addressed ns organisation of the commonwealth Employment Agency and local employment agencies) top top both die duration von unemployment and the matching probability. Subsequent aggregate studies confirm the improved equivalent efficiency following die implementation of the Hartz reforms (Hertweck and Sigrist 2013) and the increase in job-finding rates, notfall only for the momentary unemployed but deshalb for die long-term unemployed (Klinger und Rothe 2012). Launov and Wälde (2016) analyse the impact von different elements des the Hartz reforms utilizing structural equilibrium models. Compare pre- and post-reform secure states, die estimated effect of the reform of the windy employment agencies in Germany accounts for approximately 20 % von the reduction in equilibrium unemployment, whereas the Hartz IV reform accounts for another 5 %. Bring away together, these studien confirm that that ns functioning des the German labour market improved after die reforms. Die reform des the benefit system through ns Hartz IV revolutionieren seems zu be von minor importance in the macroeconomic context.

The cyclical behaviour of aggregate work turnover in Germany provides a first impression des whether jobs have become more or much less stable since die mid-nineties (Rothe 2009). Fig. 2 shows die overall trends in hiring and separation prices over the unternehmen cycle between 1996 und 2012. Beginning with die year 2001, both rates plainly drop and continue zu decrease during ns period of the Hartz reforms. Remarkably, both hiring and separation rates remain relatively low during ns cyclical upturn following die Hartz reforms (instead des returning kommen sie higher levels) and an the recession year 2008 and 2009. Thus, there has actually been a reduction in worker turnover bei Germany that began before ns Hartz reforms and continued thereafter. If job stability had actually decreased considerably after the Änderung period, us would oase expected to see a corresponding increase in the sales rates. This erste piece of evidence wollen be confirmed und complemented über analyses that account for job heterogeneity in our micro-level analysis.

Regarding present micro-evidence weil das Germany, there are lernen that have to be mentioned even if most of them do notfall refer to die Hartz reforms. These lernen look at long-term trends in job security or wages, and several des them focus explicitly on inequality und the processes of sorting or polarization in the labour market.Footnote 5

Most analyses des long-term trends an individual job duration indicate a constant or decreasing level of job stability. Bergemann und Mertens (2011) find a tendency towards shorter arbeit durations zum the duration 1984 to 1998, however other studien present evidence des a rather constant level of job stability until the middle des the last decade (Giannelli et al. 2012; leitung 2010). Giesecke und Heisig (2011) look hinweisen the long period from 1984 kommen sie 2008 and thus hülle the first years after ns Hartz reforms. They uncover that as whole mobility bolzen firms has actually remained reasonably constant, whereas much less qualified workers oase experienced a decline in job stability over time.

Boockmann und Steffes (2010) focus on the determinants of job duration in Germany and demonstrate that in addition to die socio-economic characteristics von workers, ns internal structure von firms, as captured von the existence von working councils or further training opportunities, contributes kommen sie longer arbeit duration. Furthermore, the results indicate a sorting process des workers with higher expected duration to firms offering an ext stable employment.

The recent literary works on wages consists of influential contributions that demonstrate a rise an wage dispersion in West Germany end the tonnage several decades. Dustmann et al. (2009) emphasis on wages between 1975 and 2004 and document that inequality has actually continued kommen sie rise since ns 1980s. Bei the 1980s, it was mainly the upper half of the wage circulation that was affected, but bei the early 1990s, inequality also started kommen sie increase at ns bottom für hilfe of ns wage distribution. Antonczyk et al. (2010) find specifically pronounced growth von wage inequality at the bottom von the wage distribution from 2001 zu 2006. Map et al. (2013) look at full-time workers in the years from 1985 zu 2009 bei West Germany and demonstrate that enhanced heterogeneity an the establishment component of wages strong contributes to ns rise in wage inequality. They deshalb find evidence that (positive) assortative matching has increased, with workers’ fairy potential being much more closely associated with firms’ wage premia over time. Cornelißen and Hübler (2011) estimate the influence des unobserved individual und firm heterogeneity on wage und job-duration attributes using German connected employer-employee säule for the years 1996 kommen sie 2002. Somewhat punkt odds with ns results of Card et al. (2013), the estimated individual and firm effects show that high-wage workers tend to be stable workers and are more likely kommen sie be employed in low-wage firms, conversely, low-wage employees are an ext likely to accept job with shorter durations in high-wage firms.Footnote 6

The study von Riphahn und Schnitzlein (2016) analyses individual fairy mobility, i. e., ns probability des shifting to a different quantile von the wage distribution or changes an the rank positions, and shows a substantial decline an wage mobility end time. Die results are provided for east Germany, where ns decline an wage mobility begins an the beforehand 1990s, und for west Germany, whereby decreasing wage mobility has actually been observed since ns late 1990s.

The paper über Arent und Nagl (2013) is verwandt, angemessen because that claims kommen sie estimate the effect von the Hartz reforms on wages. Their central result, a structural break an the wage equation in the year of the benefit reforms, differs from ns findings discussed above and from ours (we uncover that wages started zu decrease before 2005; see deshalb Ludsteck und Seth (2014) for a comment).

Our empirical analysis borrows from die literature discussed in this section, especially wie man we existing evidence on empirical distributions and an the specification of our modell of arbeit durations und wages. Although the sorting des workers kommen sie firms seems zu be a promising road for further research, the results von these studies are not unambiguous. Our very own analyses room performed at the individual or job level, which seems to be adequate weil das our study question. Our aim is to assess ns changes bei job top quality after die implementation von labour sector reforms that reportedly improved die functioning of the labour market and led to ns creation of neu jobs. We use stability und wages as indications of arbeit quality and analyse them for a large share of the work market.

We start our evaluation with in assessment des overall trends bei the distribution of job durations und wages. Us then estimate job duration and wage modell to determine ns size and direction von changes an job quality across the reformen period while steuerung for die heterogeneity of jobs and workers. Finally, we look punkt the job quality over time of three groups von disadvantaged workers, namely, unskilled workers, previously unemployed workers und temporary company workers.

Following a relevant strand von the economic literature, us measure job quality von the level des wages und by arbeit duration: die higher ns wage is, ns better the job is; the longer a job lasts, the better the job is (Jahn und Rosholm 2014 or Caliendo et al. 2013).

We usage a flow sampling approach und select cohorts of jobs starting an the very same year or period. This approach avoids oversampling of longer durations, well-known as length-bias (Cameron and Trivedi 2005). A drawback des this approach ist that our sampling returns a large share des incomplete arbeit spells because von the restricted observation period. Due to the fact that we deshalb censor arbeit after 24 months, ours analysis von changes an survival probabilities und job-leaving risks over time ist confined zu changes in the anfangsverdacht two years an a job and is based upon samples with huge numbers von censored job spells.

We study entry incomes (wages at the time des the first notification des a worker with a specific employer), which guarantee that us do not mix wages zum new jobs with the wages des incumbent workers who schutz already obtained some endure on ns job. Because new job will strongly reflect current labour market conditions, changes in the industry wage or in increase bei temporary, unstable job will be revealed quite quickly bei our samples. With respect to wages, we see our analyses as complementing other lernen based on stange samples.

The job duration model

To analysieren job durations, us estimate the following piecewise exponential blended proportional hazard modell (Blossfeld et al. 2007; Cameron und Trivedi 2005; Gutierrez 2002):

$$lambda _ijleft (t|x_i" , eta , u _i ight )=lambda _0left (t ight ) , exp left (x_i", eta ight ) u _i, quad i=, 1,ldots ,, N;, j=, 1,ldots ,, J$$


$$lambda _0left (t ight )=lambda _j, quad au _j-1

where (lambda _ij) zu sein the risk rate representing the risk des leaving die current employee during ns j thtime interval des the job spell belonging zu individual i. Arbeit durations are break-up into punkt most J time intervals (pieces) to modell changes in the threat of job termination conditional on die time already spent bei the current job. The jth interval starts punkt duration ( au _j-1) und ends hinweisen duration ( au _j). Ns baseline danger (lambda _0left (t ight )) ist a step function that ist constant within intervals. Unobserved heterogeneity ist modelled von a gamma-distributed frailty hatchet ( u _i) that is assumed zu be specific to the job spell. Only the variance ( heta) des the frailty term zu sein estimated.

The vector (x_i) ist a set of individual, job, firm, industry, und macroeconomic time invariant explanatory variables that are details for in individual at ns beginning von the job.

The fairy model

To analysieren wages, we calculation censored regression modell that enable us to take die threshold weil das social defense contributions into account (Cameron and Trivedi 2005).

We model the it was observed censored fairy (W_ij)of individual i in year j as the realization des a latent variable(W_ij^*):

$$W_ij^*=x_ij" , eta +varepsilon _i$$

where ns j is the threshold zum social insurance money contributions, which varies end time.

The coefficients (eta) measure ns influence of the covariates on ns latent change (W_ij^*.)The best likelihood estimation returns results for the vector of coefficients also as zum the variance (sigma ^2).

Mehr sehen: ' Game Of Thrones 8 5 Online: The Bells, The Bells (Game Of Thrones)

Data und variables

We attract our dünn from a two % sample of the German combined Employment Biographies (IEB), a large administrative charme set regularly developed at the Institute zum Employment study (IAB) an Nuremberg.Footnote 7 ns IEB includes employment liable to social insurance however does notfall include self-employment or civil servants. We emphasis on workers’ employment spells und wages indigenous 1998 zu 2010.

The analysis wollen concentrate on west Germany i m sorry accounts weil das more than 80 % of the German labour market. A common analysis zum East und West ist hindered über the big differences the continue zu exist bolzen the two labour markets.Footnote 8 Selected outcomes on in its entirety trends in East Germany are contained in the appendix.

We pick workers aged 25 kommen sie 54 wie man starting your jobs. Employees under 25 or end 54 are excluded due to the fact that the arbeit exits des younger workers often coincide through re-entry into ns educational system and the transitions von older workers could be influenced von the different option von retirement. Because our focus is on employment spells, we also exclude apprentices des any age.

The duration of a job zu sein defined as a period des employment in the same establishment.Footnote 9 successive sub-spells within die same facility are concatenated zu generate the job spells, allowing weil das a maximum von 90 days des interruption. Die start und end of the job spells space measured exactly in days. To keep the observation window as lang as possible, us look at durations up kommen sie 24 months, therefore censoring longer arbeit spells. An the case of overlapping arbeit spells, we just keep the spell with die highest amount von earnings.

Entry wages are measured as those of the first sub-spell (lasting punkt most one year) contained in a possibly much longer employment episode. We deflate the nominal amount with the German consumer preis index zu obtain ns real jeden tag wage. Right-censoring occurs at ns threshold zum social defense contributions, which zu sein adjusted nearly every year.Footnote 10 ns exact amount von wages above die threshold zu sein unobserved.

Even if the charme allow us zu distinguish bolzen full-time und part-time work, they do notfall contain die info on hours worked. Consequently, back we incorporate part-timers (those liable kommen sie social insurance) bei the analysis of job durations, we are forced zu exclude lock from the analysis of wages, i beg your pardon implies ns neglect of a considerable part des the female workforce.Footnote 11 in addition, we exclude “mini-jobs” und other part-time arbeit with brief hours or low earnings from our analyses. Mini-jobs are legally required to fall below a certain threshold (e. g., 400 € bei 2012) und are largely exempted indigenous social defense contributions. The growth von such jobs in recent years is certainly relevant to ours research inquiry (Jahn et al. 2012; Möller 2014). However, early out to ns lack of information on hrs worked and because earnings from mini-jobs often constitute a type von extra income, these arbeit are daunting to analyse without any information on die household context.

With respect zu individual characteristics, die administrative data contain info on workers’ age, gender and nationality. Info regarding workers’ ability levels combines different types of school und university training v a binary indicator of whether die worker has completed vocational training. We encompass indicators of labour market zustand preceding the neu job, differentiating “jobs taken ~ unemployment” (where unemployment zu sein defined von benefit receipt or being registered with die local employed staff agency), “job-to-job changes” (allowing zum 31 days of non-employment betwee jobs), “gaps” (periods without entries an our data) und “first spells”, which are workers’ zuerst appearance bei the IEB. Jobs with fixed-term contracts contribute to our results on durations and wages however cannot be distinguished from long-term jobs an our data. About temporary firm work, die Hartz reforms implied a considerable liberalization und the increase in the number of temporary agency workers has been one des the taste topics in the controversy on current labour sector changes bei Germany. Bei contrast zu fixed-term contracts, temporary company workers can be identified reasonably well using ns industry password (see Antoni und Jahn 2009).

We encompass industry and firm size class an our modell as info on establishments.Footnote 12 to take account of unternehmen cycle effects, we include indicators weil das the state of the local work market: GDP growth obtained from the German federal States’ Accounts and the regional unemployment price made available von the commonwealth Statistical Office, through both variables measure at ns level of the west German districts (Kreise). Seasonal effects are modelled von the inclusion von dummy variables for the quarters of job entry.

In die analysis von overall trends an job quality, we focus on yearly cohorts of new jobs. Bei our regression modell of arbeit duration and wages, we group the observations right into three sub-periods: arbeit beginning bei the duration before the reforms (1998 zu 2002), job beginning during the period des the Hartz reforms (2003 zu 2005) and jobs beginning in the post-reform period (2006 zu 2009/2010).

We zuerst discuss the results of the overall evaluation of arbeit duration and wages an new jobs (Sect. 4.1). Us then turn zu the modell analysis (Sect. 4.2) and, finally, to die results concerning the three selected groups of disadvantaged workers: unskilled workers, temporary firm workers, und previously unemployed employees (Sect. 4.3).

Trends in the distribution of job durations und wages

The duration von a new job is defined as in uninterrupted period of work with the same employee (see Sect. 3.3). We embrace Kaplan-Meier survival function estimators to plot time trends von the distributions of arbeit durations. Zum wages, we compare die quartiles (25th, median and 75th percentile) of the distribution des log wages. To visualize broad gruppe differences, we erste distinguish by gender und then von skill level.

Durations von newly started jobs

Fig. 3 shows die estimated probabilities of staying an a job hinweisen specific durations after the job begins for job spells started between January 1998 und December 2009 (December 2008 for survival probabilities of 18 and 24 months).

Fig. 3


Job stability is fairly high. Die estimated 12-month survival probabilities zum male employees are bolzen 50 und 60 %, and they are even higher zum female workers (see so Table 4 in the appendix). Over time, the survival probabilities for male workers seem zu be fairly stable, whereas zum female workers they show a tendency towards longer job durations. There ist a temporary decrease in survival probabilities zum jobs beginning in the year 2000; this decrease zu sein more pronounced for women and for much longer durations. We carry out not have a ready explanation for this result;Footnote 13 in any case, the size of this reduction is rather restricted (up zu 4 percentage points zum women’s 24-month survive probability, see Table 4 in the appendix).Footnote 14 Overall, the vertical distances betwee the survival probabilities weist different durations remain relatively constant, definition that within ns groups von male und female workers, job durations do not become much more unevenly distributed over time (see deshalb Fig. 7 an Sect. 4.3). As an intermediate result, there space no signs of a general bottom trend an job durations during or after the Änderung period, i beg your pardon confirms the aggregated turnover rates presented an Fig. 2.Footnote 15.

Fig. 4 distinguishes between three ability levels (see deshalb Table 5 in the appendix). Ns category of “unskilled” workers comprises those with lower than tool education and without vocational training. Die category des “skilled” workers has workers with up zu a medium level des education, workers v vocational training, und workers v a degree that qualifies them zum professional college or universität attendance. Workers in the “highly skilled” category have college or universität degrees.

Fig. 4


For a given duration, highly professional workers schutz the greatest survival probabilities, conversely, unskilled workers have the lowest. End time, ns survival probabilities diminish somewhat for unskilled workers, specifically at arbeit durations von 18 and 24 months. Ns opposite ist true zum highly professional workers; their survival probabilities zeigen a slight increase trend, which is stronger for longer job durations. Because die observed transforms are quite small, this evidence indicates the heterogeneity in job durations in ~ skill groups has deshalb not increased substantially over time.

Wages in newly started jobs

This analysis zu sein confined kommen sie full-time workers, because die info on hrs worked zu sein not available weil das part-time employees (see Sect. 3.3).

By now, ns growth in earnings inequality end the tonnage several decades and a steady decline in real earnings since die early 2000 s is fairly well established weil das Germany, as well as zum other industrialized nations (Dustmann et al. 2009, card et al. 2013). Through its emphasis on neu jobs, our evaluation presents further proof on this issue.

The fairy variable is measured as ns log of jeden tag real wages in 2005 prices. An Fig. 5, the 25th percentile, median, and 75th percentile des the wage distributions zum men und women are plotted over die years (exact figures zum the year 1998 und 2010 tun können be found bei Table 6 in the appendix).

Fig. 5


In line with the results of Card et al. (2013), die decrease in real incomes after 2001 is clearly visible. For both men und women an new jobs, the decline in wages ist strongest in the 25th und 50th percentiles of the fairy distribution, whereas die decline in the 75th percentiles is comparatively modest. Furthermore, fairy inequality has been increasing among both men und women. Ns lower interquartile distinction (50th percentile minus 25th percentile) has increased von 8 % zum men und by 4 % zum women, and the top interquartile distinction (75th percentile minus 50th percentile) has actually increased über 6 % zum both men and women (Table 6).Footnote 16

As formerly noted, “wage moderation” ist one potential explanation von Germany’s relatively great employment power during the great recession bei 2008/2009. In accordance through most von the studien summarized in Sect. 2, ours results zeigen that die decrease in real wages starts well before die Hartz reforms. Thus, one impact of the reforms might have been zu strengthen a pre-existing propensity towards reduced wages. Die size of wage losses bei our sample of neu jobs zu sein clearly larger than the wage reduce reported an Card et al. (2013).Footnote 17

The differences in entry wages between skill level are considerable (Fig. 6, Table 7 in the appendix). Furthermore, the disadvantage von unskilled workers zu sein increasing. The median log in wage von highly professional workers, weil das example, reduced from 4.79 an 1998 zu 4.63 in 2010, implying a daily fairy loss of 15 % (approximately 18 €). The corresponding typical wage loss zum skilled workers from 1998 kommen sie 2010 was so 15 % (10 €). However, ns median wage loss zum unskilled workers was 21 % (11 €; lakers Table 7). An addition, within-group inequality has actually been rising, which deshalb accords with ns results von Card et al. (2013). Weil das the skilled und highly professional worker groups, die decline bei wages ist strongest bei the 25th percentile (that is, for workers with ns lowest earnings in these groups).Footnote 18 Footnote 19

Fig. 6

Model analysis

Job duration

Although no general propensity towards shorter arbeit durations end time was revealed von the analysis bei Sect. 4.1.1, we jetzt consider the possibility that changes an the composition des jobs have influenced this result. We calculation a duration model that controls zum a broad range von explanatory variables. The objective ist to identify whether die period bei which the Hartz revolutionary were implemented or ns subsequent duration are damit verbundenen with changes in job durations when we take into account changes in the structure von jobs, together captured von our covariates. Since we do notfall use an experimental or quasi-experimental design, die estimated covariate effects reflect empirical associations quite than causal influences.

The model ist a single assignment model, approximated separately zum men und women, weil das employment spells beginning betwee January 1998 und December 2008. Zum persons with an ext than one new job in the monitoring window, us randomly choose only one spell.Footnote 20

The outcomes are presented in the form of hazard ratios, i m sorry – an the case des binary or categorical variables – indicate die influence of a variable on die risk von leaving the job relative to ns reference gruppe (see Table 1). Die relationship bolzen the risk of terminating a job und job duration is inverse, meaning that a value of the peril ratio better (smaller) 보다 one means a positive (negative) result on ns hazard and a negative (positive) impact on duration for that variable. Ns z‑values bei Table 1 room based on the cluster-robust estimation von the variance-covariance matrix, because the district-level variables von unemployment rate und GDP expansion give rise kommen sie a within-district correlation of regression modell errors (Cameron und Miller 2015).Footnote 21

Table 1 job Duration Models zum West Germany
Full dimension table

Starting with our variables des interest, ns time period indicators of the year von entry are smaller sized than one for both men and women during the Änderung period (2003–2005). For the post-reform duration (2006–2008), the hazard proportion continues zu be smaller than one zum women but ist insignificant for men. This results suggest that job durations ceteris paribus are somewhat longer weil das new job started bei the revolutionieren period (2003–2005) contrasted to ns reference period (1998–2002). Zum men, in the year following the Hartz reforms, durations do notfall differ indigenous those in the reference period, whereas zum women, job durations continue zu be slightly much longer than those in the duration preceding the Hartz reforms.

For the rest of the models, ns pattern of the baseline hazard, ns associations of the covariates with die hazard rate und the relevance von the frailty hatchet are des interest together well. Ns pattern des the approximated risk des leaving a job is non-monotonic: ns hazard ratios zum the time periods at ns top von the table originally slightly boost (for women, over there is deshalb a temporary decline bei the third month) und then anfang to decrease after the zuerst three months of employment. This non-monotonic pattern can be explained von the high frühen zeitpunkt risk of contract dissolved if a mismatch ist found bei combination with die stabilizing effect von job-specific human funding (Blossfeld et al. 2007, p. 121). An addition to this baseline pattern – and despite die large number des covariates – the importance of unobserved heterogeneity zu sein confirmed bei all des our models von likelihood ratio tests that are highly significant.

The results also point to a clear seasonal pattern an job durations. Die estimated danger ratios from die men’s modell are large and increase during ns year, implying the longest durations zum jobs beginning in the first three calendar months. Zum women, this pattern zu sein less regular and less pronounced.

The state of the local work market can influence job durations through fluctuations bei labour demand. Die respective manage variables room insignificant, however. Persistent differences an economic conditions between regions might deshalb influence arbeit durations. The estimated hazard ratios for the federal claims seem to reflect kommen sie some extent ns north-south divide in Germany. Die reference ist North Rhine-Westphalia, which zu sein the zustand with ns largest population and is situated bei the western centre. Significantly shorter job durations space estimated in the north states of Schleswig Holstein und Hamburg, und longer arbeit durations room estimated in the southerly states. Zum the state des Hesse, situated an the southern centre, und the economically solid states Baden Württemberg and Bavaria, which are both situated in the south des Germany, we discover lower risks des leaving jobs and thus much longer durations (in Bavaria, this is only found weil das women).

The effect of firm size is clear and rather large. Together predicted von theories von internal labour markets, arbeit duration monotonically rises with die size of the firm zum both men and women.

For part industries, we can deshalb note relatively strong und significant effects. Contrasted to unternehmen services (the recommendation group), jobs tend to be an extremely stable weil das men in manufacturing und for women an social and public services. Really large and significant danger ratios zum both men und women are estimated zum temporary firm jobs, indicating far shorter durations. Back this confirms the general id that momentary agency arbeit are of low quality, a more careful evaluation would need kommen sie look hinweisen subsequent job to see how plenty of temporary agency job lead kommen sie regular jobs (e. g., Jahn and Rosholm 2014).

Part-time arbeit with a minimum von 18 h per week are viel shorter zum men yet only slightly more stable zum women. One interpretation of this finding is that zum many men, part-time arbeit are just temporary options until they uncover a full-time job. Frauen might schutz longer durations in part-time jobs, back these job are reasonably hard kommen sie find.

Mehr sehen: Weenect Dogs Gps Tracker Hund Ohne Vertrag, Gps Sender Mini

Age und nationality oase the intended effects: arbeit duration zu sein much shorter zum younger period groups and foreigners. Die skill effect zu sein pronounced zum both men und women. Workers zum whom info on skill was missing und unskilled workers oase far greater risks des leaving their job compared to die reference gruppe (vocational maintain with at least an intermediate degree), i m sorry indicates ns importance of completed vocational training an Germany. Amongst workers with greater educational levels, only males with a university or similar degree oase more stable arbeit than ns reference group. Among ns categories describing ahead labour sector status, die reference gruppe (job-to-job changers)Footnote 22 zu sein the largest group. Persons start new job out von unemployment form another considerably group. Return the kopieren, gruppe of persons starting a new job after a gap is also fairly large, the kopieren, gruppe of persons in their first jobs and the group of persons that are zuerst unemployed und then have a gap before starting their current arbeit are both fairly small. Ns results zeigen clearly that compared to job-to-job changers, every other groups experience shorter arbeit durations. The negative association with duration ist strongest zum persons who room unemployed immediately before beginning a job.


As bei the term model, we use one randomly selected employed staff spell über person beginning bei the period 1998 to 2009. We estimate the modell separately weil das men und women in West Germany (see Table 2).Footnote 23