Facilitating structural change and preventing long-term unemployment.
Employment reacted with a long lag to the recession...
In the years prior to the current crisis, unemployment has fallen continuously with the number of unemployed decreasing by one-third from 2005 to 2008. While the decline in unemployment came to a halt at the end of 2008, the reaction of the German labour market to the crisis in the real economy was unusually small, both with respect to previous downturns in Germany and compared to other countries in the current crisis (Figure 2.1). The increase of the unemployment rate during this recession amounted to 1/2 percentage point up to the third quarter of 2009, compared with a rise by 3 percentage points in the OECD area. To some extent, this development may still reflect the extensive labour market reforms that have taken place. The reduction in the generosity of unemployment benefits through the Hartz IV reform (which merged the unemployment and social assistance benefit schemes) and the shortening of the duration of unemployment insurance benefit payments are estimated to have lowered the NAIRU by around 1/2 percentage point (OECD, 2008). This effect is spread over several years and may in 2008 still have lowered structural unemployment, thereby countering some of the adverse impact of the crisis (Boss et al., 2009). One indication for such an adjustment is the further inward movement of the Beveridge curve between January 2008 and January 2009, which suggests that the matching process continued to improve.
[FIGURE 2.1 OMITTED]
... but hours worked have been reduced significantly
More flexible working time arrangements
The focus on unemployment neglects the substantial adjustment of labour input that has taken place through a reduction in working time. (1) Hours worked per employee have fallen much more than in previous downturns (Figure 2.2). The evidence suggests that most of this adjustment during 2009 can be explained by more flexibility at the company level with regard to working hour arrangements, such as an increased use of exemption clauses from collective bargaining contracts and adjustment of working time on the company level as needed. To some extent this may reflect an increased willingness to compromise by employees as a result of the labour market reforms earlier in the decade. Such arrangements provide the opportunity to reduce working time for a defined period with a proportional reduction in wages. (2) Around 40% of the working time reduction in 2009 is accounted for by such a reduction in weekly working hours in collective wage bargaining contracts. Further factors that helped in the reduction of working time are the increased use of working hour accounts and a reduction in paid overtime (each accounting for around 20% of the decrease in hours worked in 2009). In 2005, one-third of all companies (covering half of all employees) had established working hour accounts. By contrast, in 1998 only one-fifth of all companies and one-third of employees were using such accounts. The manufacturing sector is a pioneer in this regard with almost 60% of employees having such accounts.
[FIGURE 2.2 OMITTED]
The extension of the short-time worker scheme has prevented lay-offs ...
A second factor, which contributed around a quarter to the reduction in working time in 2009, is the wide-spread use of the short-time working scheme. (3) With this scheme, the Federal Employment Agency, under certain conditions, subsidizes part of the foregone income of employees if a company reduces working time for economic reasons. Firms use short-time work as long as opportunity costs of dismissal and rehiring in the next upswing are higher than the costs of current labour hoarding via short-time work. The aim is to prevent "excessive" layoffs, i.e. dismissal of workers due to temporary difficulties of the company even though the jobs would be viable in the long run. As part of its fiscal stimulus programme the government eased the prerequisites for short-time work, lengthened the duration of entitlement and reduced the costs for companies (Box 2.1). In the recession, the number of short-time workers jumped to around 1.5 million at the peak in mid-2009, similar to the levels reached during the recession in 1993 (Figure 2.3). (4) On average, the reduction in working time per short-time worker amounts to one third. Thus, during the summer of 2009 the use of short-time work schemes prevented roughly 500 000 lay-offs, preventing a rise in the unemployment rate by around 1 percentage point.
Box 2.1. The German short-time worker scheme and recent changes The subsidy scheme for shorter working hours is meant to reduce the labour costs of companies that are in temporary distress. Companies are eligible for the subsidy if they face a major drop in orders due to economic reasons or extraordinary events, provided the drop in orders is temporary in nature. (1) At least one-third of the employees must lose more than 10% of their gross wage for the company to be eligible, though this condition is suspended from February 2009 to December 2010 (instead, the subsidy can only be paid to employees who lose more than 10% of their gross wage). The scheme can only be introduced if it is agreed in a collective agreement, a company agreement or the working contract of the individual employee. If a works council is active in the company it also has to agree to the implementation of the scheme. Employees participating in the scheme have to accept a cut in their monthly income, as the state only pays 60% of the foregone net wage (67% if the employee has children). (2) Short-time working compensation is paid to employers who must transfer it to employees. Employees remain insured in the social security system as the companies continue to pay their part of social security contributions and the employees' part on 80% of the foregone salary. (3) As part of the government's fiscal stimulus programme, 50% of the social security contributions paid by the employer (100% if the employee is in a non-firm-specific training measure) are reimbursed by the labour office in the first six months and no social security contributions have to be paid after 6 months; these measures expire at the end of 2010. The maximum duration of eligibility is generally limited to 6 months, though it has recently been extended to 24 months for all employees who become eligible before the end of 2009, and to 18 months for those who become eligible afterwards but before end-2010. This allows employers to use the scheme until mid-2012, at least for those workers who start shorttime work in December 2010 (but the costs for employers are higher from 2011 onwards as social security contributions will no longer be reimbursed). Since end-2008, the scheme may also be used for workers employed by temporary work agencies. (1.) Apart from economic reasons, short-time work may be applied by companies due to seasonal effects (e.g. construction companies in winter; so-called seasonal short-time work) or during periods of restructuring (so-called transfer-short-time-work). (2.) Some collective agreements stipulate that the benefits paid by the labour office are to be extended by the employer. (3.) This relates only to contributions for health- and long-term care insurance as well as pension contributions. Contributions to unemployment and accident insurance do not have to be paid (Bach and Spitznagel, 2009).
[FIGURE 2.3 OMITTED]
Even though firms' labour costs are reduced through short time work, the reduction is not one-to-one to the decrease in working time. This is because firms have residual overhead costs of employment. For example, wage bargaining agreements often entail some topping up of short-time work transfers paid by the Federal Employment Agency (or prevent a further reduction in salary below a certain floor, even if working time is reduced by more) or have other forms of salary guarantees. Also, indirect labour costs, such as social security contributions on the foregone part of the employee's income, other costs like special payments regulated in wage agreements, and paid annual or sick leave all add to the wage bill. These indirect costs are not reduced during short-time work as they are dependent on the existence of an employment relationship and not the actual working time. Bach and Spitznagel (2009) estimate that these residual overhead costs during shorttime work amount to between 24% and 35% of the gross hourly labour costs, depending on the extent of the government's subsidization of social security contributions. Without the subsidization of social security contributions in the government's fiscal stimulus package, the residual costs are estimated to amount to between 46% and 59% of gross hourly labour costs.
... but may also hamper structural adjustment
While in general, short-time work schemes can promote equity (by letting all employees work less rather than laying off a few and letting the others continue to work full-time) and efficiency (by preventing transitory factors from destroying valuable job matches), they may also lead to displacement effects (OECD, 2009a). These occur if the programme ends up supporting unviable jobs. One indication that this may partly be the case in Germany is the large share of short-time work in the car industry, which is well known to suffer from overcapacity (OECD, 2009b). (5) There is thus a trade-off between the immediate concern of supporting existing jobs and the longer-term objective of facilitating the reallocation of jobs and workers towards the most productive firms and sectors. Subsidising training while workers are on short-time does not change the picture that a short-time worker scheme is protecting jobs but not workers. (6)
It is very difficult to assess ex ante whether economic difficulties are only temporary and the jobs are viable after the subsidy ends. It is therefore essential to preserve strong incentives for workers and firms to leave subsidised short-time work programmes at the earliest opportunity. In this regard, a further prolongation of the extended generosity of the scheme beyond the originally legislated date should be avoided. A further option to improve incentives under the scheme would be to require employers to repay some of the short-term benefit paid to the employee if workers are laid off in the period that follows the end of short-time work. This safeguard is used for a similar short-time work scheme in the Netherlands. Also, the duration of the subsidy should be limited and employers should participate in the costs. One way to maintain search incentives for workers on short-time is to taper the replacement rate paid to the worker over time. The authorities should closely monitor the effects of the extension of the maximum duration of short-time work to 24 months (18 months for those who started short-time work in 2010) and the lowering of the costs to firms to ensure the scheme does not inhibit necessary structural change. In this respect, the generous refunds of the social contributions paid for hours not worked should be reviewed. It is not clear, for instance, why social insurance rights (notably pensions) must continue to accrue also for hours not worked. This amounts to an implicit subsidy that comes on top of the short-time work benefit paid.
Labour hoarding also due to prior skill shortages
Even after taking into account the substantial reduction in hours worked, hourly labour productivity still fell significantly, indicating that firms voluntarily hoarded labour to a larger extent than in previous recessions. One explanation for this behaviour relates to difficulties in finding qualified labour. Firms take into account the costs of hiring new employees in the next upswing and weigh these against the cost savings they could have now by laying off employees. In this regard, it is noteworthy that during the previous upswing, several sectors experienced a shortage of skilled labour. This related foremost to some manufacturing sectors where up to one half of companies said they were experiencing difficulties in filling vacancies in the final quarter of 2008, mostly because of skill shortages (Heckmann et al., 2009).
During the crisis, those sectors that suffered the largest output declines tended to be those that were experiencing labour shortages before (Figure 2.4, panel A). This may help to explain why companies are hesitant to lay off workers even if production levels are still low. Consequently, those firms also tended to be those that make most use of the shorttime work scheme in view of potential problems in finding qualified staff during the next upturn (Figure 2.4, panel B). Not least, such behaviour reflects restrictions regarding the immigration of skilled labour into Germany (Chapter 5). Lifting these restrictions would arguably make it easier for firms to adjust their employment across the business cycle, thereby lessening the need to hoard labour.
[FIGURE 2.4 OMITTED]
Strict employment protection increases the risk of labour market duality
A serious and long-standing structural labour market weakness is strict employment protection legislation (EPL). Germany ranks 9th among the 30 OECD countries in terms of the overall strictness of EPL (Figure 2.5, panel B). (7) The strictness of overall employment protection legislation is primarily due to the regulation of regular job contracts, which is one of the strictest among OECD countries, at least for companies with more than 20 employees. (8) By contrast, protection of temporary contracts is around average, primarily reflecting relatively light protection of fixed term contracts, while protection of temporary agency contracts is more in line with that of other countries. In 2008, the difference between regulation of regular and temporary work contracts was one of the largest among OECD countries, primarily reflecting significant deregulation of temporary work since 2000 and slightly stricter regulation of regular work contracts over the same period.
An important consequence of the increased divergence between regulation of permanent and temporary workers has been the growing importance of temporary workers in total employment. The share of temporary workers has risen from 12.7% in 2000 to 14.6% in 2008, while in the average OECD country the share rose from 11.6% to 12.3% over the same period. Temporary employment is thus rapidly gaining ground. For example, in the year 2007, almost half of all recruitments were done on a temporary basis; for companies with more than 250 employees, the share even amounted to two-thirds (Hohendanner and Gemer, 2010). This experience is in line with evidence showing that the strictness of regular EPL is associated with a larger incidence of temporary employment and negatively related to the transition rate from fixed-term contracts to permanent employment (Grubb et al., 2007; OECD, 2004).
The downturn shows features of growing labour market segmentation ...
One of the consequences of such an evolving labour market duality has been the divergent development of regular and temporary agency workers during the crisis. While the number of overall employees subject to social security contributions fell by half a per cent in the year to August 2009, the number of workers on temporary work agency contracts declined by almost 30% (but has since recovered somewhat since then), even though such workers are also eligible for short-time work. Around half of temporary work agency contracts had durations of less than 3 months (Bundesagentur fur Arbeit, 2009). This is a typical reaction: among OECD countries, the business cycle sensitivity of total hours worked is 2.5 times higher for temporary than for permanent workers (OECD, 2009a). Temporary work agency contracts rose sharply in the upswing, reflecting far-reaching liberalisation (OECD, 2008), but still peaked at only 3% of employment subject to social security contributions (Figure 2.5, panel A).
[FIGURE 2.5 OMITTED]
... which may adversely affect labour market outcomes over the long term
The benefits of enhancing overall firm flexibility notwithstanding, temporary work may adversely affect career progression and productivity if workers get trapped in temporary forms of employment. Countries with a large difference in regulation between regular and temporary work contracts are at risk of developing a dual labour market, as has occurred in Spain and Japan, which disadvantages those at the margin of the labour market. Apart from being socially unjust, such a development is also harmful for long-run economic development as workers on fixed-term contracts are less likely to develop skills, often because companies are less willing to invest in training short-term staff. Moreover, it is primarily the low-skilled that take up temporary employment: of all employees with atypical work contracts, more than one-third do not have vocational training and only 15% have a tertiary degree. (9) This is counterproductive in view of the fact that the relative demand for skilled labour is increasing (Chapter 5). There is also the risk that temporary workers remain stuck in precarious contracts, in particular if companies use temporary workers to circumvent strict employment protection legislation for regular work (Hohendanner and Gerner, 2010). (10) According to business surveys, German companies would prefer a lowering of regular EPL rather than a further easing of the restrictions for temporary work contracts. (11) With this background, it is unfortunate that the government envisages a further liberalisation of temporary work contracts while planning to leave regulation of regular work contracts unchanged. In particular, it is envisaged to allow temporary work contracts (after a waiting period of one year) also in those cases where the employee has worked in the company at an earlier time.
Strict EPL hinders economic adjustment
Strict EPL also hinders economic adjustment by protecting existing jobs and thereby helps to conserve existing industrial structures. In this regard, it is noteworthy that labour reallocation is fairly low in Germany. Both job reallocation (the sum of job creation and job destruction) and worker reallocation (the sum of hiring and separations) are much higher in other OECD countries (Figure 2.6). Empirical cross-country evidence tends to find a negative relationship between job reallocation and the degree of EPL (OECD, 2009a; Wolfers, 2005). (12) This may be one reason why OECD countries with stricter EPL tend to have more persistent output gaps, i.e. it takes them longer to return to potential growth rates in the aftermath of shocks (Duval et al., 2007).
[FIGURE 2.6 OMITTED]
Options to lower the degree of EPL for regular workers
Several features of the German EPL system contribute to its strictness relative to those of other OECD countries (Table 2.1):
* Notification procedures--the works council (if one exists) must approve dismissals, and if it does not, the employer has to wait for a decision by the Labour Court. Germany is the only OECD country where such a procedure exists. (13) This process adds to the delay before the notice period can start, which at 16 days for Germany is around twice as long as the OECD average of 8.5 days. (14)
* The notice period for workers with long tenure is fairly long; for example, 7 months for workers with 20 years of tenure, compared with an OECD average of 3 months.
* Compensation for unfair dismissal is high -18 months of salary for a worker with 20 years of tenure compared with an OECD average of 12 months. Many countries just give compensation equal to back pay while the court considers the case (typically 6 months). Similarly, according to the World Bank Doing Business Survey, redundancy costs are among the highest in the OECD.
Efforts to lower the strictness of EPL for regular work contracts should thus concentrate first on reducing the administrative burden of regular layoffs, notably by shortening the time period it takes to complete the necessary procedures. The notification period should be shortened, for example by reforming the requirement that the works council approves dismissals. For workers with a long tenure, the notice period should be reduced.
Reforms are also needed to lower the incentives for employees to file a lawsuit against dismissal as Germany stands out with one of the largest number of dismissal cases that are contested in court among OECD countries (with around 6 cases per 1 000 workers, against less than 2 in most other OECD countries). More than 80% of cases are resolved in conciliation, settled or otherwise withdrawn, usually in return for an additional payment (Venn, 2009). The government tried to tackle this issue in 2004 by specifying the amount of severance payments if the firm dismisses the worker for economic reasons. The aim of the legislation was to reduce the incentive of the dismissed workers to file a lawsuit against the dismissal in return for a severance payment. However, Jahn (2008) shows that this reform had no effect on the probability of receiving a severance payment for workers and thus did little to reduce the legal uncertainty that employers face. She argues that it only raised the bargaining power of workers as they may still chose to go to court knowing that the legislatively guaranteed severance payment is likely to be a lower bound for the settlement. The high average actual level of compensation payments compared with the OECD average shows that this strategy is usually successful.
In order to simplify the dismissal procedure the government should consider implementing a reform that entails replacing the court route for dismissals for economic reasons with a specified dismissal payment, making that the dismissal process more transparent and less uncertain (OECD, 2008). Such a regime would leave it up to the companies to choose whether to pay severance payment (and risk the employee challenging the decision in court) or to pay a higher dismissal payment (in effect, admitting that the dismissal is unfair) without leaving the court route open for employees (Jahn, 2005). It would truly reduce the legal uncertainty compared with the status quo. Spain introduced a similar scheme in 2002, which allows employers to effectively admit that a dismissal was unfair from the outset and pay unfair dismissal compensation at the time of dismissal. Even though the initial costs to the employer can be more than double (compared with severance pay), 75% of dismissals are now settled in this way, suggesting that employers seem to be willing to pay a higher price to avoid the cost and uncertainty of court proceedings (Venn, 2009).
Preventing the build-up of long-term unemployment
Even though employment has held up well, unemployment may still increase sharply, in particular if the upswing proves weak and firms ultimately adjust their workforce. Recent projections suggest that unemployment could rise by more than half a million in 2010 (equivalent to an increase in the unemployment rate by 1 percentage point). Labour market policy thus needs to prepare for such an outcome and the challenge will be to avoid many of the dismissed workers slipping into long-term unemployment as happened in the past (Figure 2.7). This may require rethinking the activation policies of past years. While raising incentives for the long-term unemployed was the main issue from the start of the decade, efforts now have to concentrate on the inflow of short-term unemployed which may require further adjustment of the available tools. The law for the reorientation of labour market policy instruments (Gesetz zur Neuausrichtung der arbeitsmarktpolitischen Instrumente) in late 2008 was a helpful step in this regard as it gave labour offices more freedom in adapting available tools to the specific circumstances of individual unemployed. (15)
[FIGURE 2.7 OMITTED]
Spending on labour market programmes is skewed towards passive measures In 2007, Germany spent around 2 1/2 per cent of GDP on labour market programmes, almost twice the OECD average, reflecting in part the larger number of unemployed (Figure 2.8, panel A). Compared with the average OECD country, spending per unemployed is skewed towards passive measures (Figure 2.8, panel B). Overall, benefits for unemployed are slightly higher than in other OECD countries. Net replacement rates during an unemployment spell on a five-year average (unemployment benefits, cash housing-related benefits and social assistance, taking into account different family types and two earnings levels) amount to 63% of average wages in Germany. This compares with a median value of 60% among OEGD countries (OECD, 2009a). By contrast, spending on early retirement programmes is lower thanks to substantial reform in that area (OECD, 2008).
[FIGURE 2.8 OMITTED]
While it is important to alleviate hardship through income support during and after a recession, it will be equally important, as the recovery gains steam, to encourage and assist jobseekers in finding new work and increasing their long-term employability. This is particularly so because steep recessions go hand in hand with structural change which requires adaptation of skills. This suggests an expansion of the funds available for active labour market programmes (ALMP) is warranted, even though in past recessions spending on ALMP was not increased proportionally (OECD, 2009a). Rather than increasing spending overall, however, focus should be put on those measures that have been proven successful in the past, and where a successful impact is to be expected in this particular recession. Evaluation studies suggest that effective job search assistance schemes promise high returns, notably over the short term, whereas more care must be used in employing other measures (OECD, 2009a; Card et al., 2009).
Intensifying job counselling and placement should be the priority...
The government started to address reintegration of labour by including higher spending on public employment service (PES) staffing and for training measures in the fiscal stimulus programmes. Close monitoring should be given to whether the additional funds are sufficient to maintain the case load per PES worker in view of a rising number of unemployed. A lower number of unemployed per caseworker is generally seen as being associated with better labour market outcomes (Hainmuller et al., 2009; Schiel et al., 2008). Job search assistance is important, because while jobs are being created even in recessions, matching potential candidates with suitable vacancies becomes more difficult. (16) In this context, fostering regional labour mobility is an important part of job search assistance going forward and may require intensifying the use of mobility allowances paid to jobseekers. Increased reliance on private-sector employment service providers could help to avoid bottlenecks in activation services, but service contracts need to be designed to reward good outcomes and minimise cream skimming (provider fails to enrol disadvantaged clients) and artificial manipulation of outcome measures (OECD, 2005).
... and requires institutional reforms of the Public Employment Service
Effective job counselling depends crucially on the institutional setup. Currently, the administration of job-search assistance for the long-term unemployed, unemployment benefit II (UBII) recipients, is shared in most regions between the regional offices of the Federal Employment Agency (dealing with benefit payment and job search activities, accounting for 80% of UBII costs) and municipalities (responsible mainly for housing benefits). However, the constitutional court ruled already in December 2007 against this mixed administration and required a reform by end-2010 (Box 2.2). Uncertainty over the future design of the UBII administration since then has led to loss of staff, and resources that could have been used for job counselling are instead occupied by restructuring. A rapid decision and implementation of a reform is essential in order to maintain the services provided to the currently long-term unemployed but also to deal with the likely future inflow of newly unemployed that drop out of the unemployment insurance system into the unemployment benefit II scheme.
The coalition agreement envisages to change the constitution in such a way that the procedures for the benefit recipients change as little as possible in order to ensure that the basic principle of the one-stop shop remains in place. Even though little systematic research is available to clearly favour one institutional setup among the options, preserving the prior arrangement as best as possible seems a sensible way going forward for two reasons. First, a recent evaluation study found that in those municipalities in which the Federal Employment Agency was not involved in the administration of UBII recipients (in the so-called "option municipalities"; Box 2.2), the integration of benefit recipients into the labour market and the ending of benefit dependency was working less well than in those municipalities where the administration was shared between municipalities and the Federal Employment Agency (ZEW et al., 2008). (17) This evidence thus argues against assigning the sole responsibility to municipalities and in favour of co-operation as neither the Federal Employment Agency nor the municipalities have the full expertise to cover all benefit and activation measures for the rather heterogeneous group of UBII recipients (which go beyond simple labour market measures and include social integration measures).
Box 2.2. The organisational setup of job counselling for UBII recipients The labour market reform that was enacted in 2005 (Hartz IV reform) replaced the previous social assistance and the unemployment assistance benefit scheme with the unemployment benefit II scheme (OECD, 2008). The scheme is administered jointly by the municipalities (responsible for housing benefits, one-off payments and special counselling measures) and the Federal Employment Agency (responsible for benefit payments, apart from housing, and job counselling). Three types of organisations were set up at the local level in 2005: * Jobcentres (joint associations, Arbeitsgemeinschaften or ARGEn): In 350 municipalities, the administration of UBII recipients is done jointly by the Federal Employment Agency and the municipality (within a jobcentre) using funding from both administrative layers (their number increased to 354 until September 2009). This set-up arguably comes closest to the Hartz Commission's intention to give responsibility for all jobseekers to one institution. * In 19 municipalities, both the Federal Employment Agency and the municipality offer their services separately without a joint jobcentre (getrennte Tragerschaft). Until September 2009, their number had increased to 23. * Option municipalities (Optionskommunen or zugelassene kommunale Trigger): In 69 municipalities, municipalities are offering the services themselves with the Federal Employment Agency tackling only the unemployment insurance benefit (UBI) recipients. This option was the outcome of a compromise in the second chamber of Parliament, as the government envisaged assigning responsibility for UBII policies to the Federal Employment Agency while a majority of states wanted to assign it to the municipalities. The maximum number of such option municipalities was limited to 69. While the "option municipalities" pay for UBII benefits, they are partly compensated out of the federal budget. This model was to expire in December 2010, but the coalition agreement envisages prolonging it indefinitely. In December 2007 the constitutional court ruled that jobcentres, because of their "mixed administration" (Mischverwaltung), are unconstitutional because this arrangement cannot guarantee that both institutions (municipalities and the Federal Employment Agency) take independent decisions as required by the constitution. Also, under this arrangement it was not transparent which institution had the responsibility for the fulfilment of public duties. The constitutional court required the current setup to be reformed by end-2010 at the latest. The available options after the court ruling are now to give sole responsibility to either the municipalities or the Federal Employment Agency, to legalise the mixed administration (possibly requiring a change in the constitution) or to organise the administration separately as is currently done in 23 municipalities (getrennte Tragerschaft).
Second, the current government's plan to continue the provision of integrated services follows the trend in the OECD to provide "one-stop shops" or service centres for benefit recipients and job seekers, though to different extents. Examples include the integration of benefit administration and employment services into a single agency (Work and Income) in New Zealand in 1998 and the Labour Force Service Centres (LAFOS) in Finland (OECD, 2005; Tergeist and Grubb, 2006; Duell et al., 2009). (18) However, the provision of services "under one roof" does not automatically lead to better outcomes, in particular if constitutional reasons set limits for the scope of co-operation. For example, the different institutions involved may have different objectives or misaligned incentives, leading to the risk that some actors will adopt labour market policies that are not geared toward improving employment outcomes (Immervoll, 2009). Thus, much depends on the concrete implementation of the government plans in order to reduce such risks.
Training may be more effective in this recession
When deciding on which other active labour market programmes (ALMP) to focus on, apart from job counselling, it should be noted that their effectiveness may be different in a recession than under normal circumstances. In this regard, an expansion of training programmes may be a desirable option. Germany has significantly reduced its spending on training over the past years, moving towards the lower OECD average. The reduction in spending in part reflects that the efficiency of training programmes in "normal times" is viewed controversially, notably because of lock-in effects. These concerns may be of less importance in a recession where the number of job vacancies per jobseeker declines and thus the opportunity costs of human capital investment are lower, even though the evidence for such effects is small (Lechner and Wunsch, 2009). In addition, this recession is associated with structural change, which may require more re-training of unemployed than in other downturns. In this regard, the scaling up of funds spent on vocational training and re-training in the government's stimulus programmes is a useful step. The effectiveness of training measures depends to a large degree on the type of training. For example, measures that enhance the general skills of the participant (e.g. by focussing on IT knowledge or foreign languages) tend to result more often in the take-up of a regular job than measures that improve the job application skills of the unemployed or a combination of different measures (Kopf and Wolff, 2009).
Hiring subsidies should be targeted
ALMP options also include measures on the labour demand side, such as hiring subsidies. The main advantage of hiring subsidies is that they concentrate on newly created jobs and thus are more cost-effective than "stock" subsidies (like general reductions in social security contributions). Such measures may be helpful for groups of unemployed that have problems on the labour market and whose chances to find a job are particularly bad in a recession. In Germany, the Federal Employment Agency may grant such a subsidy (Eingliederungszuschuss) for disadvantaged persons for up to 12 months and this measure is generally seen as effective in bringing these persons back to work (Ruppe, 2009). Also, employers who hire long-term unemployed with particular difficulties finding a job have received a special subsidy since October 2007. Making hiring subsidies proportional to net employment changes instead of gross hiring (marginal stock subsidies) would help to minimise the adverse effects such subsidies may otherwise have. Such adverse effects include hiring workers only for the duration of the subsidy ("churning") or the replacement of existing workers by subsidized hires ("revolving-door effect"). By imposing on employers to keep an employee for whom they received a hiring subsidy employed for one year or at least the same number of months that were previously subsidised, such effects are to some extent avoided in Germany.
Box 2.3. Recommendations regarding the labour market Short-time work scheme * Closely monitor the effects of the extension of the maximum duration of short-time work and the lowering of the costs to firms, to ensure the scheme does not inhibit necessary structural change. Avoid a further prolongation of the exceptional subsidization of the costs of short-time work beyond the originally legislated date. * Consider improving incentives under the short-time work scheme by requiring employers that join the scheme from now on to repay some of the short-term benefit paid to the employee if workers are laid off in the period that follows the end of the short-time work. Also consider to taper the replacement rate of the short-time work benefit paid to the worker over time to maintain search incentives for workers. Easing employment protection legislation * Consider shortening the period before a dismissal notice can be given, for example by reforming the requirement for the works council to approve dismissals while preserving the important consultation mechanism on the company level. * Also consider to reduce the notice period for workers with a long tenure. * In case of dismissals for economic reasons, employers should be given the right to choose between paying a severance payment (while leaving the court route open for employees), in line with current regulation, or paying a higher unfair dismissal compensation which would replace the court route. * Rethink the further liberalisation of fixed-term contracts envisaged by the coalition agreement in order to lower the risk of dualisation on the labour market. Active labour market policy * Monitor whether adequate job counselling and placement capacities are available to deal with the inflow of new unemployed. More intensive use of other active labour market instruments (training, hiring and mobility allowances) may become necessary in case of a marked deterioration of the labour market. Reform the administration of the basic income scheme for jobseekers (recipients of unemployment benefit II) quickly as required by the constitutional court's ruling. Any reform should be implemented in such a way that the procedures for the benefit recipients change as little as possible to ensure that the basic principle of the one-stop shop remains in place.
Bach, H.-U. et al. (2009), "Die Krise wird deutliche Spuren hinterlassen", IAB Kurzbericht, No. 20.
Bach, H.-U. and E. Spitznagel (2009), "Kurzarbeit--Betriebe zahlen mit--und haben was davon", IAB Kurzbericht, No. 17.
Bauer, T.K., S. Bender and H. Bonin (2007), "Dismissal protection and worker flows in small establishments", Economica, Vol. 74, pp. 804-821.
Boss, A. et al. (2009), "Deutsche Konjunktur im Herbst 2009", Kieler Diskussionsbeitrag 468/469, Institut fur Weltwirtschaft, Kiel.
Bundesagentur fur Arbeit (2009), Zeitarbeit--Aktuelle Entwicklungen, August.
Card, D., J. Kluve and A. Weber (2009), "Active Labour Market Policy Evaluations: A Meta-Analysis", IZA Discussion Papers, No. 4002.
DIHK (2008), Beschaftigungshfurden abbauen, Jobs sichern.
Duell, N., D. Grubb and S. Singh (2009), "Activation policies in Finland", OECD Social, Employment and Migration Working Papers, No. 98.
Duval, R., J. Elmeskov and L. Vogel (2007), "Structural Policies and Economic Resilience to Shocks", OECD Economics Department Working Papers, No. 567.
Grubb, D., J.-K. Lee and P. Tergeist (2007), "Addressing labour market duality in Korea", OECD Social, Employment and Migration Working Papers, No. 61.
Hainmuller, J. et al. (2009), "Do more placement officers lead to lower unemployment? Evidence from Germany", IAB Discussion Papers, No. 13.
Heckmann, M., A. Kettner and M. Rebien (2009), "Offene Stellen im IV. Quartal 2008: Einbruch in der Industrie--Soziale Berufe legen zu", IAB Kurzbericht, No. 11.
Hohendanner, C. and H.-D. Gerner (2010), "Die Ubernahme befristet Beschaftigter im Kontext betrieblicher Personalpolitik", Soziale Welt, forthcoming.
Immervoll, H. (2009), "Minimum-income benefits in OECD countries: Policy design, effectiveness and challenges", OECD Social, Employment and Migration Working Papers, No. 100.
Jahn, E. (2005), "Wie wirkt der Kundigungsschutz?", Zeitschrift fur Arbeitsmarktforschung (ZAF), No. 2/3, pp. 284-304.
Jahn, E. (2008), "Do firms obey the law when they fire workers?--Social criteria and severance payments in Germany", University of Erlangen Discussion Papers, No. 22.
Koch, S. (2001), "Flexible Arbeitszeiten fordern Beschaftigung", IAB Kurzbericht, No. 17.
Kopf, E. and J. Wolff (2009), "Die Wirkung von Trainingsmassnahmen fur ALG-II-Bezieher: Auf den Inhalt kommt es an", IAB Kurzbericht, No. 23.
Lechner, M. and C. Wunsch (2009), "Are training programs more effective when unemployment is high?", Journal of Labor Economics, Vol. 4, No. 27, pp. 653-692.
OECD (2004), OECD Employment Outlook, OECD, Paris.
OECD (2005), OECD Employment Outlook, OECD, Paris.
OECD (2008), OECD Economic Surveys: Germany, OECD, Paris.
OECD (2009a), OECD Employment Outlook, OECD, Paris.
OECD (2009b), OECD Economic Outlook 86, OECD, Paris.
Rothe, T. (2009), "Arbeitsmarktentwicklung im Konjunkturverlauf--Nicht zuletzt eine Frage der Einstellungen", IAB Kurzbericht, No. 13.
Ruppe, K. (2009), "Eingliederungszuschusse und Betriebszugehorigkeitsdauer in Westdeutschland", IAB Discussion Papers, No. 6.
Schiel, S. et al. (2008), "Offentliche Arbeitsvermittlung: Mehr Personal--mehr Zeit--mehr Vermittlungen", IAB Kurzbericht, No. 21.
Tergeist, P. and D. Grubb (2006), "Activation strategies and the performance of employment services in Germany, the Netherlands and the United Kingdom", OECD Social, Employment and Migration Working Papers, No. 42.
Venn, D. (2009), "Legislation, Collective Bargaining and Enforcement: Updating the OECD Employment Protection Indicators", OECD Social, Employment and Migration Working Papers, No. 89.
Wolfers, J. (2005), Measuring the Effects of Employment Protection on Job Flows: Evidence from Seasonal Cycles, Wharton School of Business, January.
ZEW, Institut Arbeit und Qualifikation and TNS Emnid (2008), Evaluation der Experirnentierklausel nach [section]6c SGB II--Vergleichende Evaluation des arbeitsrnarktpolitischen Erfolgs der Modelle der Aufgabenwahrnehrnung ?Zugelassene kommnunaler Trager und ?Arbeitsgemeinschaft, Bundesministerium fur Arbeit und Soziales, Forschungsbericht 387, Mannheim, Gelsenkirchen and Bielefeld.
(1.) In addition, firms have accepted a sharp fall in employees' hourly labour productivity, which is very unusual when looking at historical developments. Hourly labour productivity had never hitherto fallen in recessions since 1970, when data became available (Bach et al., 2009).
(2.) The prototype of a working time arrangement to protect employment was the 4-day working week of Volkswagen in 1994 (Koch, 2001).
(3.) The sum of the contributions of the different factors to the decrease in working time adds up to more than 100% due to other factors that increased working time during 2009, such as a fall in sickness absence.
(4.) Over the year 2009 as a whole, on average 1.06 million employees worked short-time due to economic reasons.
(5.) In June 2009, the share of short-time workers in all employees amounted to 19.1% in the car industry, the second-highest share after manufacturing of basic metals and fabricated metal products (itself a major input to the car sector) with a share of 26.2%. The average share in manufacturing was 16.9% (Bundesagentur fur Arbeit, Inanspruchnahme von Kurzarbeitergeld, October 2009).
(6.) German companies get full relief of the social security contributions they pay on the part of the employees' salary that is lost due to short-time work in case the employee is taking part in a training measure (by contrast, if the employee is not in training, full relief of the social security contributions is only provided after 6 months). However, it should be noted that this only applies if the training measure is not related to acquiring company-specific knowledge.
(7.) The EPL indicator was almost unchanged between 2003 and 2008 as a relaxation for temporary contracts (increase in the allowable duration of fixed-term contracts for employers launching a new business or hiring older unemployed) was offset by stricter regulation of regular work contracts (introduction of a legislated severance pay for dismissals).
(8.) Currently, enterprises with 10 or fewer employees (covering about 18% of total employment) are exempt from regular employment protection (while employees are still provided special protection against discriminatory and arbitrary dismissal and employers must not give notice without a minimum of social consideration). Firms with 20 or less employees are exempt from requirements for collective dismissal. Accordingly, the EPL indicator for these firms is one of the lowest among OECD countries (Venn, 2009). Unsurprisingly, firms with very few employees tend to view EPL as less of a problem than larger firms (DIHK, 2008). However, adjusting the overall EPL indicator for the small firm exemption has only a minor impact on the ranking of countries according to strictness of EPL (Box 4 in Venn, 2009).
(9.) See Destatis, Niedrigeinkommen und Erwerbstatigkeit, Press Briefing 19 August 2009.
(10.) In 2007, 48% of all workers on temporary work contracts obtained regular contracts in the company after the end of the limitation period. The share differs across sectors with two-thirds of manufacturing firms taking over workers that previously were on temporary contracts and only around one-third of companies in the social service sector doing so. Chances for temporary workers of obtaining a regular work contract are lower in larger companies that face stricter EPL and companies with a large churning rate (labour fluctuations without a change in employment level). By contrast, chances of obtaining a regular work contract are higher in companies with recruitment problems (Hohendanner and Gerner, 2010).
(11.) See DIHK, Flexibilitat am Arbeitsmarkt--weiter ganz oben auf der Agenda; Ergebnisse einer DIHK-Unternehmensbefragung, Herbst 2009.
(12.) However, looking at small firms in Germany that are exempted from EPL, Bauer et al. (2007) do not find that job worker turnover is significantly higher.
(13.) Only Austria maintains a similarly prominent role of the works council in the dismissal procedure. There, the works council needs to be notified of a planned dismissal and has 5 days to react to it. In case the council objects to the dismissal, it can appeal against it before the labour court by request of the dismissed worker.
(14.) The longer notification period in the case of Germany reflects both the time it takes for the works council to react (around one week) and the regulation that the notice applies from the 1st or 15th of the month (thus adding on average 8 days to the delay before the notice period can start).
(15.) The law replaced 17 specifically regulated labour market policy instruments with the two broader instruments "placement budget" (Vermittlungsbudget, replacing 9 individually regulated instruments such as mobility support) and "measures for activation and integration" (Massnahmen zur Aktivierung und berufllichen Eingliederung, replacing 8 individually regulated instruments) which allows for a more flexible use of private third-party service providers. Based on the law, the labour office introduced in late 2009 a 4-phase model for the integration of unemployed.
(16.) During the downturn from 2001 to 2005 on average 1 784 000 recruitments (including apprenticeships) happened each quarter, of which 690 000 were filled by the unemployed. This compares with around 1 845 000 recruitments per quarter during the boom period between 2006 and mid-2008 (Rothe, 2009).
(17.) The third existing type of organisation where both institutions offer their services separately without creating a joint jobcentre (getrennte Tragerschaft) was not part of the evaluation due to the limited dataset.
(18.) At the same time, a number of OECD countries moved towards administration of benefit recipients and job seekers at the municipal level such as the Netherlands and Denmark.
Table 2.1. Components of the OECD employment protection legislation indicator Score on a scale from 0 Regulation of ... (least restrictions) to 6 (most restrictions) OECD average Germany Regular contracts Notification 2.8 5 procedures Delay involved 1.1 2 before notice can start Length of notice 2.6 3 period at 9 months of tenure Length of notice 2.6 2 period at 4 years of tenure Length of notice 1.6 4 period at 20 years of tenure Severance pay at 9 0.8 1 months of tenure Severance pay at 4 1.5 2 years of tenure Severance pay at 20 1.3 2 years of tenure Definition of 1.4 4 justified or unfair dismissal Length of trial 4.2 3 period Compensation 2.1 3 following unfair dismissal Possibility of 2.6 3 reinstatement following unfair dismissal Maximum time to make 2.4 1 a claim of unfair dismissal Fixed term Valid cases for use 1.6 0 contracts of fixed-term contracts Maximum number of 2.4 2 successive fixed-term contracts Maximum cumulated 1.1 1 duration of successive fixed-term contracts Temporary work Types of work for 1.4 1.5 agencies which temporary work agency employment is legal Restrictions on 2.9 4 number of renewals of temporary work agency contracts Maximum cumulated 1.5 0 duration of successive temporary work agency contracts Authorisation and 3.0 6 reporting requirements for temporary work agencies Regulations 3.7 6 requiring equal treatment of regular and agency workers Collective Definition of 4.2 6 dismissals collective dismissal Additional 3.8 3 notification requirements for collective dismissals Additional delays 1.8 3 involved before notice can start for collective dismissals Other special costs 1.7 3 to employers of collective dismissals Rank of Germany Regulation of ... among OECD countries Regular contracts Notification 30 procedures Delay involved 19 before notice can start Length of notice 13 period at 9 months of tenure Length of notice 7 period at 4 years of tenure Length of notice 28 period at 20 years of tenure Severance pay at 9 24 months of tenure Severance pay at 4 19 years of tenure Severance pay at 20 22 years of tenure Definition of 24 justified or unfair dismissal Length of trial 4 period Compensation 18 following unfair dismissal Possibility of 18 reinstatement following unfair dismissal Maximum time to make 2 a claim of unfair dismissal Fixed term Valid cases for use 1 contracts of fixed-term contracts Maximum number of 12 successive fixed-term contracts Maximum cumulated 13 duration of successive fixed-term contracts Temporary work Types of work for 17 agencies which temporary work agency employment is legal Restrictions on 17 number of renewals of temporary work agency contracts Maximum cumulated 1 duration of successive temporary work agency contracts Authorisation and 18 reporting requirements for temporary work agencies Regulations 14 requiring equal treatment of regular and agency workers Collective Definition of 24 dismissals collective dismissal Additional 6 notification requirements for collective dismissals Additional delays 20 involved before notice can start for collective dismissals Other special costs 16 to employers of collective dismissals Note: The OECD average is unweighted. A higher rank in the last column denotes a stricter protection. For more complete description and details on the scoring methodology, see www.oecd.org/employment/protection.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Chapter 2|
|Publication:||OECD Economic Surveys - Germany|
|Date:||Mar 1, 2010|
|Previous Article:||Emerging from the crisis.|
|Next Article:||After the crisis: bringing public finances back to a sustainable path.|