جستجوی فرصت های شغلی مورد نیاز برای افراد بیکار مسن : انتقال به استخدام، بازنشستگی زود هنگام و مزایای از کار افتادگی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|3842||2013||27 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : European Economic Review, Volume 58, February 2013, Pages 31–57
We use a recent policy change in the Netherlands to study how changes in search requirements for the older unemployed affect their transition rates to employment, early retirement and sickness/disability benefits. The reform, becoming effective on January 1 2004, requires the elderly to formally report their job search efforts to the employment office in order to avoid a (temporary) cut in benefits. Before the new law was passed, unemployed individuals were allowed to stop all search activity at the moment they turned 57.5. Estimating various duration models using difference-in-difference and regression discontinuity approaches, we find that for several groups of individuals who are affected by the policy change, the stricter search requirements significantly increases their entry rate into employment. However, we also find evidence of a higher outflow to sickness/disability insurance schemes, a presumably unwanted side-effect of the policy change.
How can policy raise the labour market participation rate of older workers? The combination of a low participation rate of older workers and population aging puts a strain on public finances. Raising the official retirement age to 67 and beyond has been brought forward as part of the solution in many countries. However, increasing the retirement age will only be effective when the opportunity to retire early is limited. Not only early retirement pension payments, but also other social benefits such as disability insurance (DI) and unemployment insurance (UI) have been used as early retirement pathways. These benefit schemes have some features that make them especially attractive for older workers. First, the potential benefit duration is usually increasing in age. Second, older workers who receive unemployment benefits often do not need to report any job search effort to the unemployment office in order to retain their benefit payments. Little is known about the effect of such policies on the inflow into unemployment and the subsequent outflow to employment. The longer potential benefit duration and exemption of search requirements might very well contribute to the fact that the duration of unemployment for older workers is longer on average (Chan and Stevens, 2001 and De Graaf-Zijl and Hop, 2007). Intentionally or unintentionally, the policy to exempt older unemployed from job search can create an “unemployment tunnel” in which individuals receive benefit payments until they reach the official retirement age. In 2006, there were many countries in which older unemployed were not obliged to look for jobs, including Belgium, Finland, France, Germany and the United Kingdom. In other countries such as Sweden, there is no formal exemption from job search for the older unemployed. However, job search requirements are not strictly enforced on them either (OECD, 2006). In recent years many governments have raised the age from which older unemployed are exempted from job search. Some countries even equalised job search requirements for all unemployed, young or old. In Belgium, the minimum age was increased from 50 to 58 between 2002 and 2004 (Bollens, 2011). In France older unemployed are no longer exempted from the requirement to look for jobs as from January 2012. The Netherlands also used to exempt their older unemployed (defined as being at least 57.5 years old) from the requirement to actively search for a job in order to receive full UI benefits. This situation changed on January 1st 2004. From that date, the 57.5+ year olds faced the same regulations as other age groups and needed to report their (formal) search behaviour to the unemployment office. This paper exploits this policy change to examine how search requirements for the older unemployed affect the number of individuals who experience a transition to employment. Because a reduction in unemployment only raises labour force participation when early retirement benefits and disability benefits are not used as substitutes to UI, we also examine the effect of the policy change on transitions to disability payments and early retirement. To study labour market transitions of older workers, access to a dataset with a large cross-sectional dimension is needed. Labour market surveys are usually based on a representative sample of the entire working age population such that only a tiny fraction of older unemployed individuals is observed, precluding meaningful analyses of transition behaviour for this group. The present paper contributes to the existing literature by using a large administrative database covering all registered benefit and wage receipts in the Netherlands, including UI benefits. The data provides very precise information on income and labour market status, giving us a large enough sample to analyse labour market transitions of the older unemployed in the years 2001–2005. There exists a large strand of literature examining the effects of changes in the UI benefit system on unemployment duration. Most of these papers are concerned with effects of sanctions or training programs (Abbring et al., 2005, Van den Berg et al., 2004 and Van den Berg and van der Klaauw, 2006), changes in potential unemployment benefit duration (Caliendo et al., 2009, Card and Levine, 2000, Kyyrä and Ollikainen, 2008, Lalive et al., 2006, Lalive, 2008 and Van Ours and Vodopivec, 2006) or the level of unemployment benefits (Carling et al., 2001 and Røed and Zhang, 2003). Studies examining a tightening of search requirements are less common. Manning (2009) finds large flows out of claimant status upon a tightening of search requirements in the UK but does not find an effect on search intensity. Petrongolo (2009), studying the same UK reform, concludes that unemployment duration decreases, but the transition rate to disability insurance benefits increases. Our paper is closest in spirit to Heyma and van Ours (2005), who examine the effect of the same discontinuity in UI eligibility criteria for Dutch older workers. They find a substantially lower outflow to jobs for individuals that turn 57.5 and are no longer required to actively search for a job. In contrast to Heyma and van Ours (2005), our dataset follows individuals both before and after the policy change. Making use not only of variation in age, but also of variation over time, we are able to estimate treatment effects for various groups of treated individuals. Moreover, instead of focussing exclusively on unemployment to employment transitions, we also shed light on substitution between various social insurance programs by considering both DI receipt and early retirement as competing risks for the exit out of unemployment. One of the desired consequences of imposing stricter requirements for receiving UI benefits is to save on government spending by decreasing the number of individuals eligible for receipt of these benefits. This can be done directly by excluding individuals from receiving UI benefits if they do not comply with the new rules, or by making the receipt of UI benefits so unattractive that individuals start to look for alternatives themselves. However, the alternative that the government has in mind (paid employment) might not be the most attractive alternative from the point of view of the individual. The unemployed worker can instead substitute towards other benefit types, such as disability benefits or, in case of the older workers, early retirement benefits (provided that eligibility conditions for such schemes can be met). Since costly substitution between programs that insure different risks should be avoided, spill-over effects among these government programs are an important aspect of policy evaluation. In this paper, we are able to furnish empirical evidence on the importance of the various substituting pathways. We estimate flexible form competing risks duration models using difference-in-difference and regression discontinuity approaches and identify the causal effect of the policy change on observed labour market transitions. We show that for several groups of individuals, the stricter search requirements strongly and significantly increase the number of individuals that find a job after a maximum of 2 years in unemployment. For instance, unemployed males aged 57.5–59.5 have a more than 20% (6.5 percentage points) higher probability of finding a job within 24 months (37.8% instead of 31.3%) due to imposition of the new rules. Previous results for the Netherlands show that working individuals substitute between UI and DI benefits (Heyma and van Ours, 2005, Euwals et al., 2011, Koning and van Vuuren, 2007, Koning and van Vuuren, 2010 and de and Kapteyn, 2011). We find evidence that substitutability of UI and DI is also relevant for unemployed individuals. A higher number of unemployed individuals receive disability benefits when search requirements are enacted, a presumably unwanted side-effect of the policy change. In contrast, no significant substitution from UI benefits towards early retirement benefits is found. The remainder of the paper is set up as follows. Section 2 presents some important aspects of the Dutch UI system, with a focus on changes in the system aimed at the older unemployed. It also gives a brief description of the sickness/disability insurance benefit and early retirement systems and changes therein. Theoretical effects of an increase in search requirements are considered in Section 3. The empirical analysis starts out in Section 4 with a description of the data and the selection of treatment and control groups for analysis. Section 5 presents the estimation strategy and some descriptive evidence, before continuing to estimation results given in 6 and 7. Section 8 reports on post-unemployment job characteristics. Finally, Section 9 concludes.
نتیجه گیری انگلیسی
As the challenges associated with an ageing population become more prominent and many countries increase statutory retirement age, it becomes all the more important to document the effects of labour market policies on behaviour of older workers. This paper deals with one such policy change which has recently been implemented in various countries, namely the imposition of job search requirements for older unemployed. Using a large administrative database covering all wage and benefit payments to Dutch individuals in the years 1999–2005, we show the effects of a tightening of search requirements which is specifically aimed at older workers. A policy reform, coming into effect on the 1st of January 2004, makes an end to the special treatment of older unemployed individuals. Before the law was initiated, an unemployed individual aged 57.5 did not need to report his/her job search effort to the unemployment office. The new policy required older unemployed to continue actively searching for a job even after turning 57.5. Using difference-in-difference and regression discontinuity techniques within a flexibly specified duration framework, we estimate the effect of this reform on outflow to jobs, early retirement and disability benefits for the various affected groups of individuals aged 55.5 to 59.5. The main finding is that within 24 months after the start of an unemployment spell, there is a 6 (11) percentage points increase in the number of male (female) individuals that find a job. However, this strong effect on labour market participation is accompanied by a 4 (9) percentage point increase in the number of male (female) individuals who use DI benefits as an alternative exit route. In light of the evidence that UI benefits and DI benefits are alternative pathways to early retirement, one would expect outflow to retirement to increase as well. However, changes in the UI and in the early retirement system cannot be completely separated, and therefore we cannot provide evidence for an increased outflow to early retirement. These conclusions are remarkably robust: using various sources of variation generated by the policy change we show strong effects of search requirements for the older unemployed on their chances to find jobs, and their probability to receive DI payments. Moreover, the same results are found in many types of alternative regression specifications. The finding that unemployed individuals substitute between receipt of UI benefits and DI benefits can be expected to hold in many countries. The exact magnitude of both the job finding effect and the substitution effect likely differs according to institutional settings such as the way in which stricter rules are enforced, the height of penalties, and the attractiveness of other social schemes. An interesting venue for further research is to estimate the contribution of these parameters to the gross effect presented in this paper. Another suggestion for further research is to estimate the impact of search conditions on after-unemployment job characteristics. Theory predicts that an increase in search costs will decrease reservation wages and might therefore lock the older workers into low-skilled and low-paid jobs, another unwanted side-effect of a policy that is meant to induce older workers to become more active and productive labour market participants.