Displaced workers: causes, characteristics, and policy options
This project revolves around displaced workers. It has three parts. In the first part we characterize the selection into displacement: Who gets laid off and why? We revisit the classic questions of wage rigidity and selection into unemployed in previously unexplored ways.
The second part of the project investigates the effects of mandated advance lay-off notice, a policy that exists in almost all OECD countries. Despite its wide-spread use, no previous study has credibly estimated the effects of this policy. We ask two main questions: What s the value of the insurance provided by advance notice for workers? What s the size of the costs imposed on firms?
The third part of the project estimates the long-run impact of Public Employment Service (PES) caseworkers for the subsequent labor market outcomes of the unemployed. It also examines whether caseworkers are more important for some unemployed individuals. The analysis addresses the following policy question: Is it possible to improve the employment prospects by reallocating unemployed jobseekers onto caseworkers?
The project combines up-to-date empirical approaches and the exceptional quality of Swedish administrative data to estimate relationships that are also relevant from a normative point of view.
Final report
Background: Summary and aim of the project
This project has had three interrelated parts. The first part revolves around wage setting and selection into unemployment. The second part investigates the effects of mandated advance lay-off notice, a policy that exists in all OECD countries. The third part estimates the long-run impact of Public Employment Service (PES) caseworkers for the subsequent labor market outcomes of the unemployed.
1. Development of the project
Within the project we have produced six studies intended for international journal publication. There are three papers on wage-setting and selection into displacement; see papers 1, 5, and 6 in the publication list. An additional three papers relate to policies targeting displaced workers (or unemployed workers in general); see papers 2-4.
Two of the papers have been published in top field journals (papers 1-2); one paper has been resubmitted (for possible publication in a top-5 journal); another is currently under revision (for a top field journal); and the final two papers are in the initial submission process.
2. Important results
The project includes a paper on Mandatory Notice of Layoff (MN) – a common, but severely understudied, policy; see paper 3. The paper breaks a lot of new ground. We show that there may be an efficiency case for MN tied to the fact that the firm is forced to provide advance information about impending layoff. Such advance information encourages workers to search while still employed, which may allow them to find higher-quality jobs upon reemployment. A crucial part of this efficiency argument is that firms can replace inefficient MN by paying “severance” to compensate the worker. We also show empirically that firms engage in such side-payments: workers eligible for longer notice (because they just surpassed an age threshold) are paid more severance. Moreover, we find that workers who are eligible for longer notice are less exposed to unemployment and obtain higher wages when they are re-employed. The paper establishes that a key reason for these beneficial effects is that it is more efficient to search for a new job during employment compared with unemployment. Finally, the paper compares the benefits for workers to the production losses imposed on firms. In our setting, the benefits are greater than the losses.
In a paper on caseworker efficiency (see paper 4), we show that being (randomly) matched to a better caseworker shortens unemployment duration. If the quality of the caseworker improves by a standard deviation, the probability of leaving unemployment within 90 days increases by 11% (or 0.1 of a standard deviation). Little of this effect is driven by demographics or education of the caseworker (although female caseworkers perform better). But better-paid PES-workers generate more favorable outcomes. From a policy point of view, the most interesting result is that caseworkers who frequently meet with their clients tend to produce better outcomes.
Moreover, a paper on displacement effects (see paper 6) shows that the typical layoff event does not involve as severe earnings losses as suggested by the so-called mass-layoff literature. In contrast to the previous literature, the paper studies the average laid-off worker (note that only 7% of laid-off workers experience mass lay-off events). It shows that the earnings gap between displaced and non-displaced workers fully closes 7 years after the displacement event. This is very different from the estimates based on mass lay-off events, which in contrast suggest that earnings losses are permanent.
3. Contribution to the research frontier
Within the project, we provide the first causal evidence on the effects of MN for workers. We also show evidence suggesting that firms and workers agree on severance pay packages, which potentially limit the effects of inefficient MN. As far as we know, this is the first paper documenting empirically that agents use side-payments in order to circumvent legal regulations. Moreover, we provide causal evidence on the higher effectiveness of job search while employed versus unemployed. The paper further offers novel evidence on the production loss of the policy, and discusses the optimal duration of the mandate from a production efficiency point of view.
The paper on caseworker efficiency makes a unique contribution to the literature in the sense that the analysis is based on a setting where job seekers are randomly allocated onto caseworkers. Other papers in this literature have relied on so-called conditional independence assumptions. We can thus provide credible answers to the questions of interest. The unique information in the data imply that we can ask questions which have not been possible hitherto.
The paper on the long-run effect of being displaced brings a completely new empirical strategy to the literature (the previous literature has almost exclusively been based on mass layoff events). Using seniority rules, it is able to the estimate the effects of displacement for the average laid-off worker
4. New research questions
The work within this project has led to a broader Forte-funded program on structural change (Structural change: consequences and policy responses). The program is a concerted effort to, on the one hand, understand the consequences of structural change for wage inequality, job loss, and occupational mobility; and, on the other hand, evaluate policies according to their effectiveness in facilitating workers' adjustment to structural change.
5. Dissemination
Our main communication channel is publication in international journals and in discussion paper series which are disseminated internationally.
With regard to communication outside the research community, we have written a policy report on structural change (see 7). The work on case-worker efficiency attracted a lot of attention in connection with the publication of the IFAU Working Paper (e.g., Sveriges Radio and Aftonbladet). The work on displacement effects has been covered by the Wall Street Journal.
Our research results are posted of the on our own web-pages; see Peter Fredriksson’s web-page (https://sites.google.com/site/peterfredrikssonuu/) or Jonas Cederlöf’s web-page (http://jonascederlof.com)
This project has had three interrelated parts. The first part revolves around wage setting and selection into unemployment. The second part investigates the effects of mandated advance lay-off notice, a policy that exists in all OECD countries. The third part estimates the long-run impact of Public Employment Service (PES) caseworkers for the subsequent labor market outcomes of the unemployed.
1. Development of the project
Within the project we have produced six studies intended for international journal publication. There are three papers on wage-setting and selection into displacement; see papers 1, 5, and 6 in the publication list. An additional three papers relate to policies targeting displaced workers (or unemployed workers in general); see papers 2-4.
Two of the papers have been published in top field journals (papers 1-2); one paper has been resubmitted (for possible publication in a top-5 journal); another is currently under revision (for a top field journal); and the final two papers are in the initial submission process.
2. Important results
The project includes a paper on Mandatory Notice of Layoff (MN) – a common, but severely understudied, policy; see paper 3. The paper breaks a lot of new ground. We show that there may be an efficiency case for MN tied to the fact that the firm is forced to provide advance information about impending layoff. Such advance information encourages workers to search while still employed, which may allow them to find higher-quality jobs upon reemployment. A crucial part of this efficiency argument is that firms can replace inefficient MN by paying “severance” to compensate the worker. We also show empirically that firms engage in such side-payments: workers eligible for longer notice (because they just surpassed an age threshold) are paid more severance. Moreover, we find that workers who are eligible for longer notice are less exposed to unemployment and obtain higher wages when they are re-employed. The paper establishes that a key reason for these beneficial effects is that it is more efficient to search for a new job during employment compared with unemployment. Finally, the paper compares the benefits for workers to the production losses imposed on firms. In our setting, the benefits are greater than the losses.
In a paper on caseworker efficiency (see paper 4), we show that being (randomly) matched to a better caseworker shortens unemployment duration. If the quality of the caseworker improves by a standard deviation, the probability of leaving unemployment within 90 days increases by 11% (or 0.1 of a standard deviation). Little of this effect is driven by demographics or education of the caseworker (although female caseworkers perform better). But better-paid PES-workers generate more favorable outcomes. From a policy point of view, the most interesting result is that caseworkers who frequently meet with their clients tend to produce better outcomes.
Moreover, a paper on displacement effects (see paper 6) shows that the typical layoff event does not involve as severe earnings losses as suggested by the so-called mass-layoff literature. In contrast to the previous literature, the paper studies the average laid-off worker (note that only 7% of laid-off workers experience mass lay-off events). It shows that the earnings gap between displaced and non-displaced workers fully closes 7 years after the displacement event. This is very different from the estimates based on mass lay-off events, which in contrast suggest that earnings losses are permanent.
3. Contribution to the research frontier
Within the project, we provide the first causal evidence on the effects of MN for workers. We also show evidence suggesting that firms and workers agree on severance pay packages, which potentially limit the effects of inefficient MN. As far as we know, this is the first paper documenting empirically that agents use side-payments in order to circumvent legal regulations. Moreover, we provide causal evidence on the higher effectiveness of job search while employed versus unemployed. The paper further offers novel evidence on the production loss of the policy, and discusses the optimal duration of the mandate from a production efficiency point of view.
The paper on caseworker efficiency makes a unique contribution to the literature in the sense that the analysis is based on a setting where job seekers are randomly allocated onto caseworkers. Other papers in this literature have relied on so-called conditional independence assumptions. We can thus provide credible answers to the questions of interest. The unique information in the data imply that we can ask questions which have not been possible hitherto.
The paper on the long-run effect of being displaced brings a completely new empirical strategy to the literature (the previous literature has almost exclusively been based on mass layoff events). Using seniority rules, it is able to the estimate the effects of displacement for the average laid-off worker
4. New research questions
The work within this project has led to a broader Forte-funded program on structural change (Structural change: consequences and policy responses). The program is a concerted effort to, on the one hand, understand the consequences of structural change for wage inequality, job loss, and occupational mobility; and, on the other hand, evaluate policies according to their effectiveness in facilitating workers' adjustment to structural change.
5. Dissemination
Our main communication channel is publication in international journals and in discussion paper series which are disseminated internationally.
With regard to communication outside the research community, we have written a policy report on structural change (see 7). The work on case-worker efficiency attracted a lot of attention in connection with the publication of the IFAU Working Paper (e.g., Sveriges Radio and Aftonbladet). The work on displacement effects has been covered by the Wall Street Journal.
Our research results are posted of the on our own web-pages; see Peter Fredriksson’s web-page (https://sites.google.com/site/peterfredrikssonuu/) or Jonas Cederlöf’s web-page (http://jonascederlof.com)