This dissertation studies how staffing decisions of firms shape labor market prospects of men, women and ethnic minorities and how governments can influence this be- havior in order to prevent socially undesirable outcomes. Chapter 1 introduces a new approach to investigate hiring discrimination. Drawing on fine-grained click data from Job-Room, the employment website of the Swiss public employment service, it analyses how recruiters search and select job candidates on the platform. The results show that, depending on the country of origin, ethnic minorities face 4–19% lower contact rates than otherwise identical natives. Consistent with theories of implicit bias, these ethnic penalties are larger during the hours just before noon and towards the end of the workday, when recruiters spend less time evaluating each CV. The chapter also finds that women face 7% lower contact rates in male-dominated professions while men face a similar penalty in female-dominated professions. This new methodology provides a widely applicable, non-intrusive, and cost-efficient tool to monitor hiring discrimination, adjudicate between different theories of discrimination, and inform approaches to counter it. Chapter 2 delves deeper into the topic of ethnic discrimination. Applying the methodology outlined in chapter 1, it shows that linguistic and cultural assimilation of ethnic minorities on Job-Room, e.g. the acquisition of German skills, does not significantly mitigate ethnic discrimination. The same holds for human capital in general. In line with theories of monopsonistic competition we find that discrimination decreases with labor market tightness. The evidence for attention discrimination, which postulates that recruiters allocate attention asymmetrically depending on the ethnicity of jobseekers, is mixed. While we find only limited evidence that natives and ethnic minorities differ with respect to unobservable skills—an assumption underlying models of statistical discrimination—the pattern of ethnic discrimination on the platform is more consistent with taste based discrimination or theories of ethnic hierarchies. Chapter 2 provides several pieces of evidence that discrimination measured on Job-Room prolongs unemployment of minority jobseekers. By drawing on the recruiter click data from Job-Room the third chapter explores the potential of naturalization to combat ethnic discrimination. It finds that obtaining a Swiss citizenship reduces the extent of hiring discrimination on the platform by around 75%. The positive effect of naturalization on the contact rate can be explained to a large extent by a reduction in statistical discrimination, i.e. by a decrease of the uncertainty with respect to the productivity of ethnic minorities. Chapter 4 investigates whether recruiters on Job-Room prefer full- or part-time workers. The chapter exploits that jobseekers have to state whether they search a full- or a part-time job when registering on Job-Room. The results show that recruiters on Job-Room have strong preferences for candidates searching for a full- time position over observationally equivalent candidates searching for a part-time job. This part-time penalty is much more pronounced for men than for women. The most likely explanation is that recruiters punish men who do not act according to gender roles. As a consequence, men face larger disincentives to reduce their working hours than women, which might prevent a more equal division of labor within the household. The fifth chapter investigates one of the most popular policies to avoid excessive layoffs during recessions. Short-time-work (STW) programs encourage firms to reduce the number of hours worked per employee instead of the number of workers by providing income support to the affected employees. The chapter assesses whether the Swiss STW scheme helped to prevent unemployment during and in the aftermath of the the Great Recession. It draws on an establishment-level panel dataset linking administrative records on firms that applied for short-time work with the unemployment register and other administrative data sources. Applying an event study Difference-in-Differences approach it finds that STW substantially reduced dismissals into unemployment. It also shows that the associated savings in unemployment benefit payments may have been large enough to compensate the total spending on STW.
Can governmental policies mitigate the effects of recessions on unemployment? We study whether the Swiss short-time work (STW) program reduced unemployment in the 2009–2015 period using quarterly establishment-level panel data linking several administrative data sources. We compare changes in permanent layoffs into unemployment, hiring from unemployment, establishment survival and size between establishments that applied successfully to establishments that applied unsuccessfully for STW at cantonal employment agencies. The latter appear to be a valid control group for the former among others because of substantial idiosyncrasies in cantonal approval practices. We find strong evidence that STW increases establishment survival and prevents rather than postpones dismissals: the 7,880 establishments treated in 2009 would have dismissed approximately 20,500 workers into unemployment (0.46% of the labor force) until 2012. Most workers would have been dismissed in the quarters immediately following application and more than a third would have become long-term unemployed. We estimate that the savings in terms of unemployment benefit payments may have been large enough to compensate the spending on STW benefits for the Swiss unemployment insurance.
Can governmental policies mitigate the effects of recessions on unemployment? We study whether the Swiss short-time work (STW) program reduced unemployment in the 2009-2015 period using quarterly establishment-level panel data linking several administrative data sources. We compare changes in permanent layoffs into unemployment, hiring from unemployment, establishment survival and size between establishments that applied successfully to establishments that applied unsuccessfully for STW at cantonal employment agencies. The latter appear to be a valid control group for the former among others because of substantial idiosyncrasies in cantonal approval practices. We find strong evidence that STW increases establishment survival and prevents rather than postpones dismissals: the 7,880 establishments treated in 2009 would have dismissed approximately 20,500 workers into unemployment (0.46% of the labor force) until 2012. Most workers would have been dismissed in the quarters immediately following application and more than a third would have become long-term unemployed. We estimate that the savings in terms of unemployment benefit payments may have been large enough to compensate the spending on STW benefits for the Swiss unemployment insurance.
Diese Studie präsentiert eine umfassende Auslegeordnung, wie sich die Covid-19-Pandemie entlang der Einkommensverteilung in der Schweiz ausgewirkt hat. Grundlage der Analyse sind sechs Befragungswellen des sotomo/SRF-Bevölkerungsmonitorings, welche die Situation der Haushalte in der Schweiz seit Beginn der Pandemie systematisch beschreiben. Die Analyse berücksichtigt dabei den Einfluss der Krise auf das Einkommen, die Ausgaben, Ersparnisse, das persönliche Wohlbefinden und die Gesundheit der Haushalte sowie den Zusammenhang zwischen Pandemie, Haushaltseinkommen und dem Vertrauen in die Politik und die Medien. Generell zeigt sich, dass die Pandemie bestehende Ungleichheiten verschärft. Haushalte am unteren Ende der Einkommensverteilung sind in den meisten Dimensionen - teils deutlich - stärker von der Krise betroffen als reichere Haushalte. So vermelden Personen, die zu einem Haushalt mit sehr tiefem Haushaltseinkommen von unter 4 000 Franken gehören, im Schnitt einen sehr starken Einkommensrückgang von 20 Prozent. Bei Personen aus Haushalten mit einem Monatseinkommen von mehr als 16 000 Franken sind die Einkommen um 8 Prozent gesunken. Die Einkommenssituation von Befragten mit einem Haushaltseinkommen unter 4 000 Franken hat sich zudem auch nach Ende des Lockdowns im Sommer 2020 kaum verbessert. Im Gegensatz dazu sind die Einkommensrückgänge der anderen Einkommenskategorien in den zwei späteren Befragungswellen (Oktober 2020 und Januar 2021) geringer als in den beiden Erhebungen davor, die inmitten bzw. am Ende des ersten Lockdowns stattfanden. Die Einkommensveränderung infolge der Pandemie hängt stark damit zusammen, wie sich die Erwerbssituation veränderte. Bei Befragten, die weiterhin einer Erwerbstätigkeit nachgingen, ist das Haushaltseinkommen im Schnitt wenig oder gar nicht zurückgegangen. In Haushalten mit Personen, in denen der Umfrageteilnehmer arbeitslos wurde, kam es zu grossen Einkommensverlusten. Bedeutende Einkommensrückgänge verzeichneten auch viele Selbständige. Unter Befragten, die von Kurzarbeit betroffen waren, waren die Einkommensrückgange zwar etwas geringer, mit durchschnittlich rund 20 Prozent aber trotzdem gross. Betrachtet man die Entwicklung entlang der Einkommensverteilung zeigt sich erstens, dass Personen aus einkommensschwachen Haushalten häufiger von einer nachteiligen Entwicklung der Erwerbsituation betroffen waren. Rund ein Drittel der Befragten aus Haushalten mit Einkommen unter 4 000 Franken, die vor der Krise einer Erwerbsarbeit nachgingen, wurden im Verlauf der Krise arbeitslos oder mussten Kurzarbeitsgeld beziehen. Bei der obersten Einkommensklasse (Haushaltseinkommen über 16 000 Franken) war es ein Sechstel der Befragten. Zweitens zeigt sich, dass die Einkommenseinbussen in allen Erwerbssituationen für Personen aus Haushalten am untersten Ende der Einkommensverteilung jeweils am grössten waren. Besonders gross waren die Einkommensrückgänge unter den Befragten aus Haushalten in der untersten Einkommensklasse, die arbeitslos wurden (-50 Prozent). ; Finanziert durch Bundesamt für Gesundheit BAG, Enterprise 4 Society Center und KOF Konjunkturforschungsstelle. Datenzugang von Forschungsstelle sotomo
This article investigates the prevalence of traumatization and mental distress in a sample of 1055 male European long-term prisoners as part of a wider study of the living conditions of prisoners serving sentences of at least five years in Belgium, Croatia, Denmark, England, Finland, France, Germany, Lithuania, Poland, Spain and Sweden. Data were collected in a written survey using the Posttraumatic Diagnostic Scale (PDS), the Brief Symptom Inventory (BSI) as well as questions on attempted suicide and auto-aggressive behaviour. Participants experienced a mean of three traumatic events, with 14 per cent developing a Posttraumatic Stress Disorder (PTSD) subsequently. In each national sample, more than 50 per cent of the participants were in need of treatment because of psychological symptoms and nearly one-third had attempted suicide.