Hybrid Archive - Page 8 of 14 - IAB - Institut für Arbeitsmarkt- und Berufsforschung

The Social Multiplier of Pension Reform

This study is about the influence of family members, neighbors and coworkers on retirement behavior.

We study the influence of family members, neighbors and coworkers on retirement behavior. To estimate causal retirement spillovers between individuals, we exploit a pension reform in the Netherlands that creates exogenous variation in peers' retirement ages, and we use administrative data on the full Dutch population.

We find large spillovers in couples, primarily due to women reacting to their husband's retirement choices. Consistent with homophily in social interactions, the influence of the average sibling, neighbor and coworker is modest, but sizable spillovers emerge between similar individuals in these groups.

Additional evidence suggests both leisure complementarities and the transmission of social norms as mechanisms behind retirement spillovers. Our findings imply that pension reforms have a large social multiplier, amplifying their overall impact on retirement behavior by 40%.

The lasting impacts of middle school principals

The principal effects on the probability of engagement in the criminal justice system are much larger for Black than for non-Black males.

Using rich Texas administrative data, we estimate the impact of middle school principals on post-secondary schooling, employment, and criminal justice outcomes. The results highlight the importance of school leadership, though striking differences emerge in the relative importance of different skill dimensions to different outcomes. The estimates reveal large and highly significant effects of principal value-added to cognitive skills on the productive activities of schooling and work but much weaker effects of value-added to noncognitive skills on these outcomes.

In contrast, there is little or no evidence that middle school principals affect the probability a male is arrested and has a guilty disposition by raising cognitive skills but strong evidence that they affect these outcomes through their impacts on noncognitive skills, especially those related to the probability of an out-of-school suspension. In addition, the principal effects on the probability of engagement in the criminal justice system are much larger for Black than for non-Black males, corresponding to race differences in engagement with the criminal justice system.

Earning while learning: how to run batched bandit experiments

Stressing strict privacy policies and changing the location of the survey URL have no response-enhancing effect.

Researchers collect data in most experiments not all at once but sequentially over a period of time. This allows to observe outcomes early and to adapt the treatment assignment to reduce the costs of inferior treatments. This talk discusses multi-armed-bandit-type adaptive experimental designs and algorithms for balancing exploration of treatment effects and exploitation of better treatments. By design, bandits break usual asymptotic and make inference difficult. We show how a batched bandit design allows for valid confidence intervals and compare coverage of the batched bandit estimator in Monte Carlo simulations. In a real-world application, we investigate elements of a survey invitation message targeted to businesses. We implement a full factorial experiment with five elements adaptively.

Our results indicate that personalizing the message, emphasizing the authority of the sender, and pleading for help increase survey starting rates, while stressing strict privacy policies and changing the location of the survey URL have no response-enhancing effect. As a tool for researchers, we introduce bandits in Stata, which facilitates running Monte Carlo simulations to assist the design and implementation of experiments before data are collected, interactively running own bandit experiments, and analyzing adaptively collected data. Bandits implement three popular treatment assignment algorithms: ε-first, ε-greedy, and Thompson sampling. Bandits facilitates estimation, inference, and visualization.

The impact of unions on nonunion wage setting: threats and bargaining

The results indicate an important role played by union wage spillovers in lowering wages over the 1980-2010 period.

In this paper we provide new estimates of the impact of unions on nonunion wage setting. We allow the presence of unions to affect nonunion wages both through the typically discussed channel of nonunion firms emulating union wages in order to fend off the threat of unionisation and through a bargaining channel in which nonunion workers use the presence of union jobs as part of their outside option.
We specify these channels in a search and bargaining model that includes union formation and, in our most complete model, the possibility of nonunion firm responses to the threat of unionisation.

Our results indicate an important role played by union wage spillovers in lowering wages over the 1980-2010 period. We find de-unionisation can account for 38% of the decline in the mean hourly wage between 1980 and 2010, with two-thirds of that effect being due to spillovers. Both the traditional threat and bargaining channels are operational, with the bargaining channel being more important.

Patterns and Paradoxes of Migrant and Ethnic Inequalities in the UK

This lecture examines the patterns and paradoxes in observed educational and labour market attainment of migrants and minorities.

The inequalities faced by immigrants and ethnic minorities are topics of substantial salience across Europe, reflected in an every-growing body of research. Despite the insights shed by this burgeoning literature there remain a number of outstanding debates and puzzles about what leads to better or worse outcomes and the underlying mechanisms.

The UK is an interesting case for illuminating some of these debates and puzzles, due to: the diversity of its immigrant and ethnic minority population, the richness of research base, and certain counter-intuitive findings that are at odds with theoretical expectations as well as evidence from other European countries. It thus has the potential to shed further light on what drives more or less unequal outcomes in different contexts.

Drawing on a range of research from across the last 20-years, in this lecture I examine the patterns and paradoxes in observed educational and labour market attainment of migrants and minorities, with reference to the role of class background, educational aspirations, neighbourhoods and social networks, cohort change and return migration, discrimination and policy to. I explore the gendered differences in such outcomes, and what that implies for our understanding of wider national ‘gender orders’. I reflect on what this body of work can tell us about the factors that shape economic outcomes in different settings, the need for greater attention to both migrant success as well as migrant disadvantage; and I assess the key outstanding questions and implications for future research.

Job Search Autonomy

A typical reaction in unemployment insurance (UI) is to impose requirements (backed by sanctions) on the quantity of job search. We evaluate the job seeker’s reaction.

A typical reaction to the moral hazard problem in unemployment insurance (UI) is to impose requirements (backed by sanctions) on the quantity of job search, aimed at ensuring sufficient levels of effort. However, is this the most effective policy strategy? It does neither take into account the intrinsic motivation of the job seeker nor the quality of the targeted search. An alternative policy strategy that encompasses such goals is to focus on job search autonomy.

Exploiting a policy change in a region in Switzerland which followed this aim, we evaluate the job seeker’s reaction to being granted more autonomy. Using rich procedural register data, we document the effects on quantitative effort, scope of search and ultimately on unemployment duration and earnings in the found job.

Our results show that the policy change increased the average duration of unemployment spells in the area by about 8%, while increasing average re-employment earnings by about 3%. Results are heterogenous, a main driver of the variety is the interplay of effort delivery and local labour market conditions, notably tightness. This finding highlights the relevance of search externalities. Furthermore, we provide some evidence of labour demand effects.

Joint work: Patrick Arni, Amelie Schiprowski

Short-time Work and Unemployment: Long-term effects on labor market outcomes

This study sheds light on the impact of different types of job retention programs such as short-time work.

This study sheds light on the impact of different types of job retention programs such as short-time work (STW).

We analyze the causal effect of an episode of STW on labor market outcomes up to five years later and compare this to the effects of sudden unemployment episodes. Using data from German Socio-Economic Panel (1992–2022), we employ an event-study approach to analyze the effect of unemployment and STW on career trajectories and unpaid care work.

Results show that workers with periods of short-time work have higher employment and wage stability than workers with periods of unemployment. There are no gender differences in the effects of STW on employment and hours worked.

Taking a Chance on Workers: Evidence on the Effects and Mechanisms of Subsidized Employment from an RCT

This paper estimates experimental impacts of a supported work program on employment, earnings, benefit receipt, and other outcomes over a four-year follow-up period.

Case managers addressed employment barriers and provided targeted financial assistance while participants were eligible for 30 weeks of fully subsidized employment. Program access increased employment rates by 21 percent and earnings by 16 percent while participants were receiving services. Though gains attenuated after services stopped, treatment group members experienced lasting improvements in employment stability, job quality, and well-being, and we estimate the program's marginal value of public funds to be in line with other adult workforce programs. Post-program impacts are concentrated among participants who were hired by their host-site employer post-program, suggesting that encouraging employer learning about potential match quality is a key mechanism underlying the program’s impact, and additional descriptive evidence supports this interpretation.

Machine learning methods provide no evidence of treatment effect heterogeneity in a broad sample of job seekers using a rich set of baseline characteristics from a detailed application survey. We conclude that subsidized employment programs with a focus on creating permanent job matches can be beneficial to a wide variety of unemployed workers in the low-wage labor market.

Testing and interpreting the effectiveness of causal machine learning – an economic theory approach

This study documents how the application of causal machine learning methods can successfully increase sales revenue.

If treatment effects vary systematically, customizing treatment status at the individual level may help increasing the overall effectiveness of an intervention. We document how the application of causal machine learning methods can successfully increase sales revenue generated within a loss framing treatment in an online field experiment. We combine this data with a behavioral experiment measuring sensitivity to loss aversion at the individual level. Our results show that treatment status as assigned by causal machine learning is consistent with treatment assignment based on economic theory.

Joint: Kevin Bauer, Andreas Grunewald, Florian Hett, Johanna Jagow, Maximilian Speicher

Measuring and Predicting New Work in the US: the role of local factors and global trends

"New work" - we develop an algorithm that identifies new job-titles in the US economy based on their vector distance from the closest existing job title in the previous census.

"New work", namely the introduction of types of jobs that did not exist earlier, is an essential part of innovation and employment growth for advanced economies. Using text analysis, we develop an algorithm that identifies new job-titles in the US economy based on their vector distance from the closest existing job title in the previous census. We use this method to generate a measure of "new work" from 1980 to 2010 in each of 354 occupations and we construct its distribution across 766 commuting zones. We first show how this measure of "new work" is associated to task and skill characteristics of workers in the occupations and to employment growth, skill bias and innovation in the commuting zones. Then we analyze whether local population density, human capital and manufacturing intensity in the 1980, and/or local exposure to structural "shocks" in the 1980-2010, relating to trade competition, technological change, immigration and age changes predict the creation of new work.

Our main findings are that the share of college educated and the density of population in 1980 are the strongest predictors of New Work creation in the 1980-2010 period. The aging of population and exposure to computer adoption were also associated to New Work creation, while robot adoption was negatively associated to it. The exposure to immigration and trade had a more nuanced and differentiated correlation to new work.