All abstracts and locations will be shared via our mailing list. If you want to attend a specific session or receive the mailing list in the future, write a short mail to aantezza@diw.de and dbursy@diw.de. Note that the dates and speakers are subject to change.
Despite growing demographic pressure, a reform in Germany has lowered the full retirement age (FRA) temporarily for early starters. In this paper we analyze whether the reform was well targeted. Specifically, we study if the early retirement scheme benefits workers with high job strain or if workers with better jobs and the potential to continue employment at older ages are the main beneficiaries of this pathway. For the empirical analysis we use high quality administrative data from the Federal Employment Agency covering the complete employment biographies of individuals and including information about the job strain at the individual level. We focus on men in West Germany as for them information on child-related pension entitlements which is incomplete in the data is usually unimportant and the information on their employment history is included in the dataset also for periods before German reunification. We exploit policy variation induced by the pension reform using a Difference in Differences Design. The reform initially reduced average labor market exit age by about nine months corresponding to around 4.5 months for each year the FRA was lowered. The impact decreased considerably when the FRA was raised again gradually. The effect is stronger for individuals working in demanding occupations. The result applies to the average job strain over the full career as well as to the average job strain in the last years before retirement.
Siblings are the ultimate peers, deeply shaping one another’s development. Do these influences vary with a family’s cultural background? I estimate how sibling spillovers differ for girls and boys with older brothers or sisters in migrant and native families, using a regression discontinuity design on high-quality administrative data. Exploiting exogenous variation in older siblings’ achievement from their school entry age, I show that girls in migrant families with high-achieving older brothers experience significantly lower educational outcomes. This effect does not appear for native girls or for migrant and native boys. I propose a simple theoretical framework to explain these results, highlighting gender bias in parental preferences as a key factor. Moreover, the effects are more pronounced in migrant families with traditional backgrounds and are also reflected in mothers' labor supply decisions when sons, rather than daughters, enter school late.
We study the impact of group composition on the creativity of team output. To do so, we exploit the randomized allocation of first-year undergraduate students to study groups in a large Economics module over multiple years. Students are required to produce a short video on an important invention/innovation and its economic impacts. Using a variety of creativity measures rated by multiple raters, some of them with a major in film studies, we analyze how the group composition in terms of gender and background affects the quality of the video output. Furthermore, we look at who appears in the videos and in which function to study the representation of group members. Qualitative evidence from focus groups with participating students provides evidence on mechanisms.
Latin America is facing rapid population aging. To address the sustainability of contributory pension systems, many countries are considering increasing the retirement age. One such case is Brazil, where a pension reform in 2019 raised the retirement age from 60 to 62 for women. While the effects of such reforms on the labor force participation of older adults and their spouses have been widely analyzed, intergenerational spillover effects remain understudied, particularly in Latin America. Evidence from other contexts suggests that a mother’s retirement can increase her young daughters’ labor supply, particularly among households with young children. However, most of this evidence comes from European countries with broader access to formal sources of childcare. In this paper, I estimate the effect of grandmother’s retirement on the labor force participation of young adults in Brazil. Using data from Brazil’s Continuous National Household Sample Survey (PNADC) from 2023 and a fuzzy regression discontinuity design, I find that older women substantially reduce their working hours upon retirement. Moreover, adult women living with young children and an older woman who retires upon reaching the minimum retirement age increase their working hours by 0.5 for every hour the older woman stops working. No significant effects are observed for men, even when young children are present in the household. These findings suggest that, in contexts with limited access to formal childcare, grandmothers play a crucial role in supporting their daughters’ labor supply, particularly when young children are involved.
We show that inventors' personal experiences of natural disasters lead to increased green innovation through changes in their higher-order beliefs about consumer preferences. We match patent records of French and German inventors and a survey of inventive firms to detailed information on natural disasters. This allows us to exploit exogenous variation in inventors’ exposure to natural disasters. In affected areas, exposure results in an average 8.2% increase in green patents, driven primarily by increased innovation in mitigation technologies that reduce emissions. Non-green innovation remains unaffected. We do not find significant spatial spillovers in patenting, underlining the importance of being personally affected. To get at the mechanism, we investigate how natural disaster exposure changes firms' reasons for green innovation. Affected firms adjust their higher-order beliefs regarding consumers’ demand for green goods and more stringent environmental regulation. Our findings provide novel causal evidence that personal experiences affect higher-order beliefs.
This paper investigates the mental health consequences of spousal death and the role of survivor benefits in mitigating these effects. Using Dutch administrative data and a staggered difference-in-differences, we first document a significant increase in the consumption of benzodiazepines - a proxy for deteriorating mental health - following widowhood (up to +3 percentage points, or +30%, in the year after the spouse's death). We then exploit a 1996 policy reform that sharply restricted eligibility for public survivor insurance to identify the causal impact of benefits on mental health. Our findings show that the restriction of survivor benefits led to an additional 6% increase in benzodiazepine use, suggesting that survivor insurance provides important protection against the mental health risks associated with widowhood.
Economic conditions at time of labor market entry have been shown to have large negative effects on labor market outcomes for an extended period of time. The immediate effects have been shown to be worse for lower educated entrants. In the long run, the effects may be very different as low and high educated have different possibilities to accommodate this negative shock, high educated entrants can downgrade and low educated entrants cannot. This paper estimates the long-term effects of economic conditions at labor market entry on income and employment for high and low educated separately up to 40 years after graduation. I exploit regional variation in unemployment and proxy for time and location of labor market entry using birth region and nominal duration of education. For both high and low educated labor market entrants between 1971 and 1988, I find short-lived negative effects immediately after graduation. These negative effects on earnings and employment remain for high educated in the long run. Low educated benefit as they end up in more stable jobs. This can be explained by low educated entrants shifting to less cyclical sectors, which shields them from future recessions, as a result of high educated entrants downgrading.
As Western populations age, more adult children face parental health challenges. Caregiving responsibilities towards parents are becoming crucial concerns for families, healthcare systems, and societies. While this care is essential for elderly well-being, it also demands significant time, energy and emotional resources. Yet, the impact of care provision on caregivers’ lives remains understudied. This study examines how unexpected parental health crises affect young adults’ decisions to form partnerships and have children. Using Dutch administrative data, we track individuals aged 15-35 whose parents suffered non-fatal health events, that generated a long-lasting care need. Our analysis aims to explore mechanisms related to time allocation and mental health impacts, potentially creating cycles where delayed fertility leads to older parenthood and earlier caregiving duties in the next generations.
This project examines the French academic pipeline to identify when and why women become underrepresented in faculty positions: I investigate whether they are less likely to apply, less likely to succeed or a combination of both. Using comprehensive administrative data combining Thèses.fr doctoral records and administrative data from the Conseil National des Universités (CNU), I analyze gender gaps at each critical transition: from PhD graduation through qualification to securing permanent positions. I then exploit the exogenous variation in the number of job openings per year to examine whether tighter competition boosts or blocks women's access to permanent positions.
Does increased legal infrastructure empower victims to leave abusive relationships? Structural barriers often prevent victims of intimate partner violence from seeking help and leaving their abuser, with two-thirds of female victims in Europe neither reporting incidents to the police nor accessing support services. I study the introduction of the 2002 Violence Protection Act in Germany, which established a structured cooperation framework between women’s shelters, courts, and police. Using newly digitized data on shelter openings from 1970 until today, I classify counties by whether they had support services in place at the time of the reform and combine this with administrative divorce records to examine how the effectiveness of legal protection varies with existing infrastructure. My results show that divorce filings by women increase across all counties following the reform, with more persistent changes in areas lacking shelter infrastructure at the time of the legal change. Contested divorces and those involving children rise only in counties with pre-existing services, suggesting that support infrastructure plays a key role in enabling more complex or constrained exits - even under improved legal protections. These findings align with predictions from a dynamic bargaining model of household decision-making under violence, institutional support, and legal constraints.
In times of an increasing scarcity of workers, Germany has in a rare move decreased the retirement age for a sizeable section of the working force. This paper investigates the effect of this negative labor supply shock on firm and individual level outcomes using matched employer-employee data from the SIAB. We show that affected workers make use of the reform and exit the labor force earlier. The intensity of the shock varies across firms, with some experiencing substantial losses in their workforce, while others remain largely unaffected. Our data allows us to investigate effects on the careers of coworkers as well as firm-level outcomes.
In this paper, we use SOEP-ADIAB data to investigate the differential wage returns to non-cognitive skills, and explore how these returns depend on the occupational task content. The dataset enables us to extract three dimensions of non-cognitive skills capturing sociability, diligence and emotional stability. We demonstrate how individuals sort into occupations with high social-skill requirements based on their skill profiles. In our main analysis, we estimate a model of tenure-wage growth. Preliminary results indicate that individuals experience stronger wage growth with tenure in occupations where social skills are more important.
This study examines the impact of Statutory Retirement Age (SRA) reforms on individual behavior and welfare in the presence of policy uncertainty and misinformation. We develop a structural life-cycle model in which individuals are uncertain about the future evolution of the SRA and misinformed about its importance. We derive individuals’ expectations and information on the SRA from self-elicited belief data using the German Socio-Economic Panel Innovation Sample (SOEP-IS). The model accounts for key life-cycle savings and old-age labor supply determinants, such as human capital accumulation, involuntary job loss, health status, and family dynamics. We estimate the model using decision data from the core sample of the SOEP. We design counterfactual simulations to assess the effects of SRA reforms and illustrate the trade-offs of several implementation strategies.
The newspaper coverage of CEOs is highly gendered with more family-related language used in newspaper articles on female than male company leaders. In a randomized online experiment, we ask whether this stereotypical representation affects readers' beliefs about CEO competence, firm performance, and resulting financial decision-making. We show participants articles consisting of elements from real newspaper coverage on a real company and its CEO, and vary the way information about their family is (not) presented. We find that although respondents do not believe firm performance to be different by CEO coverage, investments into the firm stock are significantly lower for female CEOs when their family is neutrally mentioned. Instead, there is no penalty for highlighting a trade-off between family and career. Machine-learning analysis based on free-text questions suggests that family and gender play an important role in respondents' reasoning.
Disruptions of labor market trajectories have lasting effects on later economic success. Displacement due to forced labor conscription is a disruption that remains understudied despite its continued prevalence in contemporary contexts. I investigate the consequences of exposure to forced labor conscription for individuals' long-term labor market outcomes. I exploit the fact that cohorts of Dutch civilians faced a differential probability of temporary labor coercion in Nazi Germany during WWII in a Regression Discontinuity Design. Using Dutch census data from 1971, I find that conscripted individuals have lower education, income, and probability of employment. Analyzing heterogeneous effects, I find that exposure to harsher conditions in Germany is associated with reduced labor force participation and poorer health. My findings suggest that the negative impact on labor force participation is mitigated when individuals are conscripted to work in sectors that are also present in the Netherlands, which enhances their ability to reintegrate into the workforce.
An influential body of economic literature has consistently shown strong associations between cognitive abilities and economic preferences. However, much of this research has overlooked the potential endogeneity of cognitive abilities—an oversight given that these abilities stem from a series of investment decisions influenced by factors likely correlated with economic preferences. To address this gap, we utilize a novel dataset that includes both stated and revealed preferences, along with genetic data on respondents. We measure inherent predispositions to high cognitive abilities using a polygenic score for cognitive performance (PGI CP), providing a robust proxy for individuals’ innate cognitive potential. Our comprehensive data further allow us to mitigate concerns that individuals with a genetic disposition for high cognitive performance are also raised in more favorable environments.
This paper evaluates the impact of immigration on the incidence of severe health shocks at the workplace level. Using rich linked employer-employee data from Germany and an instrumental variable leveraging policy variation, I show that firms with a higher concentration of foreign workers experience lower rates of long-term sickness among their employees. Decomposing the rates by gender and worker origin reveals that these health improvements are concentrated entirely among native men, are particularly pronounced in medium-sized construction and trade companies, and are driven by the firm's foreign new hires. The primary mechanism is a lower relative participation of native male workers in manual tasks at the firm, particularly in industries with high health risks. Further analysis confirms that these effects are not due to the substitution of less healthy natives, the evolution of norms around absenteeism, differences in firm expansion trends, or reverse causality.
This paper studies and compares the effect of different adverse life events -- job loss, disability and health shocks, divorce and spousal death -- on individuals' income trajectories. We use a harmonized design across events in terms of methodology and data: matching difference-in-difference with exhaustive Dutch administrative registers. We assess the effect of adverse events on different margins. We compare their effect on primary and disposable household income in order to measure the public insurance to the shocks provided by the tax and transfer system. Both between different events and within different groups for a given event, we find that the importance of government insurance increases with the severity of the shock on primary income. However, we find that certain groups of the population are relatively less protected against adverse life events, such as young people facing a large health shock or secondary earners facing a divorce.
Aging societies put a strain on social security systems worldwide. Raising the statutory retirement age (SRA) is one of the most common tools that policymakers employ to respond to this pressure. We study the effect of policy reform on savings, labor supply, and welfare. Using data from the German Socio-Economic Panel (SOEP), we estimate a structural life-cycle model. The model features subjective policy uncertainty and is enriched with subjective survey-elicited (SOEP-IS) policy expectations data. It include various monetary and non-monetary drivers of life-cycle savings and old-age labor supply. We can use the model to quantify the effects not just of further increases in the SRA, but also of policies targeted at removing uncertainty by committing to certain policy trajectories.
Work-limiting disabilities pose a significant risk to the earnings potential and welfare of older workers. While coverage of public disability insurance (DI) systems is almost universal, the risk of becoming dependent on DI varies across occupations. In this paper, I study the value of public DI across different occupations using data from administrative social security records in Germany. I evaluate how public DI coverage affects occupational transitions by analyzing the employment effects of a large-scale policy reform that eliminated occupational DI coverage in Germany. I then use a structural model to study the relationship between disability insurance, occupational choice, and retirement in a life-cycle context. In counterfactual experiments, I evaluate how adjustments to the DI and retirement system affect the labor supply of different occupational groups.
Health investments are vital for maintaining physical and mental well-being throughout working life, and their importance is amplified with rising retirement ages due to demographic aging. This is the first study to examine if a longer working life causally increases institutionalized health investments. We explore the impact of a German pension reform that raised the retirement age by three years and extended the working life. Utilizing detailed administrative data from the German pension insurance on workrelated rehabilitation measures, we apply a sharp regression discontinuity design based on birth cohort cut-offs introduced by the reform. We find no evidence of increased health investments for enhancing work capacity. This null effect holds across various specifications and persists for specific diagnoses and measures. These findings challenge the predictions of the standard human capital model, which suggests that work-related health investments should rise as the working life extends in aging societies.
Does increased legal infrastructure lead to higher intimate partner violence (IPV) reporting and empower victims to leave abusive relationships? Structural barriers often prevent IPV victims from seeking help, with two-thirds of female victims in Europe neither reporting incidents to the police nor accessing support services. I study the rollout of women’s shelters and the introduction of the 2002 Violence Protection Act in Germany, which established a structured cooperation framework between shelters, courts, and police. I leverage variation in the rollout of shelters and the introduction of the reform to identify its effects, using newly digitized data on shelter opening dates combined with administrative county-level data.