Note: All abstracts are listed in the sequence of the presenter (in bold). To find one, either search for name or title of the paper. Included are only contributed papers, not keynote speakers or job market papers.

A – NAME

Salamatu Nanna Adam (CERGE-EI). (73)
Statistics and Stories: Experimental Evidence on HIV Testing in Ghana

Understanding what drives HIV testing behavior is essential for designing effective communication strategies that promote testing uptake.   In this study, I use a randomized experiment to examine whether and how the format of information affects HIV testing behavior among university students in Ghana.  Providing factual information on HIV incidence and the availability of nearby testing services increased actual testing rates by about 1 percentage point from a near-zero baseline. In contrast, adding a story about the testing experience to this statistical information did not generate any additional effect. Financial incentives, introduced non-experimentally, raised testing rates to 11 percent. Interestingly, the impact of the original information treatments diminished once incentives were available.  The analysis of belief outcomes indicates that the information treatment primarily worked by improving awareness of local testing services and correcting misperceptions about peer testing behavior, rather than by heightening perceived risk. However, stories did not enhance the treatment effect on beliefs or information recall beyond the impact of simple statistical facts. These results suggest that factual information can effectively address informational barriers to HIV testing in this context, while narrative elements offer no measurable added benefit for influencing this high stakes health behavior.

Damilola Afolabi (University of Regina) (141)
Is there a Motherhood Bonus for Immigrant Mothers in the Canadian Labor Market compared to Immigrant Non-Mothers?
Research on the labor market outcomes of immigrants has disproportionately focused on male immigrants. In this research, I examine the motherhood penalty and how it affects immigrant mothers compared to immigrant non-mothers in Canada using Statistics Canada’s 2016 and 2021 Census data. I examine how the motherhood penalty differs for immigrant mothers across immigration categories (economic immigrants, family immigrants, and refugees), age, and education levels. The OLS regression analysis indicates that a motherhood penalty does not appear to hold for immigrant mothers when compared to immigrant non-mothers. Between 2015 and 2020, immigrant mothers with children aged 0 to 5, earned 13 percent and 6 percent more than immigrant non-mothers, respectively, after the controls were added, and the results are significant. However, when we take into account factors such as immigration status, children’s age group, and the mother’s education levels, it is evident that immigrant mothers who are family immigrants with infants experience a significant income decline due to the motherhood penalty compared to family immigrant women who are not mothers. In addition, the penalty disappears over time as the children become school-aged, and immigrant mothers with some education fare better overall.

Thank you very much for accepting my submission titled “Nowcasting Monthly Macroeconomic Variables in India Using Google Trends: A Machine Learning Approach” for presentation at the Global GLO-JOPE Conference 2025. I am honoured by this opportunity to contribute to the conference.

Presenter with affiliation:

Shadab Ali (Jamia Millia Islamia), Asheref Illiyan (Professor, Jamia Millia Islamia). Nowcasting Monthly Macroeconomic Variables in India Using Google Trends: A Machine Learning Approach

Abstract:

Timely assessment of macroeconomic conditions is particularly challenging in emerging economies like India, where official statistics are released with significant delays and frequent revisions. This study examines whether high-frequency digital trace data from Google Trends can enhance the nowcasting of four key Indian macroeconomic indicators: unemployment, consumer price index (CPI), real GDP, and GDP growth. Adopting the two-step approach of Woloszko (2020), we combine Google Trends with machine learning models to evaluate their predictive content. Results show that predictive performance is indicator-specific: unemployment is best captured by Lasso regression, CPI by decision trees, and real GDP and GDP growth by gradient boosting. Stock-like variables (real GDP and unemployment) exhibit strong alignment with official series, while flow measures such as GDP growth and CPI are more difficult to track, reflecting their inherent volatility. SHAP analysis further reveals that search intensity proxies for meaningful economic drivers, ranging from trade activity to consumer sentiment, and the framework successfully captures turning points during episodes such as the 2008 global financial crisis and the COVID-19 pandemic. To the best of our knowledge, this is the first systematic application of Google Trends with machine learning to nowcast multiple macroeconomic indicators for India. The findings highlight the value of search-based behavioral data as a complementary, real-time source for economic monitoring, with important implications for policymakers and forecasters in data-scarce, rapidly evolving environments.

Deepthi Sara Anil (Indian Institute of Technology Kanpur)
Dietary Diversity and Aspirations of Young Adults. Evidence from India

This paper examines whether dietary diversity among young adults in India shapes their aspirations. Using longitudinal data on young adults, I exploit an instrumental variable framework leveraging exogenous rainfall shocks to causally estimate the effect of dietary diversity on educational and occupational aspirations. The results report that diverse diets significantly raise general life optimism, work ambitions, and perceived future wellbeing. I explain my findings through suggestive channels such as improved cognition, psychosocial reinforcement, and gendered intra-household allocation. 

B – NAME

Cynthia Bansak (St. Lawrence University), Stephen Drinkwater (Roehampton University) “Differences in Schooling Attendance and Modes of Instruction Amongst Ukrainian Refugee Children: A Cross National Study”                          No. 28, online   cbansak@stlawu.edu

 Abstract: We use cross-national multi-wave survey data on Ukrainian refugees’ intentions and perspectives administered by the UNHCR to analyse the schooling decisions of Ukrainian refugees who moved to other European countries following the Russian invasion which began in February 2022. We focus in particular on two questions: whether Ukrainian refugee children attend school (or pre-school) and if so, what is the type of schooling attended (formal schooling in host country, online/remote learning in Ukrainian curriculum or both). There are significant variations along both of these dimensions by recipient country and socio-economic characteristics of the children, their parents, and their households. The latter include educations levels and language proficiency of the head of household in the host country. Differences are also found according to how refugees are registered (e.g. refugee, visa holder, or not registered) and future return migration plans. We find there are varied approaches and challenges countries face in providing education to Ukrainian refugee children.

Andrea Berlanda (Università di Padova), Elisabetta Lodigiani, Lorenzo Rocco (84)
Immigration and Adult Children’s Care for Elderly Parents: Evidence from Western Europe

In this paper, we use the Survey of Health, Ageing, and Retirement in Europe (SHARE), complemented with register data on the share of the foreign population in the European regions, to examine the effects of migration on the level of informal care provided by children to their senior parents. Our main results show that migration decreases informal care among daughters with a university degree, while it increases the provision of informal care among daughters with low-to-medium levels of education. Viceversa, migration has practically no effect on sons’ care provision who remain little involved in care activities. These results depend on the combination of two supply effects. First, migration increases the supply of domestic and personal services, making formal care more affordable and available. Second, as immigrants compete with low-to-medium-educated native workers, while improve the labor market opportunities of the better educated, the supply of informal care can increase among the less educated daughters and decrease among the more educated.

Kulshreshtha, Shobhit & Bhattacharya, Leena & Ayyagari, Padmaja

Later Sunset, Better Health? (RePEc:zbw:glodps:1648)

Abstract: Previous research, focusing primarily on high income countries, has linked later sunsets to sleep deficits and worse health outcomes. These results might not generalize to low- and middle- income countries, which have different socioeconomic, cultural, and environmental conditions. Using data from the 2015-16 and 2019-21 waves of India’s Demographic and Health Surveys (DHS) and exploiting within-district variation in annual average sunset times, we estimate the causal impact of later sunsets on the long-term health outcomes of individuals. We find that later sunsets lead to a lower prevalence of anemia, diabetes, and thyroid disorders and an improvement in the overall health index. To explore mechanisms, we analyze variation in time allocation due to a later sunset time using the 2019 Time Use Survey. We find that individuals experiencing later sunsets sleep better and exercise more, but do not change their sedentary leisure activities. Additionally, they consume healthier food and increase labor supply. These lifestyle changes may explain the health improvements associated with delayed sunsets in India.

Keywords: health; sunset time; time use; lifestyle; fixed effects

<https://ideas.repec.org/p/zbw/glodps/1648.html>

Niels-Hugo Blunch (Washington and Lee University) (112-J)Stairway to Heaven? Human Capital and Religion in Ghana

Addressing several methodological issues of previous related research this paper finds a strong relationship between individual religious affiliation and human capital of children and adults in Ghana. Allowing religious affiliation to be endogenous, the findings suggest that previous studies may have systematically underestimated the strength of the religion-human capital link.  Further testing for the relative importance of supply and demand factors in the differential in human capital across religious affiliation mostly supports the main hypothesis of tastes for human capital related to one’s religious affiliation being an important driving force of the observed human capital differentials across religious affiliation.

Alessio Buonomo (University of Naples “Federico II”), Stefania Capecchi, Francesca Di Iorio, Salvatore Strozza. (144)
Does cultural identity influence the probability of employment during economic crises?

This paper examines the relationships between cultural identity and immigrants’ likelihood of finding employment during economic crises. Drawing on Italian data from the Social Condition and Integration of Foreign Citizens (SCIF) survey, we develop a multidimensional measure of cultural identity that incorporates both self-identification and observable factors. The analysis reveals that cultural identity significantly affects employment outcomes, with dual identity (where immigrants maintain both their identity of origin and adopt aspects of the host country’s identity) proving to be the most beneficial for securing employment, even in a labour market constrained by an economic recession. This effect is evident for both men and women, although it is more pronounced among women. Based on these findings, policies that discourage the preservation of one’s original identity are not justified if the goal is to enhance immigrants’ success in the labour market. This study highlights the necessity of policies that promote cultural integration while also supporting cultural diversity.

Luca Buzzanca (Gran Sasso Science Institute) (90)
Labor Market Effects of Climate Extremes: Evidence from Italian Agriculture

This paper investigates the impacts of extreme weather and drought events on agricultural firms and labor market outcomes in Italy, focusing on municipalities with predominant grape cultivation. Using detailed employer-employee data and a difference-in-differences event study design, we find that rapid-onset shocks, such as large hail and heavy precipitation, reduce the number of workers, average wages, and days worked, with effects varying by firm size. Slow-onset droughts exhibit limited immediate labor impacts. Our preliminary results highlight the heterogeneous and non- linear effects of climate extremes on agricultural labor markets and underscore the importance of targeted policy interventions to support workforce resilience. By documenting substantial short-term  osses in earnings and employment, our findings also suggest that climate and weather extremes can exacerbate income instability and labor insecurity in the agricultural sector, raising concerns for both social protection systems and the long-term sustainability of agricultural employment.

C – NAME

Luca Buzzanca, Carlo Caporali (Gran Sasso Science Institute) (50-J)
Drought, Mafia and Slavery: The Nigeria-Italy Case Study

We investigate the effects of climate-induced migration on criminal dynamics in destination countries. Leveraging restricted-access monthly and provincial-level data on human trafficking-related crimes, our difference-in-differences estimates show that the severe drought shock occurring in Southern Nigeria in 2016 increased human trafficking-related crimes committed by Nigerian nationals in Italy. Specifically, we estimate an increase of 18 additional crimes attributable to this climate shock, representing a 150\% rise relative to pre-shock levels. We then explore the role of existing organized crime structures and community networks in facilitating the exploitation of migrants using provincial-level data. This analysis reveals that a stronger presence of criminal organizations and larger Nigerian communities in certain provinces explains a higher incidence of crimes related to human trafficking.

Tarana Chauhan (Center for Philosophy, Politics, and Economics, Brown University), Berber Kramer, Patrick S. Ward, and Subhransu Pattnaik
Agricultural credit and women’s agency: Experimental evidence from India

Women in rural households face limited financial access and agency within the household. We hypothesize that improving liquidity constraints for households can allow for resource allocation that improves welfare for women. We test this using a randomized provision of collateral-free loans to over 1,300 small and marginal farmers across 80 villages in India. Employing ANCOVA and IV estimations, we find that households increase ownership of productive capital and consumer durables in response to the program. Women in these households report independent access to banking services through debit cards and mobile phones as well as more time for leisure. The program did not seek to impact women’s financial autonomy or time use, yet had unintended, positive consequences through resource transfer. 

Nancy Chau, Huiyi Chen (James Madison University), Oleg Firsin (62-J)
Social Networks and the Spread of Strikes

This paper examines the role of social networks in the diffusion of labor strikes in the United States. Using new data from the Cornell ILR Labor Action Tracker (2021–2024) and Facebook-based county connectedness measures, we document that strikes spread primarily through social networks rather than spatial, industrial, or political linkages. A 1% increase in network exposure is associated with a 2.3% contemporaneous and 0.8% lagged increase in strike activity. We show that both informational and behavioral channels drive diffusion. Policy and administrative environments, such as right-to-work laws and public-sector notice requirements, shape the timing and persistence of these network effects.

Chen Chen (Brandeis University) (80)
Public Health Restrictions and Household Instability: Evidence from China’s COVID-19 Lockdown

Public health restrictions generate significant health benefits, but their social outcomes remain underexplored. I use China’s COVID-19 lockdown as a quasi-experimental setting to explore its impact on household instability. In a staggered Difference-in-Difference framework, I find that each additional week of lockdown is associated with a 4% increase in divorce-related web searches across prefectures, indicating heightened household instability on the intensive margin. The surge in searches for domestic violence during the same period suggests that it played an important role in amplifying divorce concerns. Moreover, I develop more precise measures of lockdown stringency and find that more stringent shutdowns are associated with greater marital concerns. Further analyses suggest that higher levels of savings and unemployment insurance coverage help alleviate marital strains during lockdowns. Consistently, individual-level data show that lockdowns lead to a lasting decline in marital satisfaction, particularly among younger women. In contrast, the impacts on parent–child relationships are milder. My findings, based on the largest-scale lockdown, underscore the importance of addressing intra-household tensions in extensive public health interventions.

I Chun Chen (Mahidol University),Ruttiya Bhula-or (56-J)
Economic Sustainability of Community-Based Long-Term Care for Aging Populations: A Comparative Qualitative Analysis of Labor Market and Financing Challenges in the United States and Thailand

As the global population ages, CB-LTC systems face economic challenges from labour shortages and funding gaps, especially in high- and upper-middle-income countries. This affects family caregiver labour supply, lowers wages in LTC markets, and threatens the fiscal sustainability of community-based LTC. This qualitative comparative case study explores these issues in the United States (characterised by a fragmented, market-driven system) and Thailand (with universal health coverage and a reliance on volunteers). It is based on semi-structured interviews with 31 key informants (22 from the US, 9 from Thailand) conducted between January and June 2024. Using population economics frameworks , including demographic economics models on ageing populations’ impact on labour supply (e.g., demographic models addressing labour market frictions in ageing societies, and thematic analysis with Atlas. ti, four main themes emerged: the roles and motivations of informants influencing labour participation; the system’s effectiveness amid financial pressures; key challenges and adaptive strategies; and successful initiatives that support viability. To complement the qualitative findings, we incorporate descriptive statistics and back-of-the-envelope fiscal projections from official sources like OECD and World Bank, illustrating comparative cost frameworks. The findings show a ‘hybridity paradox’: Thailand’s models use cultural volunteers for savings but are hindered by silos; US models rely on advocacy and immigration for innovation but face fragmentation and $5 billion annual costs. This expands LTC research by highlighting economic incentives and ROI in training, aiding low- and middle-income countries seeking equity and scalability. Policy suggestions include hybrid approaches, like US-style caregiver benefits within Thai UHC, to improve retention and intergenerational equity, supporting UN SDGS. The insights aim to shape resilient ageing strategies globally, with calls for long-term cost research.

Carmel U. Chiswick  (George Washington University).   
 Economic Development in the 21^st Century

The goal of economic development is to raise standards of living in LDCs, to be achieved by accumulating both human and non-human capital so as to maximize production net of the cost of these investments.   An LDC economy is modelled with two sectors, modern and traditional, each of which uses its own type of human and non-human capital in production.  Sector-specific human capital is specified as an attribute embodied in its workers, who have agency to choose their sector of employment and level of education.  Earnings of labor are the sum of two components:  recovery of human capital investment costs (e.g., student loan repayments) and an economic rent (i.e., profit) available for current consumption.  The consumption-maximizing resource allocation equalizes rates of return to investments in all types of capital and allocates workers between the two sectors so that labor rents (i.e. consumption levels) are the same in both.  Policy implications emphasize removing economic, social and cultural barriers to economic mobility for all resources.    


Amelie F. Constant (University of Pennsylvania).
Dynamic migration transitions between wealthy home and host countries by natives                        

Utilizing unique panel data from the registers in Norway that encompass the entire population I study emigration and return migration by native Norwegians from 2000 to 2022. I examine the stochastic processes of migrating to different locations each year, which depend on the current location and the preceding location as individuals choose the location that offers them the highest utility. I find that Norwegians who emigrate are strongly self-selected in many characteristics. They are more likely to emigrate when they are well educated, out of the labor force, in below-average income brackets, and have wealth. Whereas the well-educated and the employed have higher odds of returning, those in higher-than-average income brackets have lower odds. Unanticipated is that Norwegians with wealth are more likely to return. The predicted transition probabilities of emigrating are quite low, making staying in Norway the dominant probability. Although the predicted transition probabilities of returning are relatively higher, staying abroad, especially in the USA, Canada, Australia, and New Zealand is also high. These results remain robust after several sensitivity tests. Norwegians in the 21st century are not very migratory and not as eager to leave their wealthy peaceful country.

D – NAME

Antonio Di Paolo (Universitat de Barcelona & AQR-IREA). (12)
Language of Instruction, Bilingualism, and Neighbourhood Quality: Do Local Language Skills Matter?

Christiaan de Swardt (RWI-Leibniz Institute for Economic Research & Ruhr University Bochum), Renate Hartwig.  
The Marriage Squeeze: Measuring and Explaining Marriage Market Dynamics in Sub-Saharan Africa.

Marriage market imbalances have been linked to social instability, crime, and reduced welfare in both developing and developed countries. We revisit this issue in the context of Sub-Saharan Africa, where high population growth and the prevalence of polygyny shape partner availability in structurally significant ways. Building on a dynamic model of the marriage market, we introduce a novel measure that accounts explicitly for the intensity of polygyny and apply it to Demographic and Health Survey data from 1991 to 2023. We show that conventional marriage ratios, which ignore polygyny, understate the extent of male-biased competition in the marriage market. When polygyny is incorporated, we find that marriage imbalances have persisted at structurally high levels since 2006, especially in rural areas and among Muslim populations. These patterns are driven by demographic transitions—declining population growth and narrowing partner age gaps—that amplify the impact of polygyny on male surplus. Our findings highlight the importance of accounting for marriage market dynamics in the design of policies related to demographic change, family formation, and social stability in low-income settings.

Maria Laura Di Tommaso (Università di Torino), Silvia Mendolia, Silvia Palmaccio, Giulia Savio (143)
Is Physical Unattractiveness a Risk Factor for Sexual Violence Perpetration? Evidence from the U.S

A comprehensive understanding of the determinants of sexual violence constitutes a crucial step toward effective prevention. While there is much research on the role of socio-economic circumstances of both victims and perpetrators, little is known about whether an individual’s physical attractiveness influences the likelihood of perpetrating sexual violence. Using U.S. data from the National Longitudinal Study of Adolescent to Adult Health (Add Health), we examine the relationship between attractiveness and sexual violence perpetration against one’s partner. Physical attractiveness is measured using interviewer-assigned scores for respondents aged 12 to 17. While no correlation is documented for women, among men, a one-point increase in attractiveness (on a 1–5 scale) reduces the likelihood of perpetrating sexual violence in adulthood by 13 percent. We also find that contextual factors, such as parental education and neighborhood cohesion, mitigate this association. Very unattractive men with highly educated parents have substantially lower predicted probabilities of perpetrating sexual violence against the partner, compared to peers with less-educated parents. Similarly, very unattractive men living in cohesive neighborhoods report markedly lower predicted frequencies of sexual violence perpetration than those in less cohesive neighborhoods.

Pawani Dasgupta (University of Groningen), Maite Laméris, Milena Nikolova. (104).
Macroeconomic Conditions during the Impressionable Years and Adult Civic Engagement

An individual’s economic experiences shape their socio-political preferences and behaviours. We study whether this extends to civic engagement, a critical factor in sustaining social, political, and economic institutions. Our paper builds on the impressionable years’ hypothesis, which views ages 18-25 as particularly important for the formation of socio-political attitudes. Combining individual-level data from the Gallup World Poll with macroeconomic indicators, our identification strategy exploits cross-country and cross-cohort variation. We find that recession experiences when 18-25 positively affect civic engagement as an adult. In addition, we also find that the intensity of recessions matter; individuals exposed to deeper recessions during early adulthood exhibit a stronger positive effect on the likelihood of civic engagement during adulthood. Our results contribute to the growing literature on the impressionable years, and extend it by linking recession experiences when young and civic engagement as an adult. These insights have broader implications for understanding how societies recover from economic crises and how such experiences can, in turn, strengthen social cohesion.

Cynthia Bansak, Eva Dziadula (University of Notre Dame), Madeline Zavodny. (21)
The Role of Coresident Grandparents in Maternal Employment among Asians in the US.

Women’s fertility and employment decisions are closely linked, and both may be influenced by the availability of childcare. We investigate the role of living with a grandparent– a potential source of childcare– in maternal employment, focusing on differences between Asian and nonHispanic white women in the United States. Specifically, we examine whether the higher share of Asians living in intergenerational households can account for the relatively small drop in Asian women’s employment after having a child, compared with their white counterparts. We f ind that living in an intergenerational household is positively related to employment among Asian women, while the opposite is the case among white women. Furthermore, having a coresident grandparent is positively related to employment among Asian mothers, but negatively among whites. Coresident grandparents appear to be the main explanation for why the smaller maternal employment gap is almost 10 percentage points smaller among Asians than among their white counterparts. We also investigate differences among Asian women by immigrant status. Ability to sponsor and support foreign parents may influence whether foreign-born Asian mothers use grandparents as a source of childcare. We find significant differences in the prevalence of coresident grandparents by Asian women’s immigrant status. However, those differences appear to have relatively little effect on differences in maternal employment among Asians by immigrant status.

E – NAME

Maye Ehab (Institute for Employment Research) and Katja Möhring (20)
Adaptation or continuation? Refugees’ labor market participation and working hours before and after migration

This study examines the relationship between refugees’ pre-migration employment behavior and their labor market participation and working hours in Germany. Recent research has emphasized female refugees’ participation in the labor market as shaped by care burdens, discrimination, and gender role attitudes and norms. However, limited attention was paid to employment trajectories in the country of origin. This paper investigates whether refugees’ employment participation and working hours in Germany represent a continuation of their pre-migration employment behavior or an adaptation to destination country normative and institutional context. Using data from the IAB-BAMF-SOEP survey (2016–2023) for refugees from Syria, Afghanistan, and Iraq who arrived between 2013 and 2019 in Germany, we apply a two-part regression model to analyze labor market participation and weekly working hours. We test whether higher pre-migration working hours increase the likelihood of employment in Germany and explore whether the number of working hours before migration continues or adapts in the host country, with potential gender differences. Preliminary results show that although pre-migration working hours influence employment likelihood for both genders, only men exhibit a positive association between working hours before and after migration. For women, no association is found, suggesting that employment trajectories for female refugees are shaped more strongly by post-migration constraints. These findings highlight significant gendered patterns of labor market integration among refugees in Germany.

F – NAME

Sofya Feygenson (Korea Advanced Institute of Science and Technology, Graduate School of Digital Humanities and Computational Social Sciences), Jun Hyung Kim (Korea Advanced Institute of Science and Technology). (86)
Unstable Jobs, Delayed Families: A Hidden Markov Model of Life-Course Transitions in South Korea

South Korea’s persistently low fertility and unstable youth employment reflect a fundamental misalignment between work, marriage, and family formation. This study examines the mechanisms linking these domains by applying a two-state Hidden Markov Model (HMM) to longitudinal data from the Korean Youth Panel Survey (YP2007), which follows individuals aged 26 to 35 from 2007 to 2020. Unlike conventional regression or event-history models that estimate static relationships, the HMM identifies latent life-course states and the probabilities of transitions between them. This approach allows for the analysis of how employment, marriage, and fertility decisions interact and evolve over time.

The model identifies two hidden life-course regimes. The Career-Oriented state is characterized by high employment, delayed marriage, and low fertility. The Family-Oriented state shows higher rates of marriage and childbearing but lower participation in the labor force. At the beginning of the observation period, 93.7 percent of respondents are in the Career-Oriented state, which reflects the dominance of employment-focused trajectories in the late twenties. As individuals reach their early thirties, transitions to the Family-Oriented state become more common, especially among women. Once entered, the Family state is highly persistent. The probability of remaining in it is 99.6 percent, while the probability of returning to the Career-Oriented state is only 0.4 percent. These asymmetric transitions suggest that family formation is largely irreversible under current institutional conditions.

Gender and age patterns provide additional insight. Women tend to transition earlier, at around age 30 on average, while men shift closer to age 32. This pattern reflects social expectations that men should achieve stable employment before marriage. Women’s steeper transition pattern indicates a compressed time window for marriage and childbirth, which contributes to the “now or never” dilemma that characterizes Korea’s low-fertility context. Men’s longer persistence in the Career-Oriented state reflects the close link between economic stability and social readiness for family formation.

Educational attainment also differentiates life-course trajectories. Highly educated individuals are overrepresented among both long-term Career-Oriented stayers and successful Family-Oriented movers. Education thus serves as a double-edged factor. It delays family transitions by extending schooling and early career investment but later enables them by improving job security and income. This pattern shows that education both postpones and facilitates family formation within Korea’s segmented labor market.

From a methodological perspective, the HMM framework provides a significant advantage over static models. It captures unobserved heterogeneity, temporal dependence, and sequential causality, and it identifies underlying life-course states rather than treating employment, marriage, and fertility as independent outcomes. This dynamic modeling approach reveals how individuals remain in or move between integrated work–family regimes, offering insights that conventional regression or survival models cannot provide.

The results have important policy implications. Since most young adults start in the Career-Oriented state, improving early career stability through predictable employment and affordable housing could reduce the uncertainty that discourages family planning. The persistence of the Family state indicates that once people form families, they tend to stay in that state. Lowering the barriers to entry, for example through childcare support, paid parental leave, and more gender-equal workplace policies, could therefore have lasting demographic effects.

In conclusion, Korea’s fertility crisis is driven not only by individual preferences but also by structural rigidity in the life course. The two-state HMM demonstrates that employment, marriage, and fertility are closely linked and path-dependent. Policies that promote flexibility across these transitions, rather than isolated fertility incentives, are essential for creating a more balanced and sustainable demographic future.

Danilo Cavapozzi, Enrico Fornasiero (Ca’ Foscari University of Venice/University of Naples Parthenope), Teresa Randazzo. (95)
The Effects of the Indian Mid-Day Meal Scheme on Cognitive and Health Outcomes of Children in Andhra Pradesh

This paper analyses the impact of the Indian Mid-Day Meal Scheme on the health and cognitive outcomes of schooling children living in the Indian State of Andhra Pradesh. We exploit the variability derived from the individual educational history of children, combined with the phased implementation of the program targeting only students in the public sector, to construct a variable measuring the monthly cumulative exposure to the Mid-Day Scheme. We provide evidence of the positive impact of the policy on children attending public schools, particularly in reducing inequalities between children enrolled in the private and public sectors. Lastly, employing a Heckman Selection model accounting for the selection issue on the type of school attended by children, we show that the impact of the policy is positive and consistent regardless of the type of school attended.

G – NAME

Despina Gavresi (University of Luxembourg), Andreas Irmen, Anastasia Litina (70-J)
Population Aging and the Rise of Populism in Europe

This paper identifies population aging as an important driver of populism using multilevel regression analysis on individuals from European countries between 2002 and 2019. Unlike individual aging, we focus on population-level demographic change measured by the old-age dependency ratio (OADR), i.e., the ratio of people aged 65 and over to those of working age. which captures the structural balance between older and economically active populations. Using data from nine rounds of the European Social Survey, we examine the relationship between population aging and populist attitudes, captured through voting for populist parties, political trust, and immigration attitudes. Our findings suggest that population aging is associated with declining electoral turnout, higher support for populist parties, lower trust in political institutions, and increased anti-immigrant sentiment. These effects appear across both younger and older voters, indicating that aging societies influence political preferences beyond individual aging. They may operate through mechanisms such as economic insecurity, cultural backlash, or shifts in collective societal priorities.

Christian Grund (RWTH Aachen University) (13-J)
Conditional Gender Pay Gaps

Although considered a problem, gender pay gaps are still prevalent in most countries. In contrast to evidence based on broad country-wide samples, this contribution complements the literature by studying the case of a homogeneous sample of middle managers within one industry. Gender pay gaps are conditional on several factors. Main parts of considerable raw gender pay gaps within the sample are explained by individual and job-based characteristics. Gaps are more pronounced at the bottom of the wage distribution and are much more relevant for contingent pay than they are for fixed salaries. Gender pay gaps are positively moderated by the level of the hierarchy and actual working hours, whereas previous parental leave and unemployment spells negatively moderate gender gaps.

Zhiming Cheng, Sarah Cook, Liwen Guo (University of New South Wales & The Kids Research Institute Australia & GLO), Massimiliano Tani. (77-J)
Environmental Policy and Gender Health Gap

Gender health gaps persist globally. We test whether stringent environmental regulation narrows these disparities. Using a nationally representative longitudinal sample of middle aged and older Chinese adults, we examine the rollout of the country’s most stringent clean air policy with three-way fixed effects and Oaxaca Blinder decompositions. The policy improves women’s health relative to men, narrowing the gap, especially among people with chronic lung disease and residents of major cities. Decomposition results show that the policy primarily reduces the discrimination component, defined as the unexplained share reflecting different returns to the same endowments, while other components change little. These findings document persistent inequality and heterogeneous policy benefits. They indicate that women face disadvantages consistent with discrimination arising from social or biological factors, and that clean air regulation advances gender health equity chiefly by narrowing returns differentials rather than by changing observable characteristics.

Laszlo Goerke (IAAEU, Trier University, IZA, CESifo, GLO), Sven Hartmann, Yue Huang. (85-J)
Councils of contentment: Works councils and income perceptions

Using data from the German Socio-Economic Panel, we provide the first comprehensive investigation of the relationship between workplace co-determination in the form of works councils and income satisfaction. Controlling for a wide range of individual, job-related, and firm-level characteristics in OLS and fixed effects specifications, we observe that employees working in establishments with a works council report significantly higher income satisfaction compared to their counterparts in non-co-determined firms. We further examine possible mechanisms behind this relationship. The rank in the income distribution, the perceived fairness of the wage, and working conditions emerge as quantitatively relevant factors in explaining the positive correlation between co-determination and income satisfaction.

Jaroslav Groero (CERGE-EI), Alena Bicakova. (122)
Beyond Test Scores: The Effect of School Entry Age on Specific Cognitive Processes

This paper examines how school starting age affects the development of specific cognitive skills that underpin human capital formation. We link school entry timing to narrow cognitive abilities using high-resolution psychometric data from children aged 8 to 13. We exploit a cutoff rule policy that assigns school eligibility based on month of birth and estimate the intention-to-treat (ITT) effect of starting school later on specific cognitive skills using regression discontinuity design. Combining the results from two models with different sets of fixed effects allow us to gain interpretive leverage on underlying mechanisms. Delayed entry improves both inductive reasoning and learning effectiveness when comparing children (of different age) within the same grade, but only enhances learning effectiveness when comparing children of the same age (across grades), suggesting that maturity drives gains in reasoning, while instruction contributes to learning effectiveness. The effects are stronger for boys and children from lower-educated families, indicating that delayed entry may serve as a compensatory input for disadvantaged groups. Our findings provide new evidence on the specific cognitive channels through which early schooling decisions affect long-run skill development.

H – NAME

Elghafiky Bimardhika, Daniel Halim (World Bank). (76-J)
To have it all? Career and family choices of college-educated Indonesian women

Are highly educated women empowered to balance their career and family objectives? This paper investigates the evolving preferences for work and family life among college-educated women in Indonesia, spanning birth cohorts from the 1950s to the 1990s. Our research reveals a growing polarization of the labor market and family formation outcomes among the younger cohorts of female college graduates. Younger women either delay marriage to stay in the labor force or opt out of the labor force altogether post-marriage. We explore two potential drivers for this trend. First, we find that this trend coincides with rising female participation in time-demanding, high-skilled, traditionally male-dominated professions. Second, on the other hand, we find that there is a rising conservatism of gender views among young men that might have contributed to marriage market frictions.

I  — NAME

Antonia K. Entorf, Miriam Gensowski, Ingo E. Isphording (Max Planck Institute for Research on Collective Goods)
Mental Health Challenges Among Teachers: The Role of Occupation and Workplaces

Teacher mental health is a growing societal and economic concern with potential repercussions for educational quality and student outcomes. Using population-wide administrative data from Denmark, we document increasingly elevated rates of mental health care utilization among primary school teachers. These elevated rates are fully explained by the specific occupational content of teaching, such as interpersonal demands, conflict intensity, and emotional labor. Second, we document large differences in mental health outcomes across schools: rates of sickness absence and mental health specialist contact are roughly twice as high in schools at the 90th percentile as in those at the 10th percentile. Exploiting teacher mobility between schools, we show that this variation reflects a causal effect of schools: incoming teachers’ health adjusts to new environments within the first years. Observable school characteristics do not predict observed differences, and estimates based on within-teacher variation in assigned student cohorts do not reveal a causal link between these characteristics and teacher mental health. Thus, we conclude that unobserved aspects of workplace quality, including leadership and organizational climate, may play a central role.

J – NAME

Souvik Banerjee, Preeti Jaiswal (IIT Bombay), Sankar Mukhopadhyay (142-J)
Motherhood and Labour Market Outcomes: Penalty or Premium?

Using nationally representative longitudinal data from the Consumer Pyramids Household Survey, we examine the effect of childbirth on female labour market outcomes in India, focusing on both the extensive and intensive margins within an event study framework. Contrary to findings from similar studies in developed countries, we do not observe any motherhood penalty in earnings, labour force participation, employment or work hours post-childbirth, after accounting for unobserved individual heterogeneity. Interestingly, we find that the birth of a child leads to a 27.4% and 32.6% increase in women’s average earnings in urban and rural regions, respectively, relative to non-mothers. This motherhood premium seems to arise partly due to higher employment after childbirth. Further, we find that the increase in the likelihood of employment is predominantly observed among women from lower caste, Hindu religion, lower income quartiles, those with primary education, and higher order births in urban regions. In rural regions, the effect is restricted to women from the lowest income quartiles. Additionally, we provide evidence suggesting that the presence of older siblings in the household increases the likelihood of women’s employment by 3.7 percentage points. These findings underscore the role of socio-economic factors in shaping the labour market outcomes of women in India.

Hugo Jales (Syracuse University), Zhengfei Yu (125)
Minimum wage and Informality in a Roy Bargaining Economy: Evidence from a Bunching Estimator

We study the determination of employment, formality, and wages using a bargaining model featuring compensating differentials and self-selection. Our framework allows us to create a novel taxonomy of formal employment that complements the taxonomy used in the literature to discuss informality (Ulyssea, 2018). This taxonomy is shown to be useful to characterize the effects of labor market policies such as the minimum wage. We use the model to estimate the effects of the minimum wage in Brazil using the PNAD dataset for the years 2001-2005. Our results suggest that the minimum wage decreases formality by approximately 8%.

K – NAME

Sua Kang (Korea University), Wookun Kim (SMU and CESifo). (30)
Kanghyock Koh (Korea University and IZA). Childbirth, Baby Bonus, and Maternal Mental Health

We study the impacts of childbirth on maternal mental health and the role of pro-natalist cash transfers. Using claims-level data from South Korea’s universal healthcare system, we find that mental health diagnoses rise by 34.8% (198.7%) after the first (second) birth. We find little evidence that cash transfers mitigate these effects. As potential mechanisms, we examine liquidity constraints, labor market changes, time use, and social stigma. Lastly, we document that poor mental health after childbirth is negatively associated with the likelihood of having another child.

Hyunji Kim (University of Washington) (65)
Maternal Health Programs and the Continuation of Unintended Pregnancies

Maternal and Child Health (MCH) programs promote safe motherhood by combining financial incentives for institutional delivery with community health outreach and family-planning education. I study how MCH programs influence reproductive behavior through pre-conception margin, by affecting contraceptive use and fertility, and post-conception margin by affecting whether unintended pregnancies continue to birth. Using a difference-in-differences design exploiting India’s national MCH program, I find that it increased modern contraceptive use by 12 percent and reduced fertility among older women. In contrast, the program increased unplanned births among younger women by four percentage points. A later phase of the program that expanded health-worker incentives shows that greater contact with community health workers contributed to this rise. These findings reveal an unintended consequence of maternal health interventions: by promoting both planned and protected pregnancy, they can inadvertently constrain women’s reproductive autonomy.

Giulia Briselli (ESCP Business School), Wookun Kim (SMU, CESifo). (29)
Unintended Consequences of Immigration Reform: Marriage Market, Intra-Household Bargaining, and Well-Being.

We examine the consequences of South Korea’s 2008–10 immigration reforms on the marriage market and intra-household outcomes. The reforms unintentionally reduced foreign bride inflows. Exploiting regional variation in exposure to the reforms and using uniquely rich data—administrative records, household surveys, and registries—we find that the reforms resulted in fewer new marriages, increased women’s intra-household bargaining power, shifted in women’s time from housework to employment, and increased well-being for both spouses. Divorce rates fell, with a shift from general incompatibility to abuse-related grounds. These findings reveal the reforms’ unintended impacts on household dynamics and broader economic implications.

Vijetha Koppa (Zayed University) (36)Does easier access to Alcohol increase Domestic Violence – Evidence from Local Option Elections

Understanding the causal relationship between alcohol access and domestic violence remains a critical policy question at the intersection of public health and economics. A substantial body of research has attempted to document the relationship between alcohol consumption and various forms of interpersonal violence, including child abuse and spousal violence (Averett & Wang, 2016; Markowitz, 2000; Markowitz & Grossman, 2000). Though much has been written in the popular press regarding the harmful effects of alcohol on domestic violence, the findings from research, however, are mixed. Some of the evidence is also correlational and/or based on weak first-stage effects of the changed alcohol policy. Therefore, establishing causality has proven difficult, with changes in alcohol access often being endogenous to local social attitudes and enforcement intensity.

This paper contributes to this literature by exploiting plausibly exogenous variation in alcohol availability generated by local option elections in Texas. Under Texas law, cities, counties, and justice-of-the-peace precincts can hold referenda to determine whether and how alcoholic beverages may be sold. These elections not only decide whether alcohol can be sold but also specify the type of liquor (beer, wine, or spirits) and the outlets permitted to sell it—whether off-premises (e.g., grocery and liquor stores) or on-premises (e.g., bars and restaurants holding a liquor license). Because these elections are initiated through citizen petitions rather than statewide mandates, their timing and geographic distribution create a rich source of quasi-experimental variation in local alcohol access.

Importantly, A 2003 court decision substantially relaxed the legal requirements for verifying petition signatures, leading to a sharp and discontinuous increase in the number of local option elections—from an average of fewer than ten per year to between fifty and seventy annually. This institutional change introduced as-good-as-random variation in the timing of elections and, consequently, in the timing of transition of jurisdictions from “dry” to “wet” statuses. My study period of 1995 to 2011, therefore, provides an ideal window to analyze how shifts in local alcohol policy affect rates of domestic violence while mitigating concerns about policy endogeneity or selection bias.

The identification strategy follows a differences-in-differences framework in which jurisdictions that remain dry serve as counterfactuals for those that change their alcohol laws. The baseline specification employs a two-way fixed-effects (TWFE) model controlling for jurisdiction and year fixed effects. However, recent methodological work (Goodman-Bacon, 2021; Borusyak, Jaravel, and Spiess, 2022) demonstrates that TWFE estimators can yield biased or even sign-reversed estimates in the presence of staggered treatment timing and heterogeneous treatment effects—both of which characterize this setting. Although much of the existing quasi-experimental literature on alcohol policy has relied on TWFE, this insight calls into question the credibility of earlier results that attribute changes in violent crime to alcohol access without accounting for these issues.

To address this concern, the paper re-estimates treatment effects using modern difference-in-differences methods that properly handle staggered adoption and heterogeneous effects. Under the traditional TWFE specification, estimates appear small and imprecise, occasionally suggesting null or even negative effects of increased alcohol access on domestic violence. When re-estimated using the CSDiD estimator, the treatment effects flip in the expected direction—showing a statistically significant increase in family and intimate-partner violence following the liberalization of alcohol laws. This reversal underscores the substantive importance of methodological advances in causal inference: conclusions about policy effectiveness can depend critically on whether the estimator appropriately accounts for treatment timing and heterogeneity of treatment effects.

Event-study analyses further corroborate the causal interpretation. Pre-treatment trends in family-violence incidents are parallel between treated and untreated cities, while a sharp divergence emerges following the policy change. The estimated effects grow over time, consistent with a gradual adjustment process as local markets for alcohol, including licenses, retail outlets, and supply chains, expand after legalization.

By leveraging an institutional feature, the petition-based local option election system in Texas, and a natural policy discontinuity introduced by the 2003 court ruling, this study provides new and credible evidence that greater alcohol availability leads to higher rates of domestic violence. The findings highlight the need for policymakers to account for the unintended social costs of expanding alcohol markets and demonstrate how careful attention to identification strategies can overturn misleading inferences drawn from conventional empirical models. More broadly, the analysis contributes to the growing literature re-evaluating causal claims in applied microeconomics by illustrating how modest methodological refinements can yield qualitatively different and more reliable insights about the consequences of public policy.

L – NAME

Anran Liu (University of California, Davis), Luoqi Yuan, Jianjun Tang. (150-J)
Property Rights and Late-Life Labor Supply: Evidence from China’s Rural Land Titling Reform

We study the impact of strengthening land tenure security on the labor supply of older workers in a major developing economy. Exploiting the staggered rollout of China’s recent, large-scale rural land titling program as a quasi-natural experiment, we find that enhanced property rights induce significant labor market transitions among the middle-aged and elderly. Our difference-in-differences estimates based on panel data reveal two primary effects. First, land titling leads to a net withdrawal from the labor force, particularly from agricultural activities, as improved tenure security enhances the wealth and insurance value of land. Second, for individuals who remain employed, titling triggers a reallocation of labor from farming towards non-agricultural sectors, often accompanied by an increase in total hours worked. We provide evidence that a key underlying mechanism is the activation of the land rental market; titling significantly increases households’ propensity to rent out their land, thereby separating their land assets from their labor decisions. Our findings highlight the dual role of property rights reform: it can function as an informal social safety net for an aging population while simultaneously promoting structural transformation by reallocating labor to more productive sectors.

Xiaoying Liu (University of Pennsylvania). (37)
Air Pollution and Under-5 Child Mortality: Evidence from China’s Coal Power Plant Phase-out Policy

This paper estimates the impact of the sulfur-control policy on under-5 mortality (U5M) throughout China during the 11th 5-year plan period (2006- 2010). The policy required shutdown of small coal power plants and installation of flue gas desulfurization (FGD). We gather information on 3969 coal power plant units over 30 megawatts that operated since 1990s and 2486 small units required to close or install FGD during 2006-2010 and compile a unique county-year panel data combining coal plants, U5M, socioeconomic, SO2, PM2.5 and meteorological conditions. We apply the IV-Lasso method to assess the impacts of air pollutants on U5M, using sums of distance-weighted capacities of retired or FGD-installed plants within different radii from a county and high-altitude wind conditions as instrumental variables. Our findings indicate the policy saved >32,000 lives of under-5 children during the 5 years, with regional heterogeneity in the policy effects, pointing to possible displacement and spillover effects.

Husame Doganay, Tony Fang, Xingfei Liu (University of Alberta), Saba Ranjbar, Arthur Sweetman. (47)
Earnings Assimilation in Canada (2006-2021): A Seemingly Unrelated Regression Approach

This paper investigates immigrant–native earnings differentials in Canada between 2006 and 2021, using nationally representative data from the Canadian Labour Force Survey. We introduce a two-step empirical strategy that constructs native-based occupational earnings benchmarks and then estimates Seemingly Unrelated Regressions (SUR) to jointly model actual and counterfactual earnings outcomes. This framework allows us to separate wage differences due to occupational sorting from those arising within occupations, capturing both structural and unobserved dimensions of economic assimilation. The results show that immigrants earn significantly less than natives even within the same occupations and after controlling for education, age, and region of origin, though the gap has narrowed modestly for recent cohorts. Years since migration improve earnings outcomes, but convergence remains incomplete, particularly among women and immigrants from non-European regions. The strong positive correlation between residuals of actual and predicted earnings suggests that unobserved productivity.

Yujia Liu (University College London) (117)

The Intergenerational Transmission in Female Labour Force Participation (FLFP): The Role of Gender Norms in Urban China.

This study investigates the intergenerational transmission of female labour force participation (FLFP) and the role of gender norms in urban China between 2010 and 2018. Using data from the China General Social Survey (CGSS) and weighted probit models, the analysis examines whether a husband’s mother’s employment status influences his wife’s likelihood of participating in the labour force. It will also examine how a husband’s gender perspectives were shaped during his adolescence by his mother’s labour force participation history, and whether these perspectives influence his wife’s decision to participate in the labour market. Consistent with Chen and Ge (2018), results from 2010 indicate a positive intergenerational transmission, suggesting that having mothers who did not work shaped men’s traditional gender attitudes and preferences for non-working wives. However, this relationship disappears in 2015 and 2018, implying a weakening of intergenerational transmission mechanisms. The findings further suggest that while husbands’ gender attitudes continue to exert a strong influence on their wives’ labour market participation, these attitudes are no longer correlated with husband’s mothers’ (wives’ mother-in-law’s) employment histories. Taken together, these findings highlight both the persistence of intra-household normative constraints on women’s labour market outcomes and the gradual weakening of intergenerational transmission (at least via the mother-in-law). This research contributes to understanding the evolving social determinants of FLFP in post-reform China and offers new insights into how economic and normative transitions jointly reshape female labour outcomes.

M  — NAME

Tista Mukherjee (Faculty of Economic Sciences, Indian Institute of Science Education and Research Bhopal (IISERB). Beyond the Street: Short-Run Spillover of Publicized Non-Partner Violence on Intimate Partner Abuse

This paper investigates whether exposure to publicized sexual violence affects women’s risk of spousal violence in India. I draw evidence from two complementary analyses. First, using the 2012 Nirbhaya case as a quasi-natural experiment, I estimate a difference-in-differences model exploiting variation in women’s media exposure and marriage timing. I find a short-run increase in emotional and physical spousal violence among exposed women. Second, combining district-level media reports of sexual assaults from the Global Database of Events, Language, and Tone (GDELT) with household data from the National Family Health Survey (NFHS), I find that districts with higher media reporting of sexual crimes exhibit greater spousal control and intimate partner violence. These results suggest that public acts of gender-based violence can spill over into private spheres, possibly through short-run backlash or reinforcement of regressive gender norms.

N  — NAME

Olena Nizalova (University of Kent), Julien Forder
Revisiting the Economic Case for Social Care Spending: Informal Care

There is a long-standing debate about the balance of long-term care delivered by family and friends as opposed to that provided formally by the state and private sector, which underpins the economic case for social care.  We explore the implications of an exogenously driven change in the use of formal care on the utilisation of informal care. Applying an instrumental variable technique to the eight waves of the English Longitudinal Survey of Ageing (2002-2017), we find that formal care use leads to an approximately 20% decline in the probability of receiving informal care when not accounting for individual heterogeneity and to about 12% otherwise. The estimated effect is smaller for men than for women. Simple calculations, based on current estimates for people aged 75 and older, suggest that one extra hour of formal care leads to up to 40 fewer minutes of informal care, or, in monetary terms, one extra pound spent on formal care brings up to 67 pence savings in informal care costs, when the latter is valued at a replacement cost.

O – NAME

David Fadiran & Adeola Oyenubi (University of the Witwatersrand, Johannesburg, South Africa) (52-J)
Spatial inequality, sub-regional governance and subjective well-being: The case of South Africa

This paper examines how spatial variation in inequality and quality of institutions interact to explain variation in subjective wellbeing. Specifically, we ask if better institutions ameliorate the relationship between inequality and subjective wellbeing? To investigate these interactions, we utilize the National Income Dynamics Study (NIDS), a nationally representative survey of individuals across South Africa.  Our results show marked variation in inequality, well-being and governance across districts. We also find that good governance ameliorates the negative relationship between inequality and subjective well-being. Lastly, our result suggests that satisfaction with public services such as electricity is a plausible mechanism explains the mederating role of institution.

P – NAME

Edumundo Pangara (Kyushu University). (14)
Welfare, and Output Volatility under Rules-Based versus Shock-Driven Taxation: Evidence from Consumption and Labor Income Taxes in Selected SADC Member States.

This paper explores optimal tax design and fiscal rule effectiveness in five Sub-Saharan African economies – Angola, Malawi, Mauritius, Mozambique, and South Africa – using a calibrated neoclassical model. We estimate Laffer Curves for both labor and consump- tion taxes, identifying revenue-maximizing tax rates and the behavioral responses underly- ing them. Results show substantial cross-country heterogeneity: labor taxes exhibit lower revenue peaks in countries with more elastic labor supply and informality (e.g., Malawi, Mozambique), while consumption taxes peak at higher rates in more formal economies (e.g., Mauritius, South Africa). We then assess welfare and macroeconomic volatility un- der two fiscal policy regimes – shock-driven versus rules-based. Across all countries and tax types, rules-based regimes yield higher household welfare (with consumption equiv- alent variation gains of 0.18–0.19%) and significantly lower output volatility (by 26–28%). These findings underscore the stabilizing role of fiscal rules and highlight the importance of tailoring tax policy to country-specific structural characteristics. Policy recommendations include prioritizing consumption tax reform in economies with limited labor tax capacity, enhancing fiscal credibility through rules-based frameworks, and investing in adminis- trative capacity to reduce informality. Overall, the study provides actionable insights for improving fiscal policy design and economic resilience in developing countries.

Petru Crudu, Giacomo Pasini (Ca’ Foscari University of Venice). (108-J)
The Health Burden of Job Strain: Evidence from Europe

This paper investigates how lifelong exposure to occupational stressors and job tasks affects health outcomes in later life. Using SHARE data and linking job characteristics to 4-digit ISCO occupations, we quantify the health impact of occupational stressors while controlling for a rich set of demographic and work history variables. We find that physical exertion, proxied by metabolic rate consumption, is the most detrimental occupational stressor, with significantly stronger effects observed among women. The results are robust to both Oster (2019) bounds and the Lewbel (2012) instrumental variable approach. To explore mechanisms, we track disease progression across the life cycle to show how sustained occupational strain contributes to long-term health deterioration. In the second part of the paper, we introduce a LASSO-based method to identify high-risk job tasks and construct a composite Job Strain Intensity index that also captures unobserved stressors. This framework provides a scalable tool for policy design, for instance, by informing targeted retirement schemes or enhancing workplace safety standards in occupations involving high-risk tasks.

Anastasia Litina, Ioannis Patios (University of Macedonia).  (98-J*)
The Impact of Natural Disasters on Migration Attitudes. 

Climate change has intensified the frequency and severity of natural disasters, making it increasingly important to understand their broader consequences on social and political outcomes. This paper examines the interplay between natural disasters measured by the number of total affected individuals and their attitudes toward immigrants, exploring whether such a shock can lead to increased solidarity or heightened resentment toward immigrants. We use a setting in which we compare Eurozone regions with non-Eurozone regions, thus exploiting the differential degree of integration across countries and the role of joint immigration policies. Linking data from the International Disaster Database (EM-DAT) and the European Social Survey (ESS), we associate disaster-affected individuals with their perceptions on various immigration issues and dimensions. We employ a difference-in-differences approach with staggered treatment adoption, where the first difference compares Eurozone and non-Eurozone countries, and the second difference accounts for the timing of being affected by a disaster. Our main findings indicate that natural disaster shocks in Eurozone countries are associated with more positive attitudes toward immigrants, particularly regarding their acceptance and perceived economic contribution. These effects emerge gradually after the shock, suggesting that disasters may foster longer-term social reflection rather than immediate solidarity responses. A plausible explanation is that Eurozone countries, being more economically and institutionally integrated, experience such shocks within a framework of shared responsibility and interdependence. In the presence of a common currency and coordinated fiscal mechanisms, these countries may also be better shielded from the economic fallout of disasters, reducing the sense of economic insecurity that can fuel exclusionary attitudes. By contrast, non-Eurozone regions facing more severe and unbuffered economic consequences may respond in more inward-looking ways, emphasizing national over collective concerns. Heterogeneity tests further highlight the critical role of factors in propagating the effect such as remittances paid, trade in services, unemployment, wage and salaried workers, and EU funding measures.

L – NAME

Sara Lemos (University of Leicester) and Jonathan Portes (King’s College London). (57)
The Impact of Immigration on Wages and Employment in the UK Using Longitudinal Administrative Data.

We study the labour market impact of immigration to the United Kingdom, focusing on the large inflows following the 2004 EU enlargement. Using the Lifetime Labour Market Database (LLMDB)—a longitudinal 1% sample of National Insurance records—we provide the first analysis of immigration’s effects on employment and wages based on high-quality administrative microdata. Exploiting individual, area and time fixed effects, as well as area-time, individual-time and individual-area fixed effects, we reduce endogeneity concerns that have limited previous work. We find limited aggregate impacts, but distributional consequences: existing immigrants—particularly those who were young or low paid—experienced modest negative employment effects, while natives faced little evidence of displacement. For wages, impacts were mixed: existing immigrants overall gained, but low-paid immigrants lost. The results suggest labour market adjustment operated through both substitution and complementarities across groups. More broadly, we provide a methodological framework for analysing the much larger and more diverse post-2021 immigration flows.

Jiaheng Li (Macquarie University) (63)
Predicting dynamic vulnerability to multidimensional poverty in China

Predicting household vulnerability to multidimensional poverty (VMP) is a vital task for designing effective anti-poverty policies. Using datasets from the China Household Finance Survey and a two-level prediction of vulnerability building upon the Alkire-Foster framework, this study estimates the conditional probability of being deprived in multiple domains. The estimation leverages 13 machine-learning techniques and 4 traditional techniques as a hybrid approach to predict vulnerability and test the intertemporal accuracy of the prediction. Counting-based methods are also employed to construct the composite Vulnerability to the Multidimensional Poverty Index. Further decomposing analysis from counting-based Foster–Greer–Thorbecke indices and machine learning-based Shapley Additive exPlanations (SHAP) allows us to identify the trend and source of VMP. We conduct several robustness checks of the estimated results using alternative methods such as intertemporal accuracy and class imbalance. Our main results on methods revolve around the performance of methods where ensemble learning outperforms other machine-learning algorithms in predicting household VMP. Our main results on estimates show that the headcount ratio of vulnerability is lower than that of poverty, the vulnerability of urban households is mainly induced by risk, while nearly all are affected by poverty among rural households. SHAP further reveals the contributions of sociodemographic, shock response and financial characteristics of the households to welfare outcomes.

M – NAME

Adeola Oyenubi, Uma Kollamparambil, Lesego Masenya (University of the Witwatersrand). (93-J)
Comparative Life Evaluation: A Relative Density Analysis of Native and Migrant Populations

While disparities in life evaluation between migrant and native-born populations are well-documented in the literature, existing evidence predominantly relies on mean-based statistical comparisons, which may obscure the underlying heterogeneity. This study employs a Relative Density method which allows for a more granular, distribution-sensitive approach to understanding wellbeing dynamics. We conduct a time-disaggregated analysis that spans 2011 to 2023/24 based on a population representative data of the Gauteng province of South Africa. We found that the life evaluation distribution for migrants is located to the left of the distribution for non-migrants. Implying that non-migrants are better off. Further, net of the location effect, the distribution for migrants is more polarized than that of non-migrants. Our analysis reveals that these effects vary over time, and the largest difference in the distributions occur prior to the COVID-19 pandemic. In addition, our analysis suggests that the pandemic coincides with a narrowing of the life evaluation gap between migrants and non-migrants. Oaxaca-Blinder decomposition of the observed distributional differences (or relative densities) identifies socioeconomic factors—specifically wealth, educational attainment, employment status, and satisfaction with public service delivery—as the principal drivers of the aggregate well-being gap. Additionally, the study finds that a higher concentration of migrants at the municipal level is negatively correlated with the well-being of both populations. However, this effect is stronger for migrants so that it contributes to the well-being gap. The findings underscore that addressing migrant deprivation is critical for improving population-level well-being and reducing wellbeing inequality within the South African context.

Kaibalyapati Mishra (Institute for Social & Economic Change, Bengaluru). (3-J*)
SCREENING, SIGNALLING, AND THE ECONOMIC VALUE OF HIGHER EDUCATION IN INDIA.

We test whether higher education in India functions primarily as a productivity-enhancing investment, a screening device by employers, or a signaling mechanism by workers, using nationally representative PLFS data from 2018-19 to 2023-24. Using instrumental variable analysis with parental education as an instrument, we find conventional OLS estimates overstate higher education returns by 227%(67.9% vs 20.7% IV estimate), revealing substantial screening bias in formal sector hiring. Regression discontinuity designs at educational thresholds show secondary completion (12 years) yields 45-87% wage premiums, confirming employers use credentials as screening devices. The screening effect intensifies in formal occupations, where education returns reach 83.9%comparedto60.0%inunscreened jobs, while experience only commands positive returns (0.47% yearly) in screened positions versus negative returns (-0.09%) in informal work. However, this credential-based sorting exacerbates structural inequalities: urban workers capture 38.9% premiums in screened jobs (versus 26.1% informal), and while affirmative action modestly reduces ST disadvantages in screened employment (-4.9% vs-10.7%), SC penalties remain severe (-21.4%). Gender gaps narrow only marginally in screened sectors (58.5% penalty versus 68.3%), suggesting formal hiring processes imperfectly mitigate discrimination. Heterogeneous effects reveal education’s signaling value is strongest for marginalized groups- SCsrealize 138% returns at secondary completion and women gain 109% versus 41% for men- indicating credentials partially compensate for discriminatory sorting. Our findings challenge human capital orthodoxy by demonstrating how screening institutions: (1) transform education into exclusionary gatekeeping mechanisms that amplify spatial (urban-rural) and social (caste) hierarchies; (2) generate credential inflation that disproportionately burdens disadvantaged job seekers; while (3) paradoxically offering marginalized groups their most viable mobility pathway through standardized certification.

João Pereira dos Santos (Universidade Nova de Lisboa) (79)
Perceptions of (mis)behavior by gender: Evidence from the Catholic World Youth Day

We provide the first field experiment dealing with prescriptive gender stereotyping on moral issues during a mass religious event that is likely to make these judgments salient. The survey experiment took place during the 2023 World Youth Day in Lisbon, an event with Pope Francis that gathered 2 million pilgrims from around the World. We asked 1800 Pilgrims from 80 nationalities to judge hypothetical norm-violating scenarios with varying genders, related to lying, sexual promiscuity, and parental care. We find that respondents exhibit high levels of judgment, with women being, in general, less harshly judged than men. We also find that women judge other women less severely in the scenario related to parental care, while men do not discriminate by gender in any scenario. Moreover, more conservative respondents are more judgmental, but do not discriminate between genders, while less fundamentalist ones are less judgmental, but discriminate in favour of women.

Bilal Ahmad Bhat, Gargi Sarkar (IIT Kanpur), Sarani Saha, Sounak Thakur (129*)
Dowries, Debts and Children’s Learning Outcomes: Evidence from India

Large sums of dowry — marital transfers from the bride to the groom — are customarily paid in Indian weddings. We show that households that have more daughters are more likely to be indebted, reflecting financial strain plausibly induced by the obligation to pay large dowries. This financial strain is mirrored in worse learning outcomes amongst children in households that have a large number of daughters and report being in debt. The results appear to be driven by lower consumption expenditure and greater maternal labor supply in indebted families with firstborn daughters. Consistent with strong son-preference in India, we present suggestive evidence that girls’ learning outcomes are more adversely affected than that of boys.

P – NAME

Giorgio d’Agostino, Donatella Lanari, Luca Pieroni (University of Perugia) (102)
Shifting Attitudes: The Impact of COVID-19 on Perceptions towards Immigrants in Africa

This paper examines the impact of the COVID-19 pandemic on native born citizens’ perceptions of immigrants in Africa. The pandemic onset generated heightened economic and social uncertainty through its adverse effects on health and economic conditions. We use Afrobarometer data from 2017 to 2022 alongside the Oxford COVID-19 Government Response Tracker to explore these changes in perceptions toward migrants. By applying a Difference-in-Differences approach to ordinal data, we preserve the complexities in migration attitudes, revealing a no table deterioration in the most positive perceptions of immigrants, particularly pronounced in countries of Central and Western Africa. Our extended model reveals significant heterogeneity in these effects, demonstrating that inefficiency in pandemic management serves as the primary driver of variation in native attitudes toward immigrants. Specifically, countries exhibiting poor governmental responses to COVID-19 experienced a sharper decline in positive attitudes. We identify dis tinct transmission channels underlying these shifts: negative expectations regarding future income, concerns about competition for public health services, and perceived threats to cultural norms, values, and societal identity. These mechanisms collectively explain how pandemic-related uncertainties translated into deteriorating perceptions of immigrants, emerging as particularly salient in contexts of governance failures.

Pragati (Indian Institute of Management Bangalore), Tirthatanmoy Das
Health Coverage and Educational Investments

Human capital theory posits that investments in health and education are complementary. Health investments boost education demand, thereby increasing educational expenditures. Drawing on this theory, this paper examines the impact of health coverage on three aspects of household educational investments: (a) the share of educational expenditures, (b) the level of educational spending, and (c) the probability of taking an educational loan. Using a health insurance scheme in India, this study employs a modified difference-in-difference strategy and two waves of the Indian Human Development Survey (2004-05 and 2011-12). The findings show that health insurance led to a 10 percent rise in the share of educational expenses, a 39 percent increase in per person educational spending, and a 185 percent rise in the likelihood of taking education loans. These effects are stronger for households below the poverty line. At the aggregate level, this externality translates to an additional one-tenth to one-fourth of every unit spent on the health coverage budget. All results are robust to changes in subsamples.

R – NAME

Enrica De Cian, Filippo Pavanello, Teresa Randazzo (University of Messina). (94)
Does social identity influence households’ adaptation to hot temperatures?

This paper investigates the role of social identity, proxied by caste, in shaping adaptation responses to rising temperatures in India. Using new rich household data, we find that marginalized castes aresignificantly less likely than upper castes to adopt cooling appliances in response to hot temperatures. This disparity persists even after controlling for income, education, and other socio-economic and demographic characteristics. To uncover the underlying mechanisms, we document the importance of (i) adherence to group social norms, (ii) discrimination in upper caste-dominated areas, and (iii) unequal access to credit. Our findings highlight how social identity can exacerbate inequalities in heat adaptation. At the same time, they point to opportunities for policy interventions that leverage intra-group dynamics to promote the diffusion of adaptive practices within marginalized communities.

Jacek Barszczewski, Sophie Brochet, Prasanthi Ramakrishnan (Southern Methodist University). (110)
Across-District Marriage Migration in India

Approximately two-thirds of Indian women migrate for marriage, with roughly one-fifth relocating across district boundaries. Given the skewed regional disparities in India’s sex ratio, we examine the influence of this on women’s marriage migration patterns and female intra-household bargaining power. We document new empirical facts associated with sex ratio, probability of marriage migration, and household characteristics. We find that women are more likely to migrate for marriage to a region with a more skewed sex ratio and to a rural household, where the household head has at least primary education. We then build a collective household marriage market model to understand the impact of marriage migration on bargaining power of women. Preliminary results indicate that an increase in the probability of marriage migration results in lower bargaining power among women.

FThomas Goldring, David C. Ribar (Georgia State University). (91-J)
How Children Combine Pre-Kindergarten and Subsidized Child Care in Georgia.

This study uses administrative data from Georgia’s Childcare and Parent Services (CAPS) subsidized child care program and its universal pre-kindergarten (Pre-K) program to examine how families combine these two types of services. Specifically, it analyzes (1) Pre-K enrollment among children in the CAPS program during the school year that they are age-eligible for Pre-K, and (2) continuing and new spells of CAPS participation among children during the school year that they enroll in Pre-K. It gives special attention to racial, ethnic, and geographic differences in program use. The study’s multivariate models reveal that non-Hispanic Black children in its analysis groups enroll in Pre-K and participate in subsidized child care at higher rates than other children. The results also indicate that children in non-metropolitan counties are more likely to enroll in Pre-K but less likely to participate in CAPS than other children. Overall, children in the CAPS program enroll in Pre-K at higher rates than other children. A significant facilitating factor is that many CAPS providers also offer Pre-K services at the same site.

Xiangqing Liu, Elisabetta Lodigiani, Silvana Robone (University of Eastern Piedmont), Elisa Tosetti, and Giorgio Vittadini. (89-J) The Effect of the Great Recession on the Mental Health Care of Immigrant and Native Workers in Italy.

In this paper we investigate the impact of the 2008 Great Recession on mental
health care utilisation among migrant and native workers who experienced job
changes during the period 2007-2011 in the Lombardy region, in Italy. We exploit
a unique administrative data set about employees residing in Lombardy, matched
with data on psychotropic drug prescriptions and hospitalisations for psychiatric
disorders, and we employ a continuous difference-in-differences approach to estimate
the causal effect of the Great Recession on health care utilisation, looking at
heterogeneous effects between natives and immigrants. Our results show that the
Great Recession significantly increased mental health care utilisation among native
workers, while its impact on immigrant workers is minimal and statistically insignificant.
To better understand the reasons behind the disparity between natives and
immigrants, we analyze the impact of the Great Recession on their employment
outcomes, considering employment status a key transmission mechanism, linking
the economic shock to mental health care utilisation. Our labour market analysis
indicates that both immigrant and native workers are negatively impacted by
the Great Recession and that immigrants are more likely than Italian workers to
transition from employment to unemployment in areas more severely affected by
the crisis. This pattern holds across gender and age groups. Therefore, we tend to
rule out the possibility that immigrants did not increase their mental health care
utilisation due to a lack of labour market impact. From a broader perspective, our
findings point at the complexity of the relationship between economic crises and
mental health care utilisation, which can be influenced by various factors, including
access to services, social stigma, and the broader economic and policy context.

S – NAME

Dimitris Vallis, Riikka Savolainen (Swansea University), Jonathan Portes. (107)The impact of the pandemic on health-related inactivity and benefit claims

We study the medium‑run effects of the COVID‑19 pandemic on economic inactivity and benefit claims among working‑age adults in the United Kingdom. Unlike most advanced economies, the UK has experienced a sharp and persistent rise in inactivity due to long‑term sickness since 2020, alongside a surge in disability – and incapacity‑related benefit claims. Using individual‑level data from the UK Household Longitudinal Study (UKHLS), linked to the COVID‑19 Study, we examine whether these developments were concentrated among clinically vulnerable individuals – those advised to shield at the onset of the pandemic. Our empirical strategy combines coarsened exact matching with difference‑in‑differences (DiD) and triple‑differences (DDD) designs. We find that clinically vulnerable individuals were 1-2 percentage points more likely to exit into inactivity due to long-term sickness or disability after the pandemic, relative to comparable non‑vulnerable peers. By contrast, we find no robust evidence of disproportionate increases in disability or incapacity benefit receipt among the vulnerable.

Jayanta Sarkar (Queensland University of Technology). (140-J)
Breaking the Mold: Norms, Childcare, and the Dynamics of Female Labor

Despite convergence in female education and formal institutions, large differences in female labor force participation persist across countries. This paper develops a dynamic overlapping-generations model in which women inherit social participation norms from the previous generation and incur utility penalties for deviating from them. Maternal time and formal childcare are direct substitutes in childrearing. When labor market returns are high enough to offset childcare costs, women optimally deviate from low inherited norms by substituting formal care for maternal time, shifting the norm upward for the next cohort. When returns are insufficient or sanctions are strong, conformity is optimal and norms persist. This mechanism generates an S-shaped best response function in labor supply, producing multiple stable equilibria that differ in participation rates, fertility, and childcare regimes. Numerical simulations demonstrate equilibrium clusters spanning a broad range of participation rates and childcare arrangements, accounting for cross-country heterogeneity in female labor force participation at similar fundamentals. Policy experiments show that isolated interventions – reducing social penalties, subsidizing childcare, or raising wages – shift outcomes within existing clusters but preserve the equilibrium structure. Combined policies operate synergistically to expand the set of attainable equilibria and support higher participation with sustained fertility.

Tomas Sarkozi (Bratislava University of Economics and Business), Martin Kahanec (97-J*)
Do Public Employment Programs Benefit Marginalized Communities?

Background: Public employment programs (PEPs) are widely used to support vulnerable jobseekers, yet evidence on post-program employment is mixed, especially for marginalized Roma communities (MRC) in Central and Eastern Europe (CEE).
Data and methods: Using two waves of the EU-SILC Roma modules (2018, 2020), we define treatment as PEP participation during the 12 months prior to the 2020 survey. We construct a matched comparison via propensity score on variables of age, sex, and employment in 2018, then estimate a Difference-in-Differences (DiD) logistic model of employment.
Results: Employment probabilities were similar across groups at baseline (2018, 6% at reference values). By 2020, the control group’s predicted employment probability rose to 6%, while the treated group’s fell to 5.6%, yielding approximately a DiD of 11.7 percentage points on the probability scale.
Conclusions: Within this observational design, PEP participation was not associated with improved near-term employment for MRC participants relative to similar non-participants over 2018–2020. Patterns are consistent with prior findings of weak or negative post-program employment effects for PEPs.
Limitations: Only two time points (one pre-trend period), a small, treated subsample, potential unobserved confounding including the COVID-19 pandemic, and limited detail on program heterogeneity.
Implications: PEPs may be more effective when embedded in broader upskilling strategies. Future work should add subsequent waves, richer treatment measures, and robustness checks.

Stefan Schneck (Institut für Mittelstandsforschung) (x)
The origins of entrepreneurship: How parental role models and socialization shape later entrepreneurial intentions

This exploratory study examines the effects of parental socialization and parental role models at ages 7 to 10 on the entrepreneurial intentions of their children in adolescence. Analysis of German household data and more than 1,400 observations shows a moderation effect between parental role models and socialization. An adolescent’s willingness to become self-employed in the future is influenced by parental role models and moderated by parental child-rearing practices related to risk-taking during childhood. While child-rearing practices not focused on risk-avoidance reinforce the parental role model effect and increase an adolescent’s intentions to become self-employed, parental child-rearing practices geared toward risk aversion nullify any positive effects of having self-employed parents as role models. Parental socialization during childhood thus casts a long-term shadow and may explain why some children with self-emp76

oyed parents have as little intention of becoming self-employed as children of employees. Early parental socialization practices may, thus, contribute to explaining the lack of willing entrepreneurs and family business successors.

Paige Schoonover (Saint Mary’s College of California). (121)
Reacting to recalls: contraceptive choice impacts of defective birth control pills in Chile

I examine the impact of an oral contraceptive recall and the news of resulting pregnancies in Chile on contraceptive choice among women aged 15–25. Using an event study approach, I compare changes in contraceptive use by women residing in areas (comunas) with a government pharmacy, who are more likely to only be offered the recalled brand of birth control pill, to women residing in comunas with private pharmacies, who would be offered a wider variety of birth control pill brands following the recalls. The results show that the recall and subsequent news coverage led to a 23.1% decrease in oral contraceptive use. There is some indication of increased use of long-acting reversible contraceptives (LARCs) and increased usage of no effective form of contraception; however, these estimates are not statistically significant and thus do not provide conclusive evidence of such shifts. This study highlights the broader implications of product recalls on consumer trust and reproductive health, demonstrating how disruptions in contraceptive supply can lead to substantial shifts in usage patterns with potential long-term consequences for unintended pregnancies and public health policy.

Natalia Danzer, Rachel E Kranton, Piotr Pawel Larysz, Claudia Senik (Paris School of Economics). (35*) Gender Identity, Norms, and Happiness.

How do gender identity and norms relate to happiness? This paper takes advantage of the 2024 European Social Survey, which asks respondents to report their feelings of femininity and masculinity, and studies the relationships between these self-assessments, (non-)conformity to gender norms, and life satisfaction. The results show a robust asymmetry between men and women. For men, feeling more masculine, behaving in ways more typical of men, and life satisfaction are all positively cross-correlated. For women, while feeling more feminine and life satisfaction are similarly positively correlated, behaving in ways more typical of women is, in contrast, associated with lower life satisfaction. These patterns vary across European regions, potentially reflecting different histories. The results are robust to alternative measures of typical behavior of men and women and subjective well-being. The findings support theories of gender identity and reveal possible trade-offs implied by gender norms for women.

Josep Amer Mestre, Manuel Serrano-Alarcon (Joint Research Centre) (99-J)
Unpacking the Current Surge in Sick Leave: Insights from Spanish Administrative Data

This paper investigates the factors explaining the sharp rise in sick leave prevalence. Using data from Spain’s Social Security records we analyse the evolution of the prevalence of days on sick leave across individuals and work factors and implement an Oaxaca-Blinder decomposition to examine each factor’s contribution to the rise in prevalence of sick leave. We document a 36% in the share of days on sick leave, common to a wide range of demographic and employment characteristics, including age, education level, sex, country of birth, industry sector, and employment contract. Observable factors such as population ageing, the expansion of employment opportunities, or longer healthcare waiting times account for roughly 25% of the rise in days on sick leave. While musculoskeletal conditions remain the most common diagnoses, mental health disorders have seen the most dramatic relative increase, more than doubling among younger workers. Notably, mental health disease accounts for between 35% and 56% of the rise in days on sick leave among individuals under the age of 40, being this burden particularly high for women. This highlights a concerning deterioration in mental health among the younger working population with broader consequences for both workers, firms and the welfare system.

Xiaogang Li, Ze Song (Nankai University), Puyang Sun, Hong Zou
Stagnation and Differentiation in Growth: Quality Effects of Consumer Goods for Chinese Households The slowdown in quality growth of consumer goods has caused household consumption stagnation. This study systematically examines the dynamic evolution of consumption quality effects among urban Chinese households. Using micro-level data, we find that durable goods quality grew at only 0.504% annually from 2002–2016, far below international benchmarks, driven by slow technological iteration and homogeneous innovation. Traditional durables exhibit re-purchase rates <5%, with stagnant demand unresponsive to quality premiums. Quality improvements accounted for 61% of living cost fluctuations but followed an inverted U-shaped trajectory, peaking at 80.72% in 2008–2010 before declining, signaling diminishing returns from conventional upgrading models. Our findings underscore critical challenges—including stagnant durable goods quality, structural innovation-demand gaps, and policy inclusivity deficits—in China’s consumption landscape, while identifying opportunities for targeted upgrades in technology, standardization, and policy design to foster sustainable consumption growth.

Ignat Stepanok (Institute for Employment Research, IAB) (25)
Migration and Intellectual Property Rights Protection

In this paper I study how intellectual property rights (IPR) protection affects migration. I build a North-South Schumpeterian model of growth with endogenous migration of high- and low-skilled workers. People can invest in education and work as skilled in the R&D sector or remain low skilled and work in production in both countries. Imitation in the South is endogenous and costly. Stronger intellectual property rights protection in the South lowers the imitation rate, decreases low-skilled and increases high-skilled migration to the North, overall migration decreases. The number of high-skilled people working in the South nevertheless increases because more Southerners invest in education. The welfare of high- and low-skilled natives in the North and in the South and also of migrants increases unambiguously. Higher migration costs, in turn, decrease migration and lead to a higher imitation rate in the South.

Jakob Madsen, Zeresh Errol, Holger Strulik (University of Goettingen). From Spirits to Crime: Two Centuries of Alcohol and Homicide in the West (59-J)

This paper presents a novel dataset covering homicides, alcohol consumption, education, and several other variables across 14 advanced countries from 1800 to 2016, and investigates the determinants of homicide rates over time. To structure the empirical analysis, we develop a simple model in which criminal behavior is influenced by alcohol consumption and key confounding factors such as education and deterrence. In this framework, intoxication from alcohol reduces self-control and increases the propensity for crime. Our empirical findings show that alcohol and drug use, alongside with deterrence measures, overall educational attainment, and the educational gap between potential offenders and victims, are significant predictors of homicide rates. These results remain robust across alternative time periods, methods addressing endogeneity, and the inclusion of additional control variables highlighted in the crime literature.

Sholeh Maani (University of Auckland, New Zealand), Olga Sudareva (University of Auckland, New Zealand). (139-J)

Assessing the Impact of Crime on Gender Disparities in Labour Market Outcomes.

This study examines the gender-asymmetric effect of crime on individual working hours, providing the first evidence that disentangles the impact of subjective safety perceptions from both current and past exposure to crime. We construct a new dataset by merging individual-level data from nine rounds of the European Social Survey (ESS) with historical and modern crime statistics, as well as macroeconomic indicators, differentiating between immigrant and native individuals residing in 30 predominantly European countries. We find that all respondents who feel unsafe work fewer hours, but the effect is stronger for women. Moreover, the effect of objective crime exposure is gendered: partnered men work longer weekly hours in higher crime environments, whereas partnered women work less, suggesting an intrahousehold compensatory work response to crime. Additionally, past exposure to extremely high crime levels has a persistent negative impact on working hours, even if the current environment is significantly safer.

T — NAME

 Bastien Bernela, Liliane Bonnal, Inès TOURE (Poitiers Economics Laboratory, University of Poitiers), and Ahmed Tritah. (148-J)
Educational mismatch, spatial mobility, and wage inequality: Evidence from France young graduates

This paper investigates the wage returns to educational mismatch among young higher education graduates in France and examines whether geographical mobility moderates these effects. Using panel data from Céreq’s Génération 2017 survey (2017-2023) and employing ORU (Over-Required-Under) wage equations with Heckman selection corrections, we find that educational mismatches generate substantial and asymmetric wage effects. Required education yields the highest returns, while overeducation generates significantly lower returns and undereducation imposes considerable penalties. Recentered Influence Function regressions reveal striking distributional heterogeneity: overeducation penalties decline across the wage distribution and become insignificant at higher quantiles, while undereducation penalties persist throughout and peak at the median. Decomposition analysis indicates that both overeducation and undereducation penalties stem predominantly from differential returns to characteristics rather than compositional differences, though this pattern is particularly pronounced for undereducated workers where coefficient effects account for over three-quarters of wage gaps at most quantiles. We document heterogeneous effects across gender, social origin, and migration background. Also, geographical mobility substantially moderates mismatch costs, with effectiveness varying systematically with distance moved.

V  — NAME

Nicholas A. Jolly, Nikolaos Theodoropoulos, Georgios Voucharas (Liverpool Hope University). (49)
Business Closures, Labor Market Policies and Gender Gaps.

Using individual-level data from the Survey of Health, Ageing, and Retirement in Europe (1986–2016), we examine the relationship between expenditures on active and passive labor market policies and a range of labor market outcomes associated with job displacement, with a focus on gender differences. Active policies are linked to higher re-employment probabilities and longer employment durations, and lower unemployment risk. Effects vary by program type and gender. Our results suggest that tailoring policies to address childcare responsibilities can improve women’s employment outcomes and influence men’s industry transitions. These findings highlight the importance of gender-sensitive labor market interventions.

W – NAME

Yunbo Liu, Zexuan Wang (Minzu University of China), Zesen Zhang, Jue Bai, Xiaoyang Ye

Occupational Cognition and Employment Choices in Manufacturing: Evidence from the Information Intervention Experiment with Vocational College Students

Ensuring high-quality employment for skilled workers in the manufacturing industry is an important issue in building a stable new industrial workforce and serving the strategy of building a manufacturing power. Based on the data from the third round of the National Vocational Education Student Development Survey, this study conducted a randomized controlled experiment involving 2,539 students in the equipment manufacturing category of higher vocational colleges across the country to examine the impact of information intervention on students’ career choices. The research findings are as follows: (1) Overall, information intervention significantly enhanced students’ willingness to choose careers in manufacturing; (2) In terms of the action path of information intervention, the significant roles of the information correction perspective and the long-termism perspective have been verified. Information intervention can enhance students’ willingness to choose careers in manufacturing by making up for information deficiency and establishing long-term development confidence; (3) Information intervention shows significant heterogeneity among different groups. The effect of information intervention is more obvious among student groups with high learning motivation, high professional identity, low socioeconomic status and low digital literacy, reflecting the incentive effect and information supplementation effect of information intervention. The research conclusion provides a practical basis for alleviating the structural employment contradiction in the manufacturing industry and the supply of skilled talents in manufacturing.

Yanxia Yu, Jacques Poot, W. Robert Reed, Weilun Wu (University of Canterbury). (58-J)Meta-analysis of the Impact of Population Age-Composition on Aggregate Saving Rates: A Global Perspective

Abstract:
According to the life-cycle hypothesis of saving behavior, the working population directly or indirectly funds the consumption of children and retired individuals. Consequently, a population’s age composition influences the aggregate gross saving rate. However, many other factors also affect aggregate saving rates, making it challenging to quantify the impact of declining fertility, increasing longevity, and changing population structures on global savings.

This study conducts a meta-analysis of 70 econometric studies of country-level saving rates published between 1969 and the present. Coefficients from these studies are standardized into two comparable effect sizes: (1) the long-run percentage point change in a country’s saving rate when the young-age dependency ratio (YDR) increases by 1 percentage point, and (2) the corresponding effect for the old-age dependency ratio (ODR).

To address heterogeneity across and within studies, we examine differences in study focus, geographic coverage, functional forms, model restrictions, data structure (cross-section, time series, or panel), covariate selection, and estimation techniques. We also apply multiple estimators to handle unobserved heterogeneity, within-study correlation, and publication bias.

Our results provide robust evidence that higher demographic dependency reduces aggregate saving rates. Specifically, a 1 percentage point increase in the YDR decreases saving rates by about 0.2 percentage points in the long run, while the effect of ODR is approximately twice as large. Using the latest global population projections, we estimate that increasing demographic dependency could lower the global gross saving rate by 5.3 to 12.7 percentage points between now and 2100, with the upper bound more likely if fertility rates continue to decline. The effect is notably stronger in OECD countries than in developing economies.

Emmi Wilén (University of Oulu), Sanna Huikari, Jouko Miettunen, Marko Korhonen. (105)
Temperament traits and longitudinal earnings: increasing returns over time and at the top

There is increasing consensus that besides standard human capital variables, non-cognitive (or soft) skills are important predictors of labour market success. Yet, we know relatively little about which are the traits that matter, when, and among whom. Combining rich longitudinal data from the population-based Northern Finland Birth Cohort 1966 with multiple national registers – including earnings records from 1995 to 2016 (n = 5033), covering individuals from ages 29 to 50 – we provide the first evidence on how neurobiologically based Cloninger’s temperament traits relate to long-term earnings trajectories. We examine how temperament traits are associated with individual earnings over time, across the earnings distribution, and by sex. Our results show these traits are predominantly rewarded at the upper end of the earnings distribution. Moreover, return patterns differ between traits, suggesting that distinct dimensions of non-cognitive ability are valued differently across earnings levels. Harm avoidance is penalized across all earnings levels, whereas returns to novelty seeking, reward dependence and persistence are concentrated at the upper end of the earnings distribution. We also find that the importance of temperament traits increases with age, indicating that they matter most during prime working years (ages 40 to 50). Finally, we reveal notable sex differences: reward dependence is substantially more detrimental for high-earning females than males. Our findings reinforce the growing evidence that non-cognitive skills are important determinants of earnings. Furthermore, we provide new insights into when and for whom these traits matter, showing that returns to temperament traits are heterogeneous across age, sex, and earnings levels.

Michael Windsor (Bankwest Curtin Economics Centre), Astghik Mavisakalyan, Loan Vu, Alan Duncan. (68)
Breathe In The Air: Institutional Quality, Political Participation and Air Pollution in Transition Countries

Air pollution is a major issue facing the majority of the world’s population. However, there has been very little research looking at how it could impact institutional trust or political participation. Our study matches three waves of the Life In Transition Survey, a study looking at transition economies which have transitioned from a socialist to a capitalist economic system, with global estimates of pollution and weather. Using a two-stage least squares (2SLS) approach instrumenting long-term exposure to pollution with temperature inversions we estimate the effects of pollution on institutional trust and political participation. We find that increases in pollution are associated with decreases in trust towards institutions and increased perceptions of corruption amongst institutions. We then look at the impacts on political participation finding that pollution decrease institutional political participation in a subsample of European transition countries and increases other, non-institutional forms of political participation such as petition signing.

James Kai-sing Kung, Wenbing Wu (University of Melbourne)
The Rise of the Chinese Clan

For centuries the Chinese clan functioned as a socio-cultural and economic organization. Yet, its evolution over time remains underexplored. By leveraging a historical natural experiment over a newly digitized dataset covering 800 years (c. 1136–1935), we find that the Ritual Reform of 1536 contributed to 31% of the growth in ancestral halls – the hallmark of clans. By allowing commoners to worship their first ancestor, the reform effectively expanded clan size and increased the demand for ancestral halls construction. Historically, the more clannish counties experienced higher population growth, more developed markets, and provided more public goods and social welfare.

Y – NAME

Siew Ling Yew (Monash University), Jie Zhang. (124)
Health externalities to labor productivity and optimal policies with endogenous fertility, labor, and longevity

We study the optimal health subsidy in a dynastic model of altruistic middle-aged and old agents who choose health spending, savings, inter-vivos transfers, and the number of children. Health spending enhances labor productivity and longevity, but firms cannot observe individuals’ health spending. This causes health externalities on labor productivity. We show that the externalities reduce health spending, labor productivity, and longevity but may indirectly increase savings and decrease fertility from the first-best levels. Public or employer-based health subsidies increase health spending to an age-dependent level, but these subsidies to old agents also increase their transfers to middle-aged agents (the transfer cost of childrearing). Age-specific labor-income taxes decrease age-specific health spending and the (time and transfer) costs of childrearing. The decentralization of the social optimum thus requires appropriate age-specific public or employer-based health subsidies and age-specific labor-income taxes to an extent that is increasing with the degree of the health externalities.

Hanming Fang, Jiayin Hu, Miao Yu (Peking University)
Maternity Leave Extensions and Gender Gaps: Evidence from an Online Job Platform

We investigate the unintended consequences of maternity leave extension on gender gaps in the labor market. Using millions of job applications on an online job platform and the staggered extension of maternity leave across Chinese provinces, we find that an average increase (22%) in the length of paid maternity leave led to a 3.7 percentage point decline in positive callbacks to female applicants relative to their male counterparts. In response, female job seekers submitted 4.4 more job applications, shifted toward jobs with 5.4% lower wages, and experienced 0.9 weeks longer job search duration than male applicants. We also find that government subsidies that partially cover firms’ wage costs of extended maternity leave help alleviate its adverse impact on gender disparities in hiring.

Z – NAME

Yaron Zelekha (Ono Academic College). (9 bank)
Systemic Bias in Criminal Justice: Evidence from Two Natural Experiments                                              

Racial disparities are pervasive in criminal legal systems worldwide, affecting both democratic countries such as the United States (Alsina & La Ferrara, 2014; Rehavi & Starr, 2014; Cohen & Yang, 2019; Agan, 2024) and autocratic regimes like China (Hou & Truex, 2022).

The literature has documented that racial bias permeates the entire spectrum of the criminal justice process, from investigation and arrest through to charging and sentencing (Rehavi & Starr, 2014; Agan, 2024). Research, predominantly focusing on Black Americans, has shown that Black individuals are more likely than their White counterparts to experience a range of adverse outcomes: they are more frequently victims of crime, more often subjected to interrogation and arrest, more likely to be detained before trial, more frequently charged with severe offenses, and ultimately more likely to be convicted and receive harsher sentences (Agan, 2024). This bias is evident not only in non-professional jury trials (Anwar et al., 2012) but also in cases judged by professional judges (Alesina & La Ferrara, 2014).

There is broad consensus that racial stereotypes play a significant role in shaping racial bias. According to these accounts, decision-makers throughout the enforcement, prosecution, and sentencing processes project their stereotypical beliefs about certain racial groups onto individuals. For instance, Black American men are often stereotypically viewed as aggressive, disrespectful, and predisposed to criminal behavior (Bridges & Steen, 1998). These stereotypes can also influence how their emotional expressions, such as remorse, are interpreted differently (Zhao & Rogalin, 2024).

Race-related salient characteristics, such as skin tone and facial features (King & Johnson, 2016), are therefore used by decision-makers—whether consciously or unconsciously—as prominent and observable informational cues. These cues help compensate for the incomplete information available when making decisions (O’Flaherty & Sethi, 2008, 2019; Lang & Spitzer, 2020; Peng & Cheng, 2022; Agan, 2024). Moreover, when cognitive load is high, the tendency to rely on heuristics and salient stereotypical cues becomes more pronounced, with these cues being given greater weight (Tversky & Kahneman, 1974; Bordalo et al., 2016; Agan, 2024).These studies correspond to large body of work in the legal sphere that has systematically demonstrated how various heuristics and cognitive biases affect legal decisions (Zamir & Teichman, 2018; Teichman et al., 2023).  

Criminal legal systems play a significant role as institutional elements of social control (Peng & Cheng, 2022). According to conflict theory, society is composed of competing groups, with the state organized to serve the interests of the wealthy and powerful (Stolzenberg et al., 2013). Within this framework, racial biases in the justice system emerge when the majority, advantaged group perceives a threat from a minority, disadvantaged group. This perceived threat may be physical, economic, or cultural (Hou & Truex, 2014). Consequently, criminal law becomes a tool for social control, aiming to protect the interests of the dominant group (Stolzenberg et al., 2013). Biases are perpetuated by police officers, prosecutors, juries, and judges, who are often members of the majority, advantaged group (Anwar et al., 2012; Hou & Truex, 2014; Rehavi & Starr, 2014; Raphael & Rozo, 2019; Agan, 2024; Zhao et al., 2024). This perspective aligns with ethnic conflict literature, which suggests that minority groups are perceived as particularly threatening due to their higher likelihood of engaging in civil conflict (Horowitz, 1985; King & Wheelock, 2007; Morelli & Rohner, 2015).

However, racial disparities can emerge not only from direct stereotypical biases but also from other sources of discrimination or disparities that indirectly affect the legal process. Economic inequalities—such as differences in wealth, education, and housing—can be correlated with criminal behavior and must be accounted for (McConnell & Rasul, 2018; Agan, 2024) while in parallel can affect juror decisions (Ahrsjo et al., 2024). Additionally, racial disparities may result from increased involvement in criminal behaviors due to broader societal discrimination that is not specifically related to the legal process (Sorensen et al., 2003).

Given the multitude of correlated socioeconomic, demographic, and cultural variables, correlational studies that aim to make causal claims about direct discrimination are prone to significant endogeneity problems and can be highly problematic (Agan, 2024). Moreover, measuring stereotypical beliefs poses challenges, as it is difficult to directly assess the extent of stereotypical tendencies among court officers.

The literature has employed various empirical approaches to address the challenges in making causal claims about stereotypical discrimination. First, some studies use controlled models to account for numerous socioeconomic, demographic, and cultural variables that may be correlated with criminal behavior (McConnell & Rasul, 2018; Agan, 2024) or criminal and case processing histories (Rehavi & Starr, 2014). Second, other models aim to eliminate sample selection bias, for example, by using sentencing guidelines (Sorensen et al., 2012). Third, some research controls for judges’ political, ethnic, and gender characteristics (Lim et al., 2016; Cohen & Yang, 2019). Fourth, studies examining appeal cases assume that appellate judges are less influenced by stereotypical biases (Alesina & La Ferrara, 2014). Finally, laboratory-based case scenarios have been utilized (Teichman et al., 2023). Despite these efforts, none of these methods can fully establish causality between direct discrimination and racial bias, as they do not involve manipulating potential discriminatory tendencies in a real field experiment.

This paper makes two central contributions to the literature on racial bias in criminal justice by employing a novel methodology and a uniquely comprehensive dataset that together enable the implementation of two natural experiments. First, it directly addresses a major empirical gap by causally identifying the presence of direct discrimination in judicial outcomes—something that previous observational and experimental approaches have struggled to conclusively establish. Second, it fills a theoretical gap by distinguishing between two underlying drivers of discriminatory behavior: stereotype-based bias rooted in perceived threat, and broader animus-driven racism toward ethnic or religious minorities.

The first natural experiment exploits exogenous variation in prosecution policy, specifically a shift toward lower prosecution rates, to test whether reducing cognitive load leads to decreased reliance on stereotypical judgments as expected by the literature (Tversky & Kahneman, 1974; Bordalo et al., 2016; Agan, 2024). The second leverages fluctuations in the perceived physical threat from minority groups—measured by terror attack intensity—to examine whether heightened threat salience amplifies sentencing disparities, thus contributing to the limited literature on the effect of terror on racial bias in criminal justice (McConnell & Rasul, 2021). In both settings, the estimators are assumed to be neutral with respect to race, enabling identification of discriminatory behavior when disparities persist.

The analysis draws on a dataset comprising over 1 million criminal sentencing records and comprehensive data on terror attack casualties in Israel between 2000 and 2022. The models control for a rich set of covariates, including sociodemographic variables, felony type, district, year, and two-way fixed effects at the court level.

Israel offers a rare and compelling context for examining variations in stereotypical bias induced by physical threats, given its history of frequent terror attacks and an ongoing conflict predominantly involving actors who identify as Muslim. The country has experienced substantial variation in the frequency and severity of these attacks, impacting both human lives and economic development (Zelekha & Bar-Efrat, 2011). During the study period (2000 – 2022), the average yearly number of terror fatalities was 48.7  (SD = 73.2) and the average yearly number of significant injuries was 525.9 (SD = 456.7).

Moreover, most of the terror attacks originate from neighboring Muslim Palestinians, while approximately twenty percent of the Israeli population is of Palestinian ethnicity (almost 93 percent of them are Muslims and approximately 7 percent are Christian Arabs). The incidence of terror attacks conducted by Palestinians with Israeli citizenship is very low. Consequently, the physical threat posed by Palestinians with Israeli citizenship is negligible. This underscores that the variation in potential physical risk is not significantly associated with Israeli Palestinians, whose sentencing this research will compare to that of their Jewish counterparts.

This research suggests that Israeli Palestinians, mostly Muslims and to a lesser extent Christian Arabs, experience direct discrimination stemming from the stereotypical attitudes of both prosecutors and judges, while Jews enjoying preferential bias. Understanding whether such direct discrimination occurs is crucial for developing effective policy responses aimed at addressing racial disparities. When direct stereotypical discrimination is identified and made known, decision-makers—such as teachers grading students—can take steps to mitigate their prejudiced attitudes (Alesina et al., 2024).

By providing causal evidence of direct discrimination and disentangling its underlying motivations, this research offers actionable insights for legal reform. It underscores the need for targeted interventions—particularly in high-stress or high-threat contexts—where cognitive shortcuts and stereotypes may be more likely to shape legal decisions.

Zhengwei YAN), Xu ZHANG (Central University of Finance and Economics). (127-J)
The Externalities of Private Tutoring on Students’ Academic and Noncognitive Outcomes: Evidence and Mechanisms

This paper examines the spillover effects of private tutoring on students’ academic and noncognitive outcomes in the classroom. Utilizing nationally representative data of middle school students randomly assigned to classes in China, we find that students receiving private tutoring have negative impacts on peers’ academic performance and noncognitive skills. We provide evidence that these effects operate through damaging the classroom environment and reducing student effort, instead of affecting teachers’ effort or teaching styles. Furthermore, we find that these effects are more pronounced among male students, students with a rural hukou, students from lower socioeconomic backgrounds, and students with lower ability. Our findings suggest that the rat race in private tutoring during compulsory education may exacerbate educational inequality.

Wenjun Zhao (Hitotsubashi University).(45-J)The Role of Collateral in Marriage: How Property Division up

on Divorce Affects Household Labor Supply
Abstract:

This study examines the impacts of the 2011 legal change in China, which shifted the property division rule upon divorce from an equal-division regime to a title-based one, on household labor supply using a difference-in-differences design. We find that this legal change led to an increase in wives’ labor supply relative to husbands’, as well as a decline in birth rates. Moreover, these impacts are independent of husbands’ preferences. Our findings suggest that joint assets serve as a commitment technology to collateralize marriage. In the absence of such collateral, marriage offers less commitment and financial protection for partners specializing in home production, leading to declines in household specialization, public goods provision, and the value of marriage.

Zhengwei YAN (Central University of Finance and Economics), Xu Zhang (Central University of Finance and Economics). (127-J)
The Externalities of Private Tutoring on Students’ Academic and Noncognitive Outcomes: Evidence and Mechanisms

This paper examines the spillover effects of private tutoring on students’ academic and noncognitive outcomes in the classroom. Utilizing nationally representative data of middle school students randomly assigned to classes in China, we find that students receiving private tutoring have negative impacts on peers’ academic performance and noncognitive skills. We provide evidence that these effects operate through damaging the classroom environment and reducing student effort, instead of affecting teachers’ effort or teaching styles. Furthermore, we find that these effects are more pronounced among male students, students with a rural hukou, students from lower socioeconomic backgrounds, and students with lower ability. Our findings suggest that the rat race in private tutoring during compulsory education may exacerbate educational inequality.

Stefan Pichler, Christopher Prinz, Stefan Thewissen, Nicolas R. Ziebarth (University of Mannheim, ZEW Mannheim). (39-J)
The Economics of Paid Sick Leave

 This article examines the economics of paid sick leave (PSL) from both theoretical and empirical perspectives. PSL research has evolved dynamically over the last decade, primarily driven by U.S. sick pay mandates, which have increased PSL access from 63% to 77% in all U.S. jobs. We begin by discussing the economic rationales for government regulation, particularly the negative externalities associated with contagious diseases when individuals work while sick. Then, we review economic modeling approaches to study optimal PSL policies. After that, we discuss key trade-offs in the general design of PSL schemes along with trade-offs when setting specific policy parameters.

Vilmundur Torfason and Gylfi Zoega (University of Iceland) (69)The impact of economic and social factors on fertility in Iceland, 2014-2022

Fertility in Iceland has declined in recent years and has remained below the replacement level since 2014. Similar trends have been observed in western countries, where both economic and social factors have been identified as potential drivers. This study examines the influence of these factors on fertility in Iceland using administrative tax return data covering the entire population of tax-liable individuals from 2014 to 2022. The results of regression and event study analyses suggest that the decline in fertility may be attributed to delays in first births and a reduced likelihood of third births. The key determinants of fertility identified in the study include age, disposable income, education, cohabitation status, homeownership, and the COVID-19 pandemic. Home ownership, cohabitation, and income disparity between partners is notably associated with the timing of first births. Unobservable factors also appear to play an important role in explaining fertility trends, possibly reflecting an increase in the cost of raising children measured in terms of sacrificed consumption.