nep-edu New Economics Papers
on Education
Issue of 2026–11–16
29 papers chosen by
João Carlos Correia Leitão, Universidade da Beira Interior


  1. Private highs: investigating university overmatch among students from elite schools By Blanden, Jo; Cassagneau-Francis, Oliver; Macmillan, Lindsey; Wyness, Gillian
  2. How is AI Affecting College Major and Course Choices? Evidence from Finland By Ciprian Domnisoru; Tuomas P. Heikkinen
  3. Estimating, Bounding, and Decomposing the Effects of a Parent-Directed School Readiness Program By Flávio Cunha; Qinyou Hu; Andrea Salvati; Kenneth I. Wolpin; Rui Zeng
  4. Age, Education, and Occupational Gradients in Artificial Intelligence Anxiety By Zimmermann, Klaus F.; Karabulut, Gökhan; Keklik, Zeynep; Altıntaş, Kıvanç
  5. The role of information frictions in study habit choice and academic performance By Affonso, Agustina; Aucejo, Esteban; El Khoury, Stephanie; Larroucau, Tomas
  6. Crime Victims' Children By Paul Bingley; Lorenzo Cappellari; Marco Ovidi; Matteo Sandi
  7. Another Empty Promise of Grade Retention? Peer Effects of Classmate Retention By Jalalian, Samuel; Pfeffer, Fabian T.
  8. From Classrooms to Careers: The Impact of Fee Reimbursement Scheme on Education and Employment By Grover, Minali; Sharma, Ajay
  9. Peer Influence in College Applications By Ana M. Gazmuri; Elena Prager
  10. Understanding Migration Responses to Local Shocks By Borusyak, Kirill; Dix-Carneiro, Rafael; Kovak, Brian
  11. Research Grants and Independent Scientific Contributions: Marginal versus Average Effects By Matej Bajgar; Suren Karapetyan
  12. Business Education and Individual Portfolios: Evidence from Norwegian College Openings By Carpena, Fenella; Modalsli, Jørgen; Zhang, Dan
  13. Prioritizing Social-Emotional Competencies for Adolescent Well-Being: Evidence from Vietnamese Lower Secondary Schools By Chuyen Thi NGUYEN
  14. Wirkt intensive Beratung nachhaltig? Soziale Ungleichheit im Bildungsweg 4, 5 Jahre nach dem Abitur By Pietrzyk, Irena; Schneider, Juliana; Erdmann, Melinda; Helbig, Marcel; Jacob, Marita
  15. Occupational Licensing of Uber Drivers By Jonathan Hall; Jason Hicks; Morris M. Kleiner; Yun taek Oh
  16. Accessibility to Higher Education and Inter-regional Mobility By Se-il Mun; Shun Fujita; Lei Qin
  17. ‘Why should students learn a skill that will become redundant?’: Teaching Quantitative Social Science in an AI era By Hume, Ross John
  18. Revisiting test score forecasts and teacher value-added By Ahn, Tom; Aucejo, Esteban; James, Jonathan
  19. AI as Shared Business Infrastructure: Firm Growth and the Distribution of Gains By Panle Jia Barwick; Lu Fang; Jinhan Liu; Huilan Xu; Zhe Yuan
  20. Severance pay reforms, worker flows, and labor market duality By Andrade, Gabriel; Ferreira, Priscila; de Almeida Vilares, Hugo
  21. Engines of Change: Automation, Wages, and Labour Demand in Slovakia’s Automotive-Centred Economy By Matej Belin
  22. Education providers’ take-up of government grants: Evidence from a grant allocation discontinuity By Rosenqvist, Olof; Sauermann, Jan
  23. Pregnancy and Interrupted Schooling: Evidence from Monthly Histories By Jonghoon Park; R. Alan Seals
  24. Financial Pessimism and the Timing of Saving By Alexander Hansak
  25. Whose values does the HDI reflect? Evidence from a discrete choice experiment in Tanzania By Shaun Da Costa;; Koen Decancq;
  26. Who Responds to Intermittent Energy Nudges? Heterogeneity by Socioeconomic Status and Female Education in Armenia By Yermone Sargsyan; Salim Turdaliev; Silvester van Koten
  27. Beyond the degree: social class and meritocratic conversion in elite public employment By Fercovic Cerda, Malik; Danus, Bárbara
  28. Aggregating Epigenetic Clocks to Study Human Capital Formation By Giorgia Menta; Pietro Biroli; Divya Mehta; Conchita D’Ambrosio; Deborah Cobb-Clark
  29. Pipeline vs. choice: the global gender gap in STEM applications By Ahimbisibwe, Isaac; Altjmed, Adam; Artemov, Gregory; Barrios-Fernandez, Mauricio Andres; Bizopoulou, Aspasia; Kaila, Martti; Liu, Jin-Tan; Megalokonomou, Rigissa; Montalban, Jose; Neilson, Christopher; Otero, Sebastian; Sun, Jintao; Ye, Xiaoyang

  1. By: Blanden, Jo; Cassagneau-Francis, Oliver; Macmillan, Lindsey; Wyness, Gillian
    Abstract: Inequality in elite college attendance is a key driver of intergenerational mobility. This paper shifts the focus upstream to examine how elite high school attendance - specifically, enrollment in UK private, fee-paying schools - shapes university destinations across the academic ability distribution. Using linked administrative data, we show that the main advantage conferred by private schools is not that their highachieving students are more likely to access elite degree courses, but rather that their lower-achieving students are more likely to 'overmatch' by attending more selective degree courses than might be expected given their grades. In particular, we show that lower attaining pupils from fee-paying high schools enrol in university courses around 15 percentiles higher ranked than similarly qualified state school students. The greater propensity of private school students to overmatch is driven largely by differences in application behavior, with even the weakest private school students aiming higher than their higher achieving state school peers.
    Keywords: higher education;educational economics;college choice;mismatch;private schools
    JEL: I22 I23 I28
    Date: 2025–08–07
    URL: https://d.repec.org/n?u=RePEc:ehl:lserod:141559
  2. By: Ciprian Domnisoru; Tuomas P. Heikkinen
    Abstract: Have concerns about the potential job displacement effects of artificial intelligence affected the education choices of students? We analyze administrative data on the university applications and course choices of Finnish students. We find, after the launch of ChatGPT 3.5 in November 2022, a persistent drop in applications to fields of study classified as more exposed to AI under several different measures. The drop is particularly large in computer science and other ICT fields. Examining the elective course choices of non-ICT students, we find declining enrollment in programming and algorithms courses, particularly among students who had not taken mathematics as part of their high school exit exam. At the same time, non-ICT students enrolled more in AI/ML courses, and this increase was relatively larger among men.
    Keywords: Artificial Intelligence, College Major Choice, Higher Education, Unemployment
    JEL: I21 J24 J64 O33
    Date: 2026–11
    URL: https://d.repec.org/n?u=RePEc:crm:wpaper:26293
  3. By: Flávio Cunha; Qinyou Hu; Andrea Salvati; Kenneth I. Wolpin; Rui Zeng
    Abstract: We evaluate JumpStart, a parent-directed school-readiness program run by a Houston-area school district, in a randomized trial spanning three cohorts. Being offered the program raises scores on a curriculum-aligned assessment by 7.6 percentage points and parental investment by 0.26 standard deviations. A model of skill formation with endogenous inputs and program participation yields an average treatment effect of completing the program of about 9 percentage points, and attributes roughly a quarter of this effect to higher measured investment. Nonparametric bounds corroborate the average effect, and a mediation analysis that relaxes part of the model's structure supports the decomposition.
    Keywords: Early Childhood Development, Human Capital Development
    JEL: C93 I21 I24 I28 J13 J24
    Date: 2026–11
    URL: https://d.repec.org/n?u=RePEc:crm:wpaper:26291
  4. By: Zimmermann, Klaus F.; Karabulut, Gökhan; Keklik, Zeynep; Altıntaş, Kıvanç
    Abstract: Higher levels of anxiety about artificial intelligence (AI) among older and less-educated adults have now been documented in large population-based samples from high-income countries. Much less is known about how these differences arise, whether age and education interact, and whether occupational position helps understand them, particularly in middle-income economies. It evaluates the measurement structure of the scale, tests flexible functional forms and age-education interactions, and decomposes both gradients by occupational skill level, task class, and routine-task intensity. AI anxiety increases approximately linearly with age, whereas its association with education is strongly negative and nonlinear, with the largest gap between the lowest and highest education categories. Age and education largely show additive associations with AI anxiety. The age gradient is weaker among university graduates, although evidence for an interaction is limited for the total score and more evident for several subdimensions. Among the 409 employed respondents, occupational skill level and task class substantially reduce the education gradient but leave the age gradient unchanged: labor-market position helps explain educational but not age differences. Routine-task intensity contributes little beyond broad occupational categories, while professionals and managers report relatively high anxiety conditional on education and task class. Because the cross-sectional design precludes causal and age-period-cohort interpretations, the estimates are conditional associations. Even so, the results show that AI anxiety is not randomly distributed: it is systematically structured by human capital, age, and occupational position in a large middle-income economy.
    Keywords: artificial intelligence anxiety, human adaptation to AI, AI adoption, age and education gradients, occupational exposure, task content, human capital, technological change, automation, Turkey
    JEL: J24 O33 D91 J14 I21
    Date: 2026
    URL: https://d.repec.org/n?u=RePEc:zbw:glodps:1863
  5. By: Affonso, Agustina; Aucejo, Esteban; El Khoury, Stephanie; Larroucau, Tomas
    Abstract: We study whether information frictions about the relative effectiveness of study methods distort how college students allocate multidimensional learning efforts and, in turn, affect their academic performance. Our baseline survey shows that students widely disagree about which study habits are more effective, and many devote substantial time to passive learning strategies such as rereading, despite evidence that active retrieval (eg, self-quizzing) is more effective. We show that these students' beliefs, and consequently their associated choices, explain academic performance. To distinguish biased beliefs from efficient heterogeneity, we randomize treated students into general or personalized feedback. We find that personalized feedback widens the perceived effectiveness gap between rereading and retrieval methods by 42 percent, shifts time toward more effective practices, and raises subsequent exam scores by 0.05-0.08 standard deviations. Finally, we develop and estimate a dynamic model of study efforts and convex allocation costs to quantify and shed light on the mechanisms behind the intervention. For students with the weakest baseline beliefs, the personalized treatment is equivalent to a 16.5 percent reduction in the cost of adopting more effective methods.
    JEL: N0
    Date: 2026–09–28
    URL: https://d.repec.org/n?u=RePEc:ehl:lserod:141785
  6. By: Paul Bingley; Lorenzo Cappellari; Marco Ovidi; Matteo Sandi
    Abstract: Crime can disrupt victims’ working lives and family circumstances, potentially affecting the environment in which their children develop. We provide causal evidence that non-domestic parental victimization harms children’s human capital. Using Danish registers, we implement a matched difference-in-differences design that exploits victimization timing relative to the end of compulsory schooling. Violent victimization reduces mathematics scores by 0.03 standard deviations on average, reaching 0.09 following paternal victimization in early childhood, while property victimization has no detectable effect. Violent victimization of mothers disrupts educational progression, particularly for daughters. Sons experience lower well-being, while daughters report greater perceived safety and have lower rates of police-recorded victimization. Achievement losses are larger in families predicted to experience weaker parental labor-market attachment, a greater likelihood of parental charges for violent crime, and greater relational or residential instability.
    Keywords: parental criminal victimization, intergenerational spillovers, children’s human capital, early childhood
    JEL: K42 I24 J13
    Date: 2026
    URL: https://d.repec.org/n?u=RePEc:ces:ceswps:_13141
  7. By: Jalalian, Samuel; Pfeffer, Fabian T. (LMU Munich)
    Abstract: Grade retention is a common but costly educational practice and studies have repeatedly shown that, in many contexts, it offers no academic benefit to retained students. A remaining justification for its continued use is the presumed benefit for the promoted peers from whose class the retained student was removed. This study tests that assumption using longitudinal data from the German National Educational Panel Study (NEPS) and a difference-in-differences approach, focusing on grades and competencies from Grades 1 to 9. Results predominantly yield null findings, with a small positive effect on mathematics grades in lower secondary education (0.08–0.09 SD) as the only exception. Overall, our findings do not provide support for presumed peer benefits as a justification for grade retention.
    Date: 2026–10–29
    URL: https://d.repec.org/n?u=RePEc:osf:socarx:thqm9_v1
  8. By: Grover, Minali; Sharma, Ajay
    Abstract: The paper looks at the Fee Reimbursement Scheme (FRS) which is a conditional higher education scheme reimbursing tuition fees for scheduled caste (SCs), scheduled tribes (STs), and other backward classes (OBCs) along with economically weaker sections students. The policy was implemented in 2008 in Andhra Pradesh (a state in India) with the aim to increase access to higher education to disadvantaged social groups. This study analyses the impact of FRS on educational and labour outcomes. Using nationally representative individual-level data, we employ triple difference methodology to estimate the causal impact of FRS. Overall, we find two main results. First, there is an increase in likelihood of graduation for the treated cohort with a decrease in the private tuition fee expenditure as the potential mechanism. Secondly, looking at the potential labour market entry age-cohort, the findings indicate no improvement in the labour market outcomes. The two suggestive mechanisms identified are over-education and out-migration for the treated cohort. The results are validated using alternate method using semi parametric difference-in-difference (SDID) method, falsification tests and other robustness analysis. The study highlights how conditional higher education intervention not necessarily improve employment opportunities.
    Keywords: education, subsidy, India, fee reimbursement, employment
    JEL: I23 I20 I28 J24
    Date: 2026
    URL: https://d.repec.org/n?u=RePEc:zbw:glodps:1862
  9. By: Ana M. Gazmuri; Elena Prager
    Abstract: This paper studies how exposure to peers affects college application behavior. We use neighbors who attend different high schools to mitigate concerns about common shocks. Using granular classroom-level data, we find that exposure to peers who have applied to a program increases application rates. Consistent with an information mechanism, there is no additional effect of peers who are admitted. In counterfactuals, the effect is large enough to close urban-rural gaps and gender gaps in applications to selective college programs. We make the (generally NP-hard) application portfolio choice problem tractable using a setting where admission probabilities are observable and conditionally independent.
    JEL: I20 I23 I24
    Date: 2026–11
    URL: https://d.repec.org/n?u=RePEc:nbr:nberwo:35993
  10. By: Borusyak, Kirill (University of California Berkeley); Dix-Carneiro, Rafael (Duke University); Kovak, Brian (Carnegie Mellon University)
    Abstract: Abstract A commonly used migration regression relates population changes to local labor demand shocks, often yielding small or statistically insignificant coefficients despite sizable effects on labor market outcomes. We argue that such findings need not imply that mobility is unresponsive to local economic conditions. Rather, conventional migration regressions can be biased, as they do not reflect how workers choose among competing locations. In a parsimonious equilibrium model, local population growth depends on differences between local shocks and those in relevant destinations; when these shocks are correlated, estimated responses are attenuated. Moreover, with industry-level shocks, standard shift-share measures misrepresent locations’ effective exposure, leading to further attenuation. Using simulation experiments based on Brazilian and German administrative data, we show attenuation can exceed 50% and that conventional regression estimates can be highly imprecise. We propose theory-grounded empirical approaches that correct the biases, deliver estimates with a clear structural interpretation, and improve the precision of estimated mobility responses.
    Keywords: internal migration, local labor markets, labor demand shocks, shift-share designs, spatial spillovers, sectoral mobility
    JEL: R23 J61 C21 F16
    Date: 2026–10
    URL: https://d.repec.org/n?u=RePEc:iza:izadps:dp19019
  11. By: Matej Bajgar; Suren Karapetyan
    Abstract: We examine whether competitive research grants generate independent research by the funded investigator. We distinguish first- and last-author publications—indicating leading contribution in biomedical research—from total output. Using data on Czech medical research grants awarded between 2015 and 2019, we combine a regression discontinuity (RD) design around the funding cutoff with augmented inverse probability weighting (AIPW) that estimates average effects across a broader applicant sample. RD estimates show that grants increase total publications by 3.1 papers (22%). First- and last-author publications increase by 59% and 41% respectively—substantially exceeding the total effect. These gains extend to hightier journals and to citations. AIPW estimates are qualitatively consistent but smaller in magnitude. Within AIPW, first- and last-author effects are larger for funded projects closer to the cutoff, supporting heterogeneity in returns to funding. Exploratory subgroup estimates suggest comparatively large local effects for female and less-senior investigators, although the smaller subgroup samples and differing AIPW patterns mean that these results should be interpreted cautiously.
    Keywords: Regression discontinuity design, research funding, scientific productivity
    JEL: O38 O30 I23
    Date: 2026–10
    URL: https://d.repec.org/n?u=RePEc:cer:papers:wp824
  12. By: Carpena, Fenella; Modalsli, Jørgen; Zhang, Dan
    Abstract: We study whether access to business education changes portfolio choice. Exploiting the staggered establishment of two-year business programs at Norwegian regional colleges, we compare cohorts young enough to attend a newly opened local college with slightly older cohorts in the same municipality, relative to the same contrast in never-treated municipalities. Using administrative register data and staggered difference-in-differences, we find that access to a business college increased mid-life participation in risky financial assets and raised the risky share of financial wealth, with a corresponding decline in the safe share. Effects on income or debt appear limited. These results are consistent with a financial-knowledge channel: exposure to even short business programs durably changes portfolio risk-taking without materially altering income and leverage.
    Date: 2026–10–19
    URL: https://d.repec.org/n?u=RePEc:osf:socarx:4znt8_v1
  13. By: Chuyen Thi NGUYEN (Hanoi National University of Education, Vietnam Author-2-Name: Son Duc NGUYEN Author-2-Workplace-Name: Hanoi National University of Education, Vietnam Author-3-Name: Thanh Hai NGUYEN Author-3-Workplace-Name: Hanoi National University of Education, Vietnam Author-4-Name: Author-4-Workplace-Name: Author-5-Name: Author-5-Workplace-Name: Author-6-Name: Author-6-Workplace-Name: Author-7-Name: Author-7-Workplace-Name: Author-8-Name: Author-8-Workplace-Name:)
    Abstract: " Objective - This study examines which social-emotional learning (SEL) competencies make the largest relative contributions to adolescent well-being, in order to inform the prioritization of school-based SEL interventions in Vietnam. Methodology/Technique - Grounded in the Collaborative for Academic, Social, and Emotional Learning (CASEL) competency framework and Seligman's PERMA model of well-being, a cross-sectional survey was conducted with 486 lower secondary school students (Grades 6–9) in Hanoi, using the Vietnamese versions of the Social Emotional Competence Questionnaire and the PERMA-Profiler. Data were analyzed using correlation analysis, nested and hierarchical multiple regression, and relative weight analysis. Findings - These findings suggest that school-based SEL programs in Vietnam should prioritize self-management and relationship skills to maximize students' psychological well-being and provide evidence-based guidance for educational resource allocation. Results - Results showed that all five SEL competencies positively predicted well-being, but their contributions differed substantially. Self-management emerged as the strongest predictor, followed by relationship skills, whereas social awareness demonstrated the weakest effect. The competency-based model explained 38% of the variance in well-being, exceeding the explanatory power of the overall SEL model. After controlling for demographic characteristics, the prioritization pattern remained stable. Type of Paper - Empirical"
    Keywords: social-emotional competencies, student well-being, SEL, lower secondary school students, Vietnam, PERMA
    JEL: I21 I31 J24
    Date: 2026–09–30
    URL: https://d.repec.org/n?u=RePEc:gtr:gatrjs:gjbssr680
  14. By: Pietrzyk, Irena; Schneider, Juliana; Erdmann, Melinda; Helbig, Marcel; Jacob, Marita
    Abstract: In der Schule angebotene intensive Beratungsprogramme erhöhen nachweislich die Studienaufnahme benachteiligter Schüler:innen. Doch ist bislang unklar, ob sich diese Effekte auch langfristig halten und wie sie sich auf herkunftsbedingte Unterschiede in den Bildungswegen auswirken. Auf Basis einer randomisierten kontrollierten Studie haben wir die nachschulischen Bildungswege von Schüler:innen 4, 5 Jahre nach dem Abitur untersucht und heterogene Treatment-Effekte für Schüler:innen niedriger und hoher sozialer Herkunft geschätzt. Bei benachteiligten Schüler:innen erhöhte die Beratung die Wahrscheinlichkeit eines Hochschulwegs um 7, 5 Prozentpunkte. Bei privilegierten Schüler:innen verschob sie die Bildungswege in Richtung berufliche Ausbildung, auch wenn dieser Effekt über verschiedene Modellspezifikationen hinweg weniger stabil ausfiel. Diese gegenläufigen Effekte führten dazu, dass sich die herkunftsbedingten Unterschiede in den nachschulischen Bildungswegen deutlich verringerten. Die Studie liefert damit langfristige experimentelle Evidenz, die über die Betrachtung der Studienaufnahme hinausgeht.
    Abstract: High school counseling has been shown to increase higher education enrollment among disadvantaged students. However, it remains unclear whether these effects persist in the long run and how they shape origin-based differences in educational pathways. Drawing on a randomized controlled trial, we examine students' post-secondary educational pathways 4.5 years after high school graduation and estimate heterogeneous treatment effects for students from low and high social origins. Among disadvantaged students, counseling promoted higher education outcomes by 7.5 percentage points. Among privileged students, it shifted educational pathways toward vocational education and training (VET), although this effect was less stable across model specifications. These opposing effects led to a substantial reduction in origin-based differences in post-secondary educational pathways. The study thus provides long-term experimental evidence that moves beyond initial enrollment effects.
    Keywords: High school counseling, Educational inequality, Long-term effects, Social origin, Experiment, Beratung, Bildungsungleichheit, Langzeiteffekte, soziale Herkunft, Experiment, randomisiert kontrollierte Studie
    Date: 2026
    URL: https://d.repec.org/n?u=RePEc:zbw:wzbslm:344523
  15. By: Jonathan Hall (Amazon, Inc.); Jason Hicks (University of Victoria); Morris M. Kleiner (University of Minnesota Hubert H. Humphrey School of Public Affairs and NBER); Yun taek Oh (University of Nevada - Reno)
    Abstract: We examine whether occupational licensing improves service quality and safety using trip-level Uber data that include driver ratings and telematics-based measures of driving behavior. Exploiting quasi-random assignment from proximity-based dispatch, we compare trips served by licensed and unlicensed drivers in two settings: a crossborder comparison between New York City and New Jersey, and a deregulation event in Houston. Across settings and specifications, including instrumental variable estimates, we find no consistent evidence that licensing improves consumer outcomes. In Houston, post-deregulation entrants are indistinguishable from previously licensed drivers on ratings and driving behavior, despite differing markedly in experience and age.
    Date: 2026–08
    URL: https://d.repec.org/n?u=RePEc:vic:vicddp:2603
  16. By: Se-il Mun; Shun Fujita; Lei Qin
    Abstract: We develop an equilibrium model of the undergraduate higher education market that incorporates interregional mobility for enrollment. An individual chooses a college to attend, seeking to maximize utility depending on the quality of education and the costs of attendance. Costs include tuition fees, the effort required to pass the entrance exam, and moving costs from one's hometown to the region where the college is located. Each college admits students whose entrance exam scores exceed a certain threshold, which we call the college's qualifying score. In our model, the qualifying score of each college is determined endogenously as the shadow price of the constraint that the number of enrolled students must not exceed capacity. We structurally estimate the model's parameters by simulating equilibrium using Japanese data. Based on empirical results and counterfactual experiments, we evaluate the welfare costs of disparities in access to higher education due to uneven spatial distribution of colleges among regions.
    Date: 2026–10
    URL: https://d.repec.org/n?u=RePEc:arx:papers:2611.00630
  17. By: Hume, Ross John (University of Edinburgh)
    Abstract: This report comprises a review of the state of the art concerning Generative AI’s impact upon quantitative social science education. Insights are combined with results of a survey of undergraduate social science students to suggest initial recommendations and avenues for further discussion. While policy recommendations are based in the context of the University of Edinburgh, the literature review and analysis presented are applicable to teaching statistics and data science in general, particularly through the lens of social science. Findings suggest that while broadly current AI models remain a procedural tool and less reliable for methodological judgements, the educational impact of AI is largely dependent upon its use cases and the educational objectives set. Students expressed different perspectives from wanting to avoid the technology to wanting to gain AI-enabled skills for their future career. Findings reflect a broader question about the purpose of university education in an AI world. However, students broadly agree on a desire for clear-cut AI policies to follow rather than simply more restrictive or permissive ones.
    Date: 2026–10–19
    URL: https://d.repec.org/n?u=RePEc:osf:socarx:mj57s_v1
  18. By: Ahn, Tom; Aucejo, Esteban; James, Jonathan
    Abstract: We develop a framework for comparing forecasts in teacher value-added models. We apply it to interpret the teacher-switching test of Chetty, Friedman, and Rockoff (2014), showing that the teacher-switching coefficient cannot be mechanically interpreted as a test of forecast unbiasedness: the correct null may differ from one, and a coefficient of one is insufficient for forecast unbiasedness. We also analyze forecast choice for ranking teachers and estimating teachers' effects on long-term outcomes. Although forecasts including current-year data may contain contemporaneous classroom shocks, they contain more teacher signal than leave-year-out forecasts, which improves rankings and may increase precision in long-term-outcome regressions.
    Keywords: teacher value added;forecast construction;forecast aggregation;forecast unbiasedness;teacher rankings
    JEL: I21 J24 C53 C52
    Date: 2026–09–28
    URL: https://d.repec.org/n?u=RePEc:ehl:lserod:141783
  19. By: Panle Jia Barwick; Lu Fang; Jinhan Liu; Huilan Xu; Zhe Yuan
    Abstract: Generative AI improves performance on individual tasks, but whether those gains translate into firm growth, and for which firms, remains an open question. We study a leading Chinese e-commerce platform that in 2024 embedded a generative AI system in the operating software firms use to run their businesses, drawing on centrally developed models and platform-wide data to deliver ready-to-use applications to every firm. Complete records of each firm’s AI use, linked to its monthly operations and sales, let us trace adoption through changes in business capabilities, such as reaching customers and introducing new products, to sales growth. Using a matched difference-in-differences design and focusing on firms with annual sales below RMB 5 million, we find that adoption raises monthly sales by 51.6%, primarily through a 47.0% increase in the number of orders; instrumental-variable estimates confirm these results. Adopters reach more customers, update their product lines faster, and deliver better product and service quality, with the largest proportional gains for the smallest firms. Two mechanisms help explain the results. First, AI lowers the cost of routine work: adopters produce more product content and change listings more often while spending less staff time per order, and the gains are largest for firms that faced the highest costs relative to sales before adoption. Second, AI reduces information frictions: adopters post richer product information, and the gains are largest in markets where demand is hard to predict and matching quality is poor, especially for small firms. At the market level, product categories with greater pre-existing information frictions see more sales by existing firms, more entry, and more new products. AI expands markets rather than merely shifting sales from non-adopters to adopters.
    JEL: D22 L25 L81 L86 O33
    Date: 2026–11
    URL: https://d.repec.org/n?u=RePEc:nbr:nberwo:36003
  20. By: Andrade, Gabriel; Ferreira, Priscila; de Almeida Vilares, Hugo
    Abstract: This paper studies how cuts to statutory severance pay affect worker flows, contractual allocation, firm-level hiring composition, and wages in the Portuguese labor market. The analysis exploits the 2011-2013 reforms, which reduced the transfer component of dismissal protection for both fixed-term and open-ended contracts while leaving most procedural safeguards intact. Using matched employer-employee data for 2010?2019, we combine sharp legal-timing contrasts around the November 2011 and October 2013 cutoffs with an exposure-based IV/control-function design that leverages heterogeneity in admission cohort, tenure, contract type, and grandfathered rights. The results show that lower severance pay does not uniformly improve reallocation. Around the November 2011 cutoff, separations shift away from exits from private-sector employment toward job-to-job moves, whereas the October 2013 cutoff does not reproduce this pattern, instead displaying weaker job-to-job responses and a more uneven reduction in exits. In the broader exposure estimates, lower severance entitlements are associated with higher turnover, but the composition of that turnover varies across worker groups and contract types, combining selective conversion from fixed-term to open-ended jobs with increased reliance on temporary contracts. Within continuing worker-firm matches, wage-severance associations are economically small, providing little evidence that lower severance rights were offset by higher wages. Overall, the reforms reshaped turnover and contract use without delivering uniform improvements in labor-market reallocation, and temporary-permanent duality remains a central feature of the post-reform labor market.
    Keywords: severance pay reforms;employment protection legislation;worker flows;wages;labor-market duality
    JEL: J65 J63 J41 J23 J31
    Date: 2026–11–05
    URL: https://d.repec.org/n?u=RePEc:ehl:lserod:141899
  21. By: Matej Belin
    Abstract: This paper estimates the effects of robot adoption on wages and employment in Slovakia, one of the most automation-exposed economies in Europe. Using matched employer-employee administrative data combined with survey-based measures of occupational task exposure, we identify the impact of automation through an nstrumentalvariables strategy that exploits variation in robot adoption from major non-European adopters. Across specifications, we find no evidence of economically meaningful effects of robot adoption on labour-market outcomes. The estimated impacts on wages and job separations are small and precisely estimated around zero. The Slovak estimates point to increased productivity among robot-adopters, which, in turn, creates new employment opportunities rather than displacing human labour.
    Date: 2026–06
    URL: https://d.repec.org/n?u=RePEc:cer:papers:wp821
  22. By: Rosenqvist, Olof (IFAU - Institute for Evaluation of Labour Market and Education Policy); Sauermann, Jan (IFAU - Institute for Evaluation of Labour Market and Education Policy)
    Abstract: In decentralized school systems where core funding comes from local governments, education providers can sometimes be entitled to central government grants on top of the core funding (e.g. Title I in the US). In Sweden, government grants that public and private education providers can claim have grown over time; from around 3 % of compulsory school costs in 2010 to 7 % in 2023. The grants, which typically are tied to specific purposes, are a way for the government to exercise control over various school policies in a decentralized system. However, for the grants to have an impact, the providers must claim their entitlements which requires submitting a detailed plan for fund usage and not all providers do this, especially not smaller private providers. We study how the size of the grant entitlement affects providers’ take-up decisions. This is a margin that readily can be adjusted by e.g. merging smaller grants to a single larger. We exploit variation from a large grant in Sweden that allocated substantially more money to providers with a socioeconomic student composition below a certain threshold value. Using a regression discontinuity design, we find that take-up among private providers increased from 70 % to 88 % as the grant amount increased from about 4 % to 13 % of the total cost per student. Importantly, the effect on take-up appears to be sticky as it remains in a later year when the compensatory part was removed from the grant and all providers were entitled to the same amount per student.
    Keywords: Education provider; government grants; take-up; grant size; discontinuity
    JEL: H81 I22 I28
    Date: 2026–10–12
    URL: https://d.repec.org/n?u=RePEc:hhs:ifauwp:2026_023
  23. By: Jonghoon Park; R. Alan Seals
    Abstract: Pregnancy can disrupt schooling before infant care begins. We use monthly histories from the National Longitudinal Survey of Youth 1997 to study school interruptions during pregnancy and after childbirth and whether students later return to school or earn a diploma or GED. We jointly model conception and sustained interruptions in enrollment, allowing persistent unobserved differences to affect both processes. Under the timing-of-events identifying assumptions, pregnancy raises the monthly interruption rate to nearly four times the rate without pregnancy. Interruption rates are elevated before delivery and remain elevated afterward. After an interruption, students already exposed to pregnancy return to school and earn a diploma or GED less often within two years. These findings show that schooling disruption is already present during pregnancy, before the additional demands of infant care begin.
    Date: 2026–10
    URL: https://d.repec.org/n?u=RePEc:arx:papers:2610.14317
  24. By: Alexander Hansak (University of Graz, Austria)
    Abstract: Does financial pessimism encourage precautionary saving or discourage households from saving? I show that these apparently conflicting interpretations can arise from the same financial forecast, depending on when saving is measured. Using British Household Panel Survey data from 1991-2009, I compare the conditional association between year-ahead financial expectations and reported saving participation at the forecast interview and at the following interview. Both estimates use identical respondent-interview pairs, controls, fixed effects, and longitudinal weights. Among respondents aged 18-34, expecting worse rather than unchanged finances is associated with a 4.19-percentagepoint higher probability of saving at the forecast interview, but a 5.15-point lower probability at the following interview. The directly estimated difference is -9.35 percentage points (95% confidence interval [-12.30, -6.40]). The timing contrast is also negative, though smaller, among respondents aged 35-59. Later Understanding Society data provide a less directly comparable and less precise extension. Supplementary analyses reveal a similar sign reversal in reported monthly saving amounts and show that pessimistic forecasts predict less favourable subsequent gross-income changes. In descriptive specifications, conditioning on observed income changes and entry into unemployment leaves the timing contrast largely unchanged. These findings highlight an important distinction: financial expectations may reflect both households’ motivations for saving and private information about their future economic circumstances. The evidence documents a conditional timing regularity rather than a causal effect of pessimism on saving, underscoring why expectations-saving associations should not be interpreted independently of their observation horizon.
    Keywords: financial expectations, saving participation, household finance, survey timing, panel data
    JEL: D14 D84 E21 G51
    Date: 2026–10
    URL: https://d.repec.org/n?u=RePEc:grz:wpaper:2026-25
  25. By: Shaun Da Costa;; Koen Decancq;
    Abstract: Multidimensional development indices such as the Human Development Index (HDI) embed value judgements about trade-offs between dimensions. Yet these are rarely tested against the valuations of the people they describe. We use a pictogram-assisted discrete choice experiment with 361 respondents in Tanzania’s Dodoma region to elicit trade-offs between income, schooling, and life expectancy. We estimate preferences using the 264 respondents who make compensatory choices and compare their valuations with the equal-weight trade-offs embedded in the HDI. The typical respondent values an additional year of schooling more highly than an additional year of life expectancy, consistent with the HDI’s qualitative ranking of dimensions. However, the implied schooling-life expectancy value ratio is around 1.5, compared with 7.2 under the HDI at Tanzania’s observed achievement levels. Allowing for observed and unobserved preference heterogeneity reveals substantial variation in valuations across respondents, which provides little support for treating any single weighting scheme as representative of this population’s valuations. The results show that the value judgements embedded in multidimensional indices can be tested empirically, and that even in a small, geographically concentrated sample the trade-offs implied by the HDI differ markedly from respondents’ valuations.
    Date: 2026–10
    URL: https://d.repec.org/n?u=RePEc:hdl:wpaper:2604
  26. By: Yermone Sargsyan; Salim Turdaliev; Silvester van Koten
    Abstract: Using a long-term randomized field experiment in Yerevan, Armenia, we study heterogeneous household responses to behavioural energy reports in a post-Soviet transition-economy setting. Households were assigned to a control group, a consumption-comparison report arm, or a consumption-plus-cost report arm that added monetary bill information to the same peer comparisons. Reports were sent monthly during two report-delivery periods separated by a prolonged pause. Average treatment effects mask substantial heterogeneity by socioeconomic status (SES), female education, baseline electricity use, and electric heating. High-SES households respond differently to the added monetary information than low-SES households, while households with a tertiary-educated female respondent exhibit more conservationoriented responses. Both patterns are clearest during report-delivery months under the consumption-plus-cost reports. Households using electricity for heating show strong conservation-oriented responses, whereas low-baseline users can exhibit boomerang-type increases under consumption-plus-cost reporting. Responses during pause months also differ across groups, implying that intermittent reporting is not uniformly beneficial but depends on subgroup targeting and report content.
    Keywords: demand-side management, intermittent nudges, peer comparison, socioeconomic status, principal component analysis, heterogeneous treatment effects, post-Soviet economy, developing country, Armenia
    JEL: Q41 Q48 Q58 C93 D91
    Date: 2026–09
    URL: https://d.repec.org/n?u=RePEc:cer:papers:wp823
  27. By: Fercovic Cerda, Malik; Danus, Bárbara
    Abstract: To what extent can higher education equalise opportunities and career outcomes within elite organisations? While research documents ‘class ceilings’ in elite private-sector organisations, less is known about how class inequalities operate within meritocratic public institutions, particularly in the Global South. Drawing on survey data and 33 indepth interviews, this paper examines a highly professionalised elite public organisation in Chile. It shows that formal meritocratic procedures coexist with stratified professional pathways: recruitment is concentrated among graduates of the country’s two most prestigious universities, while internships, international education, English proficiency, professional networks, and classed cultural resources shape access and advancement. Although the workforce is gender-balanced, women remain underrepresented in senior positions. We argue that meritocratic conversion—the process through which socially unequal resources become organisationally legible and valued as merit—helps explain persistent class advantage under formal meritocratic conditions, limiting the equalising potential of higher education. Class bearing, or prestancia de clase, is one observable form of this conversion. Meritocracy thus reshapes, rather than eliminate, how privilege is expressed and recognised in elite public organisations.
    Keywords: ARRAY(0x620707231578)
    JEL: R14 J01
    Date: 2026–11
    URL: https://d.repec.org/n?u=RePEc:ehl:lserod:141989
  28. By: Giorgia Menta; Pietro Biroli; Divya Mehta; Conchita D’Ambrosio; Deborah Cobb-Clark
    Abstract: Early-life environments can influence human capital and biological aging long before their consequences are visible in conventional outcomes. By aggregating age-related epigenetic markers, epigenetic ‘clocks’ have become the leading tool for studying biological aging. We develop a novel integrated measure – the Multi EpiGenetic Age (MEGA) clock – that combines existing clocks to reduce measurement error and improve efficiency. Using three empirical applications, we show that accelerated epigenetic aging in adolescence predicts worse educational, mental health, and labor market outcomes; that child maltreatment increases epigenetic aging; and that delayed school entry causally accelerates aging by age seven, especially among disadvantaged children. Together, these results show how an integrated measure of epigenetic aging can be systematically incorporated into economic models to study the dynamics of health and human capital formation.
    Keywords: epigenetic clocks, DNA methylation, child abuse, human capital, ALSPAC data
    JEL: I12 I14 J24
    Date: 2026
    URL: https://d.repec.org/n?u=RePEc:ces:ceswps:_13132
  29. By: Ahimbisibwe, Isaac; Altjmed, Adam; Artemov, Gregory; Barrios-Fernandez, Mauricio Andres; Bizopoulou, Aspasia; Kaila, Martti; Liu, Jin-Tan; Megalokonomou, Rigissa; Montalban, Jose; Neilson, Christopher; Otero, Sebastian; Sun, Jintao; Ye, Xiaoyang
    Abstract: Women account for only 35% of global STEM graduates, a share unchanged for a decade. We use administrative microdata from centralized university admissions in ten systems to deliver the first cross-national decomposition of the STEM gender gap into a pipeline gap (academic preparedness) and a choice gap (first-choice field conditional on eligibility). In deferred-acceptance platforms where eligibility is score-based, we isolate preferences from access. The pipeline gap varies widely, from -19 to +31 percentage points across education systems. By contrast, the choice gap is remarkably stable: high-scoring women are 25 percentage points less likely than men to rank STEM first.
    Keywords: gender inequality;centralized application platforms;STEM gender gap
    JEL: I23 N30
    Date: 2025–08–27
    URL: https://d.repec.org/n?u=RePEc:ehl:lserod:141543

This nep-edu issue is ©2026 by João Carlos Correia Leitão. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at https://nep.repec.org. For comments please write to the director of NEP, Marco Novarese at <director@nep.repec.org>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the Griffith Business School of Griffith University in Australia.