Research Seminars

This research explores how mobility big data can be used to understand urban travel behavior and evaluate policy implications. The data sources include public transport smart cards, public bike systems, and mobile phone data. In particular, the transport card–based analysis demonstrates the benefits of Seoul’s integrated fare system with free transfers. The presentation will also briefly introduce data from autonomous delivery robots deployed in Seoul as a potential emerging urban data source.

The death of a child is a devastating and life-altering event for the entire family, extending beyond the initial grief and shaping the short and long-term life trajectories of the surviving siblings and parents. Although a growing literature documents its negative impact on parental outcomes, very little is known about its causal impact on the human capital accumulation of surviving siblings. This paper examines the effect of losing a sibling during childhood on educational outcomes, using detailed register data from the entire population of Finland. I exploit the timing of an unexpected sibling loss relative to the time of 9th-grade GPA measurement. I find that losing a sibling 2 years before the 9th grade has a negative impact of 19% of a standard deviation on the 9th-grade GPA. The effect is more pronounced and prevalent across different ages at the time of sibling loss for children with a lower socioeconomic background. Further, I find a 12-14 percentage points decrease in the probability of general track choice in the upper-secondary school following a sibling loss. Exploring potential mechanisms, I find notable negative mental health effects on both surviving children and their parents, along with adverse effects on the labor market outcomes for mothers. These findings highlight the large social and economic consequences of such losses, emphasizing the need for targeted policies to support bereaved family members.

We study how secondary markets for durable goods interact with consumers’ social-responsibility motives to mitigate environmentally harmful new production. On the positive side, secondary markets may allow responsible consumers to acquire used goods that would otherwise be discarded, reducing premature waste. On the negative side, secondary markets introduce two major harmful forces. First, the possibility of buying used goods and thereby causing less harm can raise the demand of responsible consumers, often increasing the production necessary to serve the market. Second, said demand can increase the price of used goods, encouraging purchases of new goods. These forces imply that if used goods have positive private consumption value, then secondary markets always erode the benefits of social responsibility. If, instead, used goods may have negative private value, then secondary markets can enhance or erode the benefits of social responsibility.

We study the effect of working from home (WFH) on health-related absenteeism. We draw on a monthly panel (2018–2023) of nearly 1.9 million workers insured under a major German statutory health insurance fund. Our identification strategy leverages the differential exposure to the unexpected shift to WFH in 2020, resulting in about one-quarter of German employees regularly working remotely today. Specifically, our difference-in-differences design compares sick leave take-up across workers with different WFH potential, i.e., the teleworkability of their occupation in February 2020. Our results imply a nontrivial lasting response to WFH. Compared to the pre-treatment mean, the monthly likelihood of sick leave take-up declined by 3.3% by 2023 due to a 10-percentage-point increase in WFH potential (comparable to a shift from performing and entertainment professions to media documentation and information services). These effects are driven by reductions in respiratory infections and musculoskeletal disorders, such as back pain. The response reflects not only behavioral changes in sick leave take-up, but also genuine health improvements.

This project examines how an exogenous decline in immigration flows induces sorting across the distribution of health conditions in the local labor market adjustment process and documents significant changes in the community health indicators. In the 1920s, a set of immigration laws in the United States imposed quotas based on national origin and restricted immigrant flows. These policies cut immigration flows from constrained countries to around 150,000 individuals a year. The effects of this policy change are identified in a difference-in-difference framework utilizing the LIFE-M, historical labor survey, and full count census data. Results indicate that individuals in areas facing larger declines in immigration are observed transitioning into occupations and industries with worse associated health measures. Increasing native-born employment in high health cost (HHC) positions offset lost immigrant labor leading to no change in HHC employment share. Further, policy exposure was associated with declining average lifespan and increasing mortality rates among counties’ working age US-born population.

Accurate perception of social relationships is central to human social cognition, yet little is known about how well individuals perceive incoming social ties within their natural social environments. We collected data from 3,077 adolescents across 117 classrooms and analyzed complete social networks comprising four directed relational layers—friends, best friends, enemies, and worst enemies—alongside adolescents’ predictions of who would nominate them in each layer. Across all layers, perceptual accuracy was strikingly low. Only 0.29% of students assessed all of their relationships correctly. Errors were overwhelmingly driven by underestimation: adolescents systematically failed to recognize how many peers named them, both positively and negatively. Misperception was structured rather than random. Popularity (in-degree) was strongly associated with underestimation, whereas sociability (out-degree) was associated with overestimation. These structural patterns were consistent across relational layers and error types and showed no systematic improvement with age. Together, these findings reveal robust structural asymmetries in social cognition, suggesting that young people perceive far less of their social worlds than there is.

 This paper investigates the effects of a price transparency policy on spatial competition in a retail fuel market. Using post-policy data, I examine how increased consumer-side transparency affects competition. I develop an empirical model of demand and supply that accounts for both fully informed consumers and those facing price uncertainty. Prior research shows that major brands coordinated before the policy and continued to do so afterward. Building on this evidence, I model coordination among major brands and allow its stability to depend on the share of informed consumers. I find that as more consumers become informed, price dispersion and profit margins decline. However, the extent of coordination among major firms remains stable. These findings suggest that while consumer-side transparency increases competitive pressure, its pro-competitive effects may be limited or even reversed if coordinating firms can also access the same transparency tools.

This paper examines how the origin of income mobility—whether driven by individual effort or luck—affects preferences for redistribution. The existing literature emphasizes how expectations of future income shape redistributive support, while abstracting from the mechanisms that generate income mobility. I study this question using a laboratory experiment in which individuals face identical income transition structures but differ in whether their mobility prospects are generated by relative effort or by luck. The results reveal that the mobility mechanism triggers significant heterogeneity across the income distribution. Among low-income individuals, effort-based mobility increases support for redistribution relative to luck-based mobility when upward mobility prospects remain insufficient to reach top-income positions. This pattern suggests that meritocratic processes can be perceived as procedurally unfair when advancement opportunities are limited, highlighting the political relevance of procedural features of income mobility. Conversely, high-income individuals tend to reduce their support for redistribution under effort-based mobility when they are more productive than the average. The paper also documents that biased beliefs about relative performance moderate redistributive preferences. Participants who are poorer than average but expect to become richer than average in the future, due to overconfidence in their relative performance, tend to support lower levels of redistribution. Overall, the findings indicate that preferences for redistribution depend not only on expected future income but also on the perceived legitimacy of the process through which mobility is generated.

This paper examines how prison inmates anticipate being stigmatized by the inmate label and whether a brief self-affirmation intervention can recalibrate these beliefs. In Study 1, we assess the perceptions of inmates (N = 297) about the extent of trust and altruism that will be given to them by general society by manipulating disclosure of their prison status in an incentivized lab-in-the-field experiment. The findings reveal that when their prison status was disclosed, inmates anticipated greater altruism, reflecting optimism rather than anticipated stigma, although their expectations of trust did not differ. We find no significant effects of self-affirmation on their beliefs about altruism, trust, or trustworthiness; nor on the proposed psychological channels of inmates’ confidence and risk taking. In Study 2 (N=489), we conceptually replicate optimism about altruism in a one-shot setting using incentivized spectator predictions, and identify the likely underlying mechanism as the perception that society will treat them favorably due to hardships they face during incarceration spells and uncertainty about the nature of their crime. Their expectations were more realistic for longer-horizon trust-based domains such as securing employment and housing, and for general kindness ratings. 

I provide new evidence on the labor market impacts of military conscription using employer-employee administrative data covering the universe of men born between 1958 and 1975 in Argentina. For identification, I exploit the random assignment of draft eligibility generated by 18 draft lotteries. I find that draft eligibility depresses early-career outcomes: At ages 20-24, eligible men exhibit lower employment and earnings. These penalties fade with age and do not accumulate into sizable long-run losses. Eligibility to conscription increases the likelihood of holding formal employment and a permanent contract, with no average effect on medium- or long-run earnings. The enduring consequence of conscription is a modest reallocation of employment away from agricultural activities and toward service-oriented and higher-paying sectors, reducing employment in the lowest-wage activities among the most geographically disadvantaged individuals. Overall, these results indicate that conscription can affect industry sorting rather than merely disrupt labor market entry in early adulthood, but the magnitudes are modest and do not translate into measurable gains in earnings.

Older workers are often viewed as obstacles to innovation, suggesting that their exit allows firms to reallocate resources toward new capital and technology. I argue instead that older, experienced workers support both the continuity of current production and the capacity to integrate new technologies into existing operations. The key empirical challenge is that when retirements are anticipated, firms have time to transfer knowledge internally, making the productivity value of older workers difficult to observe. I address this by studying a 2014 German pension reform that unexpectedly lowered the early retirement age for experienced workers by up to 29 months, inducing a sudden and unanticipated loss of long-tenured employees. Firms exposed to the reform reduce capital accumulation, delay technology adoption, and experience subsequent declines in revenue and value added, consistent with the erosion of firm-specific human capital. To interpret these findings, I develop a stylized model in which older workers transfer uncodified, firm-specific knowledge that is essential for maintaining legacy capital and integrating new technologies into firms’ operations. The model predicts, and the data confirm, that unexpected retirements weaken firms’ ability to sustain production and slow the pace of technological upgrading.

This paper studies the relationship between patient cost-sharing and the demand for care after hospital stays (post-acute care), and how this relationship is mediated by non-medical custodial care needs and the presence of the family. Using plausibly exogenous variation in the number of free skilled nursing facility (SNF) days at hospital discharge, I first show that higher cost-sharing for SNFs leads to substitution away from SNFs to home health and routine discharges with no care. Moreover, for those who are already in a SNF, discharges rise sharply when cost-sharing begins. I further document that people with family are less likely to use SNFs, while people with custodial care needs are more likely to use SNFs. I use these findings to motivate a dynamic structural model that begins at hospital discharge and follows patients through further discharge decisions in different post-acute settings. Using the model, I explore the implications of alternative cost-sharing structures. I find that eliminating the non-linearity in SNF copay can generate large Medicare savings at small costs to the patients. Medicare subsidizing custodial care at home enhances patient welfare substantially for individuals without family, at no additional cost to Medicare. For patients with family, the same policy, while still beneficial to the patient, will increase Medicare costs substantially, resulting in a net zero effect overall.

We assess the welfare impact of the introduction of universal daycare services in Qu´ebec in 1997. Unlike the standard sufficient-statistic metric, which assumes marginal changes in fiscal policy, our approach accounts for the non-marginal nature of the program and quantifies nonpecuniary benefits. Through a structural model of childcare demand, we estimate substantial welfare gains from the policy, yielding a Marginal Value of Public Funds (MVPF) above 3.5. Using the sufficient-statistic approach underestimates welfare gains by half. Counterfactual simulations and a difference-in-differences analysis suggest that increasing availability, rather than solely improving affordability, is crucial for the effective design of universal programs.

Despite years of transparency reforms, their impact on civic participation is not well understood. We study whether government performance information from one of the largest transparency initiatives can motivate citizens to participate and whether changes in beliefs are the mechanisms driving this effect. The initiative tracks 112 of India’s 806 most underperforming districts across key development indicators and posts performance information online for public accountability. We conducted a randomized field experiment with 2,107 adults in a district targeted by the initiative. Over 90% in our setting were unaware of the initiative, and 75% overestimated their district’s performance. Our low-cost, scalable video intervention informed citizens that their district was included, revealed its declining performance relative to neighboring districts, and listed formal grievance channels. Citizens randomly assigned to watch the video were 5 pp more likely to sign the petitions and 8.4 pp more likely to attend workshops on complaint filing portals. They updated beliefs about their district’s low rank and became more aware of the initiative. Citizens who initially overestimated their district’s rank updated their beliefs and increased participation, while we observed no effects among those who had not overestimated prior to treatment. Beliefs about citizens’ influence over service delivery remained unchanged. One month later, beliefs about the district’s rank converged, yet citizens in the intervention group continued to meet civic officials and file complaints. Those who initially overestimated their district’s rank remained significantly more likely to participate. To explain why citizens continue participating despite low perceived influence, we develop a model where participation depends on expected performance gaps relative to aspirational service levels and perceived influence over government. Larger perceived gaps alone can sustain participation through an accountability channel. These findings have implications for 250 million Indians in most-underperforming districts and highlight how prior beliefs mediate effectiveness of transparency programs.

We study whether machine learning forecast combinations can improve on the equal-weight mean of survey forecasts from Slovak commercial banks for GDP growth and inflation. Using monthly bank forecasts with full histories (2007–2024; 216 observations per series) and real-time data vintages that respect publication lags, we run a rolling out-of-sample evaluation. We compare linear and penalised regressions (OLS, ridge, lasso, elastic net), nonlinear methods (trees, random forests, XGBoost, kNN, SVR), an ARIMA benchmark, and stacked ensembles estimated via non-negative least squares. Performance is measured by relative MSE versus the bank mean. For inflation, regularised linear combinations deliver consistent gains (RelMSE at around 0.81–0.82), while nonlinear and stacked approaches underperform, suggesting overfitting in small panels. For GDP, several nonlinear models and a restricted stack improve materially on equal weights.

How do parents respond to their peers’ choices when deciding whether and how much to invest in their children’s extracurricular education? Using a field experiment, we identify distinct mechanisms through which peer information shapes investment decisions: positive and negative social learning, as well as social utility. Consistent with positive social learning, we find that peers’ willingness to purchase a course increases both parents’ willingness to purchase and their level of investment. Negative social learning effects, however, depend on the nature of the signal.  Explicit peer criticism significantly reduces both willingness to invest and spending, whereas peer non-purchase without negative evaluation has no effect. Social utility effects also vary by peer purchase price: high-price peer purchases increase spending, whereas low-price peer purchases raise a parent’s willingness to invest but do not lead to higher expenditures. We further document heterogeneity in responses to negative peer signals across socioeconomic and residential contexts. Using an enhanced causal forest approach to complement our regression analyses, we find that peer non-purchase without explicit criticism increases purchase preference among parents in less central areas with smaller socioeconomic gaps, but reduces it among parents in central areas with larger gaps.

This paper investigates whether severe economic hardship undermines preferences for honesty. We use controlled, incentivized measures of cheating for private benefit in a large, diverse sample of 5,676 Kenyans, exploiting three complementary sources of variation: experimentally manipulated monetary incentives, randomized increase in salience of own financial situation, and the Covid‑19 income shock, exploiting randomized survey timing as a natural experiment with respondents surveyed before and during the crisis. We find that severe economic hardship—marked by a 50% drop in monthly earnings— leads to a sharp increase in the prevalence of cheating, from 43% to 72%. Cheating behavior is highly responsive to financial incentives and increases gradually with prolonged hardship. The effects are largest among the most economically impacted and are amplified when salience of own financial situation is experimentally increased. Predictable seasonal income fluctuations, in contrast, do not affect honesty. The results demonstrate that while most individuals exhibit a strong preference against cheating under normal conditions, severe economic hardship substantially erodes honesty.

This study investigates the effects of California’s Senate Bill 328 (SB 328)—the first state legislation requiring later school start times for middle and high schools—on adolescent sleep, mental health, and academic outcomes. Drawing on data from the Youth Risk Behavior Surveillance System (YRBS) for high school students (grades 9–12) and the American Time Use Survey (ATUS), which includes sleep duration for adolescents aged 15 above and bedtimes and wake-up times for children under 13, we analyze shifts in sleep patterns and mental health metrics. We then examine the impact on academic performance using district-level data from the Stanford Education Data Archive (SEDA) for students in grades 3–8 and SAT scores for high school students. Employing difference-in-differences (DID) and matched DID methods, we find that SB 328 led to a 14% increase in the share of students sleeping at least 8 hours per night, consistent with CDC recommendations for adolescents. We find suggestive evidence of a reduction in certain mental health problems, particularly difficulties concentrating, though the results are imprecisely estimated and not robust across specifications. Finally, we find evidence of significant improvements in math and English scores in grades 3–8 (approximately a .1 standard deviation increase), while SAT scores rose by 2%. We uncover substantial heterogeneity in the policy's effects, with larger improvements in sleep and mental health among boys and Hispanic students, and larger academic gains among Hispanic students.

Despite the potential of economics to serve as a vehicle for social mobility and its impact on society, persistent disparities in participation and progression raise important concerns for the discipline. This presentation examines the structure of the economics education pipeline in the UK, focusing on patterns of entry, attrition, and academic outcomes across key demographic dimensions. Using institutional-level data from UK universities, I analyse variations in enrolment, continuation rates, and degree attainment by gender, ethnicity, and socio-economic background. The analysis identifies systemic barriers and points of attrition that disproportionately affect students from underrepresented groups.

The notion of a “preference” lies at the heart of economic analysis. Yet, because preferences are typically inferred from observed behavior - and because such behavior is often shaped by cognitive biases and bounded rationality - this raises a fundamental challenge: how can we recover individuals’ true economic preferences when their decisions are systematically distorted? This paper addresses that challenge by proposing a novel method to identify true preferences in the presence of attenuation biases - a pervasive distortion that arises when individuals are uncertain about what choice is optimal and thus respond only weakly to economic incentives. Under mild assumptions, the two elicitations allow the direction and magnitude of the bias to be bounded; and under stronger but plausible assumptions, the method delivers a unique inference of the true preference. Evidence from two experiments confirms the validity and generalizability of the method. First, in an induced-value experiment that approximates product demand decisions, observable demand curves imply non-trivial attenuation bias and associated welfare distortions. Second, in a contingent valuation experiment where people evaluate policy programs, observable valuations are insensitive to the scope of the program – a pattern known as “scope neglect”. Applying the proposed method shows that scope neglect is largely attributable to attenuation bias and that true sensitivity of valuation to scopes are likely much higher than from observable behavior. 

The spread of disinformation is widely regarded as one of the most serious short-term global risks and a major societal challenge. In one laboratory and three online experiments, we measured trust in true, false, and disinformation statements related to the Russo-Ukrainian war, politics, climate, and health. We examined the effects of refuting disinformation using a corrective message containing factual information. By manipulating the timing of the message, we tested whether it is more effective to intervene before exposure to disinformation (i.e., to prebunk) or after exposure (i.e., to debunk). We found that debunking significantly outperformed prebunking. The positive effects of debunking persisted for at least two weeks. In the fifth experiment, we examined both direct and indirect (i.e., spillover) effects of endorsing some of the true statements and refuting some of the disinformation statements with either fact-based or logic-based corrective message. The two interventions performed similarly overall, but fact-based corrective messages resulted in marginally larger positive direct effects, while logic-based corrective messages generated significantly larger positive spillover effects.

While technology can magnify existing inequalities by excluding marginalized groups in society, paradoxically, it can also serve as a means for them to move from isolation to inclusion. In this paper, we report a study illustrating how individuals belonging to one marginalized group—people with lifelong disabilities (PWLD)—used technology to navigate this path. Our study was part of an initiative by the Norwegian government exploring how PWLD can attain social inclusion (SI) through the use of virtual worlds (VW), specifically Second Life (SL). Using a conceptualization of SI specific to PWLD, which consists of two interrelated domains—interpersonal relations and community participation—we trace how the actualization of affordances in VW enabled PWLD to achieve SI. This outcome was built upon the realization of five interrelated affordances: Communicability, Mobility, Personalizability, Engageability, and Self-actualizability. Based on these insights, we demonstrate the role of VW affordances in achieving SI and propose a model titled the affordance-based pathway model of social inclusion. We contribute to the discourse on affordance theory by extending the role of facilitating conditions. Specifically, the outcomes achieved by the actualization of certain affordances can serve as facilitating conditions that support the perception and eventual actualization of other affordances.

Household and ambient air pollution (HAAP), a major global health risk, is linked to a lower quality of life and is responsible for over six million premature annual deaths globally. We investigate country-level socioeconomic, environmental, energy, and health determinants of HAAP mortality rates, using regression analyses and global GIS mapping of predicted probabilities on high HAAP mortality risk. While related studies are predominantly country-specific based on micro-level factors, our study provides global evidence (n=150 countries) based on a broad range of macro-level indicators. Our findings reveal that rural access to clean cooking fuels and technology, as well as greater healthcare expenditure are critical for reducing HAAP deaths, whereas rurality and energy deprivation significantly increase mortality risks. While advanced economies show clear resilience to HAAP mortality risks, emerging and developing economies are disproportionately vulnerable. Contrary to related literature, our analyses also show that males are more at risk of HAAP mortality than females. Our results are robust to alternative model specifications and remain consistent over repeated samples. We further contextualise our global evidence with previous country-specific case studies on HAAP risks. Our research helps to appraise the progress towards achieving the UN Sustainable Development Goals 3, 5, and 7, addressing their associated targets and indicators, providing guidance for policymakers to strengthen efforts to reduce HAAP mortality and improve living conditions globally.

This paper examines how refugees' destinations and preferences evolve, focusing on Central and Eastern European countries. While our research findings are focussed on Bulgaria, Romania, and Hungary—regarded as ‘transit only countries’ prior to the Ukraine refugee crisis - we consider potential lessons to be drawn for more recent refugee inflows. Using a mixed-methods approach we show that prioritising safety significantly increases the likelihood of asylum seekers reconsidering a transit or temporary migration country as a settlement destination. Other influential factors include asylum conditions, migration costs, and educational background, with more educated individuals more likely to revise initial plans. Although our primary focus is on asylum seekers, we find that high migration costs affect decisions on destinations, suggesting a need to ‘recover’ investments through settlement in more stable or economi­cally attractive countries. The qualitative findings support the results, highlighting the role of legal stability, social networks, and perceived opportunity in shifting preferences for settlement. Finally, as the Czech Republic hosts most Ukrainian refugees per capita relative to its population, with a high share of women migrants, we add a reflection on the gendered aspects of migration, linking our research to the wider literature on the topic. 

We investigate whether departures from stationary intertemporal preferences, as identified in parametric estimations, reflect genuine nonstationarity in time preferences or arise from inconsistent or error-prone choice. Using data from a controlled lab experiment with a large number of intertemporal choices per subject, we applied nonparametric revealed preference tests to evaluate whether subjects’ behavior satisfies utility maximization and stationarity. We find that apparent violations of stationarity in parametric estimations are often better explained by misidentification due to inconsistency of choices with utility maximization or misspecification of the functional form, rather than by true nonstationary time preferences.

Job referrals through informal networks are an essential channel for disseminating information about the qualifications of job candidates. As such, they play a crucial role in determining the outcomes of hiring and promotion decisions. In this paper, we study gender biases in the referral process. We investigate this question through an online experiment in which university students are asked to nominate their highest-scoring classmates in either a math or a verbal task. Using administrative data, we reconstruct the students’ co-enrollment network. This allows us to identify who is chosen as well as everyone else who was not. In other words, we can measure the quality of the referrals and the characteristics of candidates who are better but not chosen. We find that participants are more likely to refer men than equally qualified women in the math task but not in the verbal task. This difference is partly explained by gender differences in network structure, i.e., who is linked with whom. However, equally important are gender biases in the referral of known contacts.

We study belief dynamics of an agent who repeatedly 'chooses' new beliefs, trading off costs from making worse decisions with benefits from anticipatory utility. We argue that some naiveté about past belief distortions is necessary to explain evidence on motivated reasoning and derive testable implications for belief dynamics when the agent naively treats each choice of beliefs as the first (and last) such choice. Such naiveté causes persistent insecurity and fragile optimism in important domains of life and dogmatic optimism in less important domains. It can also generate 'information avoidance' where a naive agent repeatedly, and indefinitely, postpones receiving information.

Background: This study investigates the potential relationship between obesity and self-ranking of poverty, as a proxy for self-awareness and happiness. To the best of our knowledge, this issue has not been previously explored based on self-ranking of poverty when income is controlled. Method: Ordered Probit Regressions. We propose a new measure for the influence of western social values and norms associated with discrimination against obese women. Results: Based on a follow-up survey after two years, findings demonstrate a drop in the projected probability of self-ranking as “not poor” with the BMI from 0.73 to 0.37 (females) – 0.48 (males) when the level of income is controlled. Similar outcomes are obtained when the independent variables are lagged and thus avoid endogeneity concerns. Finally, additional outcomes support the conclusion that the lagged BMI Granger-cause self-ranking of poverty for women, but not for men. Findings support the awareness of more obese women to lower prospects of finding a job. Conclusion: Since according to twin studies, approximately 80% of obesity emanates from genetic factors, research findings stress the need to educate the public against prejudices on the grounds of obesity. In particular, our study seeks to evoke awareness among potential employers, which, in turn, might motivate avoidance of, or at least reduction in, an implicit wage penalty against obese women.

Allowing voters to trade votes for money may generate either negative or positive externalities: A voter is harmed (benefited) if voters she agrees with end up with reduced (increased) voting power. However, the plausibility of each kind of externality crucially depends on the electoral rule in place (i.e., on the way votes are aggregated into a political outcome). Focusing on a private-values setting with two alternatives and on the family of qualified majority rules (ranging from a simple majority to unanimity), we first argue that trade: a) cannot generate any negative externalities in the interim stage for any rational voter (and, ex post, to any non-trader) if and only if a unanimity rule is in place, and b) can generate positive externalities to agents under any qualified majority rule. That is, under unanimity rule, vote trading is less objectionable in normative and/or welfare-related terms than under any other electoral system. We conduct a laboratory experiment to test these predictions and show that despite the fact that vote-trading improves utilitarian welfare under unanimity rule, several subjects are worse off compared to when vote-trading is not allowed. This raises important caveats regarding its use even under this most favorable institutional setting.

This paper aims to assess the social welfare benefits of open access competition for long distance passenger rail in Europe, using GB data as a case study, importantly taking into account not only benefits to passengers through lower fares and increased frequency, but impacts on total costs and impacts on government finances. This takes the cost analysis beyond Stead et. al. (2019), first by extending the analysis into an overall social cost benefit analysis with broader policy implications; and second by permitting more up-to-date and more years of data to assess cost differences between the incumbent and open access operators, taking into account that open access volumes have grown in the meantime.