José Alberto Molina

Social connection is a key determinant of emotional well-being, yet the role of solitude in shaping both momentary affect and well-being remains understudied. This paper investigates how being alone while engaging in daily activities relates to subjective well-being, using rich time use diary data from the UK covering four distinct periods: pre-pandemic (2015–2016), the Covid-19 lockdowns (2020–2021), the relaxation phase (2021), and the post-pandemic period (2023). We find that being alone is negatively associated with momentary enjoyment, particularly in the post-pandemic period, but not during lockdowns or the initial relaxation phase, suggesting that the emotional cost of solitude depends on its perceived voluntariness and social norms. The enjoyment penalty is strongest for leisure and unpaid work episodes, and most pronounced among remote workers. Our results contribute to the literature on experienced utility, labor supply, worker time allocation, and remote work, highlighting the need to account for the emotional toll of isolation in welfare analysis and policy design.

Are inferences based on synthetic samples from large language models (LLMs) robust to experimenter demand effects? We bound demand effects across eight LLMs from three providers and nine canonical economic tasks, benchmarking our estimates against a human sample of 7,517 participants. Our results are consistent with large demand effects in LLM samples. Making the researcher's expectation explicit—weak-demand treatment—shifts responses by 1.223 standard deviations for LLMs, compared with 0.157 among humans. Making the researcher's request explicit—strong-demand treatment—shifts responses by 1.714 standard deviations for LLMs and 0.571 among humans. On a common unit-interval scale, demand wording shifts average LLM choices by 22.3% of the feasible action range under weak demand and 48.1% under strong demand, compared with 3.8% and 15.2% in the human benchmark. The magnitude of these effects varies across tasks, providers, and individual models, but no provider or model family approaches the relatively low level of demand responsiveness observed in the human sample. These findings suggest that LLM agents may be useful for some experimental applications, but inferences based on synthetic samples should explicitly assess and bound potential demand effects.

We estimate the effect of the Tax Cuts and Jobs Act of 2017 (TCJA) on household time devoted to financial management activities. The TCJA nearly doubled the standard deduction, inducing a large share of itemizing middle-income households to switch to the simpler standard deduction and thereby reducing their annual documentation burden. Using the American Time Use Survey (ATUS) for 2013–2023 and a difference-in-differences design comparing households with incomes of $50,000–$100,000 against lower-income households below $35,000, we find no statistically significant average treatment effect on financial management time (estimates of -0.95 to +0.53 minutes per day across the primary specifications, p > 0.06 throughout). Pre-reform event-study coefficients are jointly insignificant (F(4,50)=1.26, p=0.30), validating the parallel-trends assumption. Heterogeneity analysis reveals that college-educated treated households experienced a significant reduction (-1.35 min/day, p=0.04), that men drove most of the negative trend (triple interaction p=0.08), and that high-SALT-state households show a marginally significant response. Using the ATUS Well-Being Module, we further document a post-reform decline in the subjective happiness associated with financial management activities (-0.47 points, p=0.03), consistent with a compositional shift toward more complex and burdensome tasks. Our findings suggest that aggregate time-cost savings from tax simplification are modest and concentrated among the most financially sophisticated households.

This article examines how expecting to receive an inheritance impacts household savings. Life‐cycle consumption models indicate that the expectation of inheriting should reduce current savings for forward‐thinking consumers. We investigate this economic prediction, considering factors such as liquidity constraints and education. To do so, we use household fixed effects to account for time‐invariant factors and exploit within‐household variation over time by using the Japanese Panel Survey of Consumers (2003–2019), which provides individual‐level information and overcomes endogeneity concerns commonly present in cross‐sectional studies. Our findings reveal that households adjust their current savings in anticipation of receiving future inheritances. Specifically, men decrease their current savings by 5% if they expect to receive an inheritance in the future. Additionally, we find more pronounced changes among households with higher education and income levels. These findings inform inheritance fiscal policies, such as inheritance taxes, revealing that future inheritance expectations influence current savings decisions.

Household and Population Social Sciences cover a high number of topics in Economics, but also in other fields such as Sociology, Psychology and Demography, among others. In this paper, we analyze the production and networks of a total of 6472 authors who have published 5070 papers in a selection of top cross-disciplinary journals (indexed in SSCI of WOS) which cover these topics of Household and Population Social Sciences between January 1969 and January 2021 (Journal of Population Economics, Journal of Human Resources, Feminist Economics, the Review of Economics of the Household, the Journal of Demographic Economics, Demography, Population and Development Review, and the European Journal of Population). Using the Impact Factor (IF) corresponding to the year of publication, our results first identify the academic leaders among those authors, as well as other results in terms of communities. Results reveal that the largest community is led by the sociologist Trude Lappegard; the community with the largest production is led by the economist Samuel H. Preston and the one associated with the most central author, the psychologist Ariel Kalil, is led by the economist Greg J. Duncan. Additionally, we note that collaborations among these authors are very rare, with only their first neighborhoods collaborating, which shows the need for greater collaboration between these social sciences to improve authors’ academic visibility.

This paper analyzes the impact of lottery wins on household labor supply in the United Kingdom, using data from the British Household Panel Survey. We show that lottery wins do not have significant effects on hours of work of males, while female hours of work decrease in response to lottery wins. When we control for different lottery prize amounts, we find that large lottery wins reduce female annual hours of work by 120 h one and 2 years after the prize. The estimates are heterogeneous across age groups, levels of educational attainment, and household composition. These results suggest that shocks in unearned income may take some time to appear but have a lasting impact.