Joacim Tåg

We examine how gender‐based occupational sorting before the release of ChatGPT relates to predicted exposure to generative AI and its potential implications for the gender wage gap. Using Swedish administrative data, we document that women are overrepresented in occupations predicted to be more affected by generative AI. Mechanical partial‐equilibrium simulations, based on hypothesized deviations from the 2021 occupational and wage distribution and incorporating predicted AI exposure and task complementarity, indicate that generative AI may widen the gender wage gap through existing patterns of gender‐based occupational sorting.

Many countries have established new local colleges to increase access to education for disadvantaged populations. However, many of these expansions have not reduced educational inequality. Drawing on evidence from a large-scale college expansion initiative, we find that increased college availability did not lead to a differential increase in attendance among students from parents with less education. Rather, the expanded access primarily benefited students with marginal academic ability. These results suggest that higher education enrollment is largely determined by inherent scholastic ability and that the expansion of higher education tends to attract students at the upper margin of this ability distribution.

Startups face a trade-off between short-term profitability and long-term growth. Their cash flows are said to follow a so-called J-curve. The shape of the curve depends on investors' financing capacity: their ability to sustain prolonged periods of negative cash flow. US venture capitalists are often believed to have greater financing capacity. We examine a large Swedish dataset with detailed cash flow information. Swedish startups backed by US venture capitalists experience deeper J-curves, with larger short-term losses and higher long-term sales, relative to those backed by non-US venture capitalists. These results are consistent with US venture capitalists having greater financing capacity: they can provide more funding directly and have better access to later-stage investors.

The creation and scaling of startups are inherently linked to risk-taking, with various types of owners handling these risks differently. This paper investigates the influence of an active venture capital (VC) market on startups’ decisions regarding research and scaling. It outlines conditions under which VC-backed startups prefer riskier, yet potentially more rewarding strategies compared to independent startups. VC firms, by means of temporary ownership and compensation structures, introduce ”exit costs” that make high-risk strategies more attractive to VC-backed startups. Moreover, an active VC market prompts startups to undertake higher initial risks, as VC firms provide support for pivoting after setbacks. Additionally, the presence of VC intensifies research risk among established firms, as their research initiatives are strategic complements to the risk choices of startups.

We study the mental health of PhD students in Sweden using comprehensive administrative data on prescriptions, specialist care visits, hospitalizations, and causes of death. We find about 7% (5%) of PhD students receive medication or diagnosis for depression (anxiety) in a given year. These prevalence rates are less than one-third of the corresponding survey-based prevalence rates reported in the literature, and even after adjusting for difference in methodology, 44% (72%) of the prevalence rates reported in the literature. We also document PhD students’ mental health significantly worsens relative to their peers after they have entered the program. This deterioration is consistent with doctoral studies having a negative causal effect on mental health.

Does the matching between workers and jobs help explain productivity differentials across firms? To address this question we develop a job-worker allocation quality measure (JAQ) by combining employer-employee administrative data with machine learning techniques. The proposed measure is positively and significantly associated with labor earnings over workers' careers. At firm level, it features a robust positive correlation with firm productivity, and with managerial turnover leading to an improvement in the quality and experience of management. JAQ can be constructed for any employer-employee data including workers' occupations, and used to explore the effect of corporate restructuring on workers' allocation and careers.

The media often cast foreign private equity firms as villains who gamble with local jobs. We use detailed registry data from Sweden to show that foreign buyouts have not affected workers' labor market outcomes. But domestic buyouts have. They have increased unemployment incidence by a fifth, duration by a third, and lowered labor income by seven percent.