Research

This paper studies the impact of Working from Home (WFH) on voluntary quits in the context of the Italian labor market. It investigates how and to what extent the adoption of WFH during the COVID-19 pandemic shaped workers' quitting behavior and how this translated into changes in employers' market power. Using very rich administrative employer-employee data, I implement a Difference-in-Differences (DiD) analysis that exploits occupational differences in WFH feasibility to identify the causal effect of WFH on workers' likelihood of voluntarily quitting and their wages. Then, I investigate whether this shift translates into changes in workers' voluntary quit elasticity. My results show that workers in teleworkable occupations experienced a decline in voluntary quits and relatively lower wages, suggesting that WFH can act as a non-wage amenity that supports job attachment. Moreover, while workers' quit elasticity declined in the post-pandemic period, this reduction was less pronounced for those in teleworkable occupations. These findings suggest that WFH mitigated the loss of wage responsiveness that characterized the post-pandemic labor market, partly protecting workers’ outside options while reinforcing attachment through non-monetary amenities.

Presented at: 2025 Remote Conference - Stanford University, 40th Conference of the Italian Association of Labour Economics - University of Milano-Bicocca, RES 2025 Annual Conference - University of Birmingham, 2025 Law, Markets, and Public Policy Workshop - Acciaroli, HEC Economics PhD Conference, 2nd Verona Early Career Workshop in Economics - University of Verona, 20th Doctorissimes (2025) - Paris School of Economics, Summer School in Economics and Finance (NSEF), University of Naples Federico II, NSE PhD Internal Seminars - University of Naples Federico II 

This paper studies how the expansion of Working from Home (WFH) during the pandemic affected women’s fertility decisions in Italy. Using data from the Italian Labor Force Survey (2015-2022), I exploit variation in occupations’ "teleworkability", i.e, the extent to which each occupation can be performed from home based on pre-pandemic task-based characteristics, to study maternity leave take-up as a proxy for childbirth. I implement a Difference-in-Differences (DiD) design and a Two-Stage Least Squares strategy (2SLS), instrumenting actual WFH with pre-pandemic occupation-level teleworkability interacted with the post-2020 period. My results show that women in teleworkable occupations became significantly more likely to take maternity leave after the pandemic. The effects are concentrated among highly educated, high-income, and full-time women, those with larger families, as well as in service-sector, white-collar jobs, highlighting that remote work reduces the opportunity cost of childbearing where career-family trade-offs are most binding.

This paper studies how retirement affects household portfolio allocation using rich panel data from the Italian Survey on Household Income and Wealth (SHIW) between 2012 and 2022. Exploiting exogenous variation in pension eligibility rules, we estimate a two-stage least squares with individual and time fixed effects to identify the causal impact of retirement on financial investment behavior. We find that retirement leads to a significant increase in stock holdings: the share of financial wealth invested in stocks rises by 8.2 percentage points, and the probability of participating in the stock market increases by 23 percentage points. This reallocation is mostly driven by publicly traded stocks. We interpret these findings through the lens of precautionary saving: as retirement reduces income risk exposure, the incentive to hold liquid, low-return assets weakens, and people shift toward higher-yielding investments. This effect is particularly pronounced for seniority-eligible individuals - i.e., those with longer, more stable career paths and higher pension benefits - while it is not statistically significant among old-age retirees, who are more likely to have experienced long unemployment spells and remain liquidity-constrained.

Presented at: NSE PhD Internal Seminars - University of Naples Federico II