A new equilibrium: COVID-19 lockdowns and WFH persistence
Laura Ketter Todd Morris Lizi Yu · 2025
Read A new equilibrium: COVID-19 lockdowns and WFH persistence by Laura Ketter Todd Morris Lizi Yu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-16671.txt.
Introduction
Laura Ketter, Todd Morris, and Lizi Yu study in 2025 how COVID-19 lockdowns set lasting work-from-home habits in Australia. Office workers in three hard-hit states endured far longer lockdowns than peers elsewhere. Rich yearly survey data let the team compare the groups over time.
Locked-down workers still worked from home forty-three percent more through 2023, about half a day a week extra. The effect grew with lockdown length. Both sides of the market locked it in: firms shrank offices and posted remote jobs, while workers moved outward and built home offices. A shock, the authors argue, moved the whole system to a new balance.
Catalog record
| CREATOR | Laura Ketter Todd Morris Lizi Yu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 106 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-16671
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A new equilibrium: COVID-19 lockdowns and WFH persistence For helpful comments, we thank Peter Kuhn and Tim Moore. This paper uses unit record data from the Household, Income and Labour Dynamics in Australia (HILDA) Survey. The unit record data from the HILDA Survey was obtained from the Australian Data Archive, which is hosted by The Australian National University. The HILDA Survey was initiated and is funded by the Australian Government Department of Social Services (DSS) and is managed by the Melbourne Institute of Applied Economic and Social Research (Melbourne Institute). The findings and views based on the data, however, are those of the authors and should not be attributed to the Australian Government, DSS, the Melbourne Institute, the Australian Data Archive or The Australian National University and none of those entities bear any responsibility for the analysis or interpretation of the unit record data from the HILDA Survey provided by the authors. Laura Ketter Todd Morris Lizi Yu Abstract This paper documents a robust link between COVID-19 lockdowns and the uptake and persistence of working from home (WFH) practices. Exploiting rich longitudinal data, we use a difference-in-differences strategy to compare office workers in three heavily locked-down Australian states to similar workers in less affected states. Locked-down workers sustain 43% higher WFH levels through 2023 — 0.5 days per week — with a monotonic dose–response relationship. Persistence is driven by adjustments on both sides of the labor market: employers downsize office space and open remote/hybrid positions, while employees relocate away from city centers and invest in home offices and technology. Keywords: Work from home, WFH, Persistence, COVID-19, Lockdowns, Habit formation JEL Classification: I18, J22, M54 Nothing is so permanent as a temporary government program. — Milton Friedman, 1984 1 Introduction Emergency government interventions focus on immediate challenges but may induce lasting behavioral changes. The COVID-19 pandemic provides a rare natural experiment to examine this possibility, as governments around the world imposed sweeping lockdowns that abruptly disrupted established work patterns. While a growing body of research has documented the short-term economic effects of these policies — including impacts on employment, consumption, and business activity ( Chetty et al. 2024 , e.g.,) — much less is known about whether such emergency interventions can produce enduring changes in how people work and live. This paper focuses on arguably the pandemic’s most enduring legacy: the widespread adoption of work from home (WFH) arrangements, which have shifted from a niche practice to a lasting feature of modern labor markets ( Barrero et al. 2023 ) . Despite the clear and sustained rise in remote work, the importance of public policies in driving this trend remains unclear, due to limited pre-pandemic data on WFH and difficulties in identifying appropriate counterfactuals. Australia offers a compelling setting to study the impact of lockdowns on the uptake and persistence of WFH behaviors. While the country implemented a six-week national lockdown from March to May 2020, significant geographic variation in restrictions emerged thereafter. As shown in Figure 1 , some states and territories --- Victoria, New South Wales (NSW), and the Australian Capital Territory (ACT) --- endured prolonged lockdowns and strict stay-at-home orders, whereas others experienced only short and sporadic lockdowns until mass vaccination was achieved in October 2021. 1 1 1 Australia contains six federated states and two territories; for simplicity, we hereafter refer to all eight as “states”. Figure 1 provides a map of the eight states. On average, workers in Victoria, NSW and ACT experienced 143 more days under lockdown than workers in the other five states. These geographic differences resulted mainly from localized outbreaks rather than deliberate policy decisions — like many Asian countries, all Australian states pursued an elimination strategy to COVID-19, under which stringent lockdowns were a standard response to local outbreaks, with lockdowns lasting until the outbreak was eliminated or mass vaccination was achieved. Our study exploits the variation in lockdown duration across states — comparing Victoria, NSW, and the ACT to the other five states — in a difference-in-differences (DiD) framework, using rich individual-level data from the Household, Income and Labour Dynamics in Australia (HILDA) Survey. HILDA is a nationally representative longitudinal study that has followed thousands of Australians over time, with annual interviews conducted between August and November. It collects detailed and consistent measures of WFH since 2002 ( Botha et al. 2023 ; Laß et al. 2025 ) , allowing us to construct six outcome variables capturing WFH adoption on both the intensive and extensive margins. 2 2 2 These variables are: (i) share of total hours worked from home; (ii) total WFH hours; (iii) indicator for any WFH; (iv) indicator for WFH under a formal agreement; (v) indicator for WFH \geq 60% of total hours; and (vi) indicator for fully remote work. We focus on a panel of “office workers”, such as managerial, professional, or clerical roles. This group, which makes up half of the Australian workforce, was the most directly affected by stay-at-home orders during the pandemic and also the most relevant for understanding the rapid increase in WFH practices (Figure A1 ). Our office worker sample comprises over 10,000 individuals and nearly 60,000 worker-year observations from 2002–2023. Relative to other studies, which rely on cross-sectional online surveys initiated after the pandemic began, we benefit from an existing high-quality survey with a long pre-COVID period, longitudinal structure, high response rates, and consistent WFH measures. The long pre-COVID period and consistent measurement of WFH outcomes allow us to reliably examine the parallel-trends assumption required for the DiD estimation. In our event-study analysis, treated and control states exhibit nearly identical WFH adoption rates and trends across all six of our WFH measures, from 2002 to 2019. This parity breaks with the onset of the pandemic, as WFH outcomes rise sharply in the treated states relative to the controls, and the gap begins to narrow in 2022. This clear divergence motivates our formal DiD analysis to assess how lockdown duration affects WFH outcomes in the short run during the lockdown period (2020–21), and in the medium run after restrictions were lifted (2022–23). 3 3 3 Relatedly, prior studies exploit the extended lockdown in Melbourne, Victoria in 2020 to examine its immediate impacts on mental health ( Butterworth et al. 2022 ) , broad human activities ( Schurer et al. 2023 ) , and education outcomes ( Gillitzer and Prasad 2024 ) . The specification controls for state-by-occupation fixed effects, which allow WFH feasibility for the same occupation to vary across states, and survey wave fixed effects to account for national trends over time. To flexibly account for individual, job, and location characteristics, whose impacts on WFH may have changed since COVID-19 ( Barrero et al. 2021 ) , we interact these controls with indicators for the pre-COVID period (2002–19), lockdown period (2020–21), and post-COVID period (2022–23). The estimated treatment effects are strikingly large and statistically significant at the 1% level across all six WFH measures, indicating that the prolonged lockdowns in Victoria, NSW and ACT caused a massive and persistent uptake of remote working that would not have occurred to the same extent otherwise. On the extensive margin, workers in treated regions are 46% (19.6 pp) more likely to engage in any