Literature Review – Remote Work and Employee Productivity Since 2020

Introduction

The abrupt relocation of white-collar work into private homes from March 2020 created the largest natural experiment in work organisation in a century. In Australia, the share of employed people working from home most days rose from around 8 per cent before the pandemic to roughly 40 per cent at the height of public health restrictions (Productivity Commission 2021), and the practice has proved durable: just over one third of employed Australians (36 per cent) still regularly worked from home in August 2024 (ABS 2024). Whether this reorganisation of work helps or harms productivity has become one of the most contested questions in contemporary management research.

This review synthesises the management and economics literature published since 2020, anchored by a small number of foundational earlier studies. It clarifies terminology, examines the measured productivity evidence, and then considers moderators, collaboration and innovation effects, wellbeing and boundary management, and monitoring and trust. A methodological critique and a synthesis of gaps follow, before the review draws implications for hybrid policy design in the Australian regulatory environment shaped by the Fair Work Act 2009 (Cth).

Definitions and Scope

Early commentary conflated arrangements that the literature now separates. Fully remote work locates all paid days away from the employer’s premises; hybrid work splits the week between home and a central workplace, most commonly with two or three home days. The distinction matters because effects differ sharply by intensity. Barrero, Bloom and Davis (2023) show that working from home in the United States stabilised at just under 30 per cent of paid full days after 2022, with hybrid schedules dominating among university-educated professionals. Productivity is equally contested as a construct: economists prefer output per unit of labour input, firms rely on performance ratings or call and ticket metrics, and surveys elicit self-assessed efficiency. Because conclusions vary with both the arrangement and the measure, this review flags each study’s definitions where material.

Evidence on Measured Productivity

Self-Reported Evidence

Survey evidence returns consistently favourable verdicts. Workers in repeated large-scale United States surveys assess themselves as at least as productive at home as in the office (Barrero, Bloom & Davis 2023). In a harmonised survey across 27 countries including Australia, full-time workers valued the option of two to three remote days at about 5 per cent of pay, with commuting time savings the dominant reason (Aksoy et al. 2022). Self-reports, however, embed known inflators: respondents commonly count reclaimed commuting time as a productivity gain, and social desirability encourages favourable self-assessment among those who wish to retain the arrangement. Self-reported data therefore establish that employees experience remote work as productive, not that measured output rises.

Objective and Quasi-Experimental Evidence

Objective designs paint a more conditional picture. The foundational benchmark remains the nine-month randomised trial at a Chinese travel agency, where volunteer call-centre staff working four home days per week raised performance by 13 per cent, largely through quieter conditions, while attrition halved (Bloom et al. 2015). Post-2020 personnel-data studies are less sanguine about fully remote arrangements. Among information technology professionals at a large Asian services firm, working hours rose markedly, including substantial out-of-hours work, while output was flat, implying a productivity decline of 8-19 per cent driven by heavier coordination and meeting loads (Gibbs, Mengel & Siemroth 2023). In United States call centres, an office closure that forced workers remote reduced calls handled by about 4 per cent and lowered call quality; critically, workers who had chosen remote roles were already less productive than office peers, so raw cross-sectional gaps overstate the causal penalty (Emanuel & Harrington 2024).

Hybrid arrangements fare better. In a six-month randomised trial covering 1,612 professionals at Trip.com, two home days per week left performance reviews and promotion outcomes unaffected over two years while reducing quit rates by one third, with the largest retention gains among non-managers, women and employees with long commutes (Bloom, Han & Liang 2024). Read together, the studies summarised later in Table 1 suggest that fully remote work without deliberate redesign tends to be mildly negative for measured output in interdependent settings, whereas well-structured hybrid work is approximately output-neutral and clearly positive for retention.

Moderating Factors: Job Type, Autonomy and the Home Environment

The variance across studies is not noise; it is moderation. Task characteristics matter first. Independent, individually measurable work such as call handling, coding and claims processing transfers home most successfully, while interdependent and novel work suffers larger coordination penalties (Gibbs, Mengel & Siemroth 2023). Work design research provides the second lens. Drawing on interviews and survey data collected during pandemic lockdowns, Wang et al. (2021) identify loneliness, procrastination, work-home interference and ineffective communication as the dominant remote-work challenges, and show that job autonomy and social support buffer these challenges while close monitoring and heavy workload aggravate them.

The home environment supplies a third moderator. A dedicated workspace, reliable connectivity and the absence of simultaneous caring responsibilities separate productive home offices from compromised ones, a distinction with equity implications given that managers and professionals are far more likely than machinery operators or labourers to hold jobs that can be performed remotely (ABS 2024). Career stage matters as well: remote early-career workers receive measurably less feedback and mentoring, which depresses skill accumulation even where current output holds up (Emanuel & Harrington 2024).

Collaboration and Innovation Effects

Task-level productivity may conceal slower-moving collaborative costs. Analysing communication metadata for 61,182 Microsoft employees, Yang et al. (2022) found that firm-wide remote work made the collaboration network more static and siloed, with fewer bridging ties across groups and a shift from synchronous conversation towards email and messaging, changes the authors argue impede the transfer of complex, novel information. Experimental evidence points the same way for idea generation: videoconferencing pairs produced significantly fewer creative ideas than face-to-face pairs in laboratory and field experiments, apparently because narrowed visual focus on the screen narrows cognitive focus, although idea selection was unimpaired (Brucks & Levav 2022).

These findings caution against inferring innovation-neutrality from short-run output results. Notably, the Trip.com trial detected no performance or promotion penalty within two years under hybrid conditions (Bloom, Han & Liang 2024), suggesting that co-located days, deliberately coordinated so that teams overlap, capture most collaborative benefits while retaining flexibility.

Wellbeing and Boundary Management

The clearest welfare gain from remote work is reclaimed commuting time, which workers across countries reallocate to a mix of additional work, caregiving and leisure (Aksoy et al. 2022). Wellbeing effects beyond the commute are double-edged. British survey evidence predating the pandemic already showed that remote workers report higher organisational commitment and job satisfaction alongside greater work intensification and difficulty switching off (Felstead & Henseke 2017), a pattern the lockdown literature replicated in the form of work-home interference and extended hours (Wang et al. 2021; Gibbs, Mengel & Siemroth 2023).

Boundary management has therefore become a managerial competency in its own right, spanning temporal routines, spatial demarcation and explicit communication norms. The Australian regulatory environment now reinforces these practices: a statutory right to disconnect entered the Fair Work Act 2009 (Cth) in 2024, permitting eligible employees to refuse unreasonable out-of-hours contact, and employers’ work health and safety duties extend to home workstations (Productivity Commission 2021). Wellbeing is consequently not peripheral to the productivity question; unsustainable intensification erodes precisely the retention advantage that hybrid work delivers.

Monitoring, Trust and Managerial Practice

Remote work strains line-of-sight supervision, and many employers responded after 2020 by expanding digital surveillance. The work design evidence suggests this is counterproductive: close monitoring aggravates the challenges of remote work and undermines the autonomy that makes it effective (Wang et al. 2021). The alternative is trust-based management anchored in outcome measurement, and Australian evidence is instructive. Surveying approximately 6,000 Australian public servants during the first pandemic wave, Colley and Williamson (2020) found that most managers judged their teams equally or more productive at home and reported increased trust in employees, a marked shift from the presenteeism culture that had previously rationed flexible work. Productivity under flexibility is therefore less a technology problem than a performance-management problem: organisations that specify outputs clearly can afford autonomy, while those that manage by observation import the office’s weakest habit into a setting where it functions worst.

Methodological Critique

The post-2020 evidence base carries identifiable weaknesses. First, pandemic confounds contaminate studies conducted during 2020-21: school closures, health anxiety and hastily improvised arrangements mean that estimates from that period measure working from home during a crisis rather than working from home as a settled practice (Productivity Commission 2021). Second, selection pervades non-experimental comparisons. Workers sort into remote roles on unobserved characteristics, and the call-centre evidence shows that negative selection accounted for much of the raw productivity gap attributed to remote work (Emanuel & Harrington 2024). Third, self-reported productivity is systematically inflated relative to measured output (Barrero, Bloom & Davis 2023).

Generalisability is a further constraint. The most credible causal designs each rest on a single firm, typically in travel services, online retail or information technology, and may not transfer across industries or countries. Australian firm-level causal evidence is particularly thin, leaving local inference dependent on surveys and overseas designs. Finally, observation windows remain short relative to the horizons over which mentoring deficits, weakened networks and slowed innovation would surface (Yang et al. 2022), so null short-run findings should be read as provisional.

Synthesis and Research Gaps

Table 1 consolidates the principal empirical studies reviewed above, and the pattern it displays is coherent once arrangement, measure and setting are distinguished. Self-reports are uniformly positive; objective studies of fully remote work in interdependent settings find modest declines; randomised evidence on hybrid work finds output neutrality with large retention gains; and collaboration studies identify slow-accruing risks to networks and idea generation that short-run metrics miss.

Table 1: Summary of key empirical studies on remote and hybrid work and productivity, 2015-2024

Substantial gaps remain. Five stand out for management research:

• Objective, firm-level Australian evidence is nearly absent; the strongest local studies rely on self-report in the public sector.
• Long-horizon effects on innovation, promotion trajectories and skill formation, especially for graduates and early-career employees, remain unmeasured.
• Small and medium enterprises, which employ the majority of Australian workers, are missing from a literature dominated by large firms.
• The productivity consequences of the 2023-24 Australian legislative changes, including flexible-work arbitration and the right to disconnect, have not yet been evaluated.
• Interactions between hybrid work, housing choice and regional settlement patterns are only beginning to be studied.

Implications for Australian Hybrid Policy Design

The Australian setting gives this evidence unusual practical force because flexibility is partly codified. Section 65 of the Fair Work Act 2009 (Cth) grants eligible employees, including parents, carers and older workers, a right to request flexible working arrangements, and amendments operating from 2023 broadened eligibility and empowered the Fair Work Commission to arbitrate disputed refusals. Employers may refuse only on reasonable business grounds after genuinely attempting to reach agreement, which obliges them to engage with evidence rather than preference. Blanket return-to-office mandates sit awkwardly against randomised findings that hybrid schedules preserve output while cutting quits by a third (Bloom, Han & Liang 2024), and refusals grounded in generalised productivity concerns are increasingly contestable.

The reviewed literature supports five design principles for Australian organisations. First, adopt hybrid rather than fully remote defaults for interdependent roles, since the sharpest measured declines arise under full-time remote arrangements (Gibbs, Mengel & Siemroth 2023). Second, coordinate anchor days so that co-located time actually overlaps within teams, addressing the network and creativity risks identified by Yang et al. (2022) and Brucks and Levav (2022). Third, manage by specified outcomes rather than surveillance, consistent with work design evidence and Australian public sector experience (Wang et al. 2021; Colley & Williamson 2020). Fourth, weight office time towards early-career employees, whose mentoring and feedback losses are best documented (Emanuel & Harrington 2024). Fifth, treat boundary protection, home workstation safety and disconnect norms as productivity infrastructure rather than compliance overhead, because retention is where hybrid work pays. Policy attention should also extend to the occupational skew in access, since flexibility is concentrating among professionals and managers (ABS 2024).

Conclusion

The literature on remote work and productivity has matured rapidly since 2020, moving from speculation to credible causal designs. Its central lesson is conditionality: outcomes depend on the arrangement, the measure and the moderators, not on remote work in the abstract. Fully remote arrangements can depress measured productivity in interdependent work, self-reports overstate gains, and collaboration and innovation face slow-accruing risks. Hybrid work, by contrast, emerges as a robust compromise, roughly output-neutral, strongly retention-positive and highly valued by employees. For Australian organisations operating under an increasingly codified flexible-work regime, the research agenda is now local: firm-level causal evidence, long-horizon career effects and evaluation of the new legislative machinery. Until that evidence arrives, the reviewed studies favour deliberately designed hybrid arrangements over both blanket office mandates and unstructured remote work.

References

Aksoy, C.G., Barrero, J.M., Bloom, N., Davis, S.J., Dolls, M. & Zarate, P. 2022, ‘Working from home around the world’, Brookings Papers on Economic Activity, Fall, pp. 281-360.

Australian Bureau of Statistics (ABS) 2024, Working arrangements, August 2024, ABS, Canberra.

Barrero, J.M., Bloom, N. & Davis, S.J. 2023, ‘The evolution of work from home’, Journal of Economic Perspectives, vol. 37, no. 4, pp. 23-49.

Bloom, N., Han, R. & Liang, J. 2024, ‘Hybrid working from home improves retention without damaging performance’, Nature, vol. 630, pp. 920-925.

Bloom, N., Liang, J., Roberts, J. & Ying, Z.J. 2015, ‘Does working from home work? Evidence from a Chinese experiment’, The Quarterly Journal of Economics, vol. 130, no. 1, pp. 165-218.

Brucks, M.S. & Levav, J. 2022, ‘Virtual communication curbs creative idea generation’, Nature, vol. 605, pp. 108-112.

Colley, L. & Williamson, S. 2020, Working during the pandemic: From resistance to revolution?, UNSW Canberra Public Service Research Group and CQUniversity, Canberra.

Emanuel, N. & Harrington, E. 2024, ‘Working remotely? Selection, treatment and the market for remote work’, American Economic Journal: Applied Economics, vol. 16, no. 4, pp. 528-559.

Fair Work Act 2009 (Cth).

Felstead, A. & Henseke, G. 2017, ‘Assessing the growth of remote working and its consequences for effort, well-being and work-life balance’, New Technology, Work and Employment, vol. 32, no. 3, pp. 195-212.

Gibbs, M., Mengel, F. & Siemroth, C. 2023, ‘Work from home and productivity: Evidence from personnel and analytics data on information technology professionals’, Journal of Political Economy Microeconomics, vol. 1, no. 1, pp. 7-41.

Productivity Commission 2021, Working from home, Research paper, Productivity Commission, Canberra.

Wang, B., Liu, Y., Qian, J. & Parker, S.K. 2021, ‘Achieving effective remote working during the COVID-19 pandemic: A work design perspective’, Applied Psychology: An International Review, vol. 70, no. 1, pp. 16-59.

Yang, L., Holtz, D., Jaffe, S., Suri, S., Sinha, S., Weston, J., Joyce, C., Shah, N., Sherman, K., Hecht, B. & Teevan, J. 2022, ‘The effects of remote work on collaboration among information workers’, Nature Human Behaviour, vol. 6, no. 1, pp. 43-54.