Remote work has surged and persisted in not only the US (Ramani and Bloom 2021) but also across the world (Zarate et al. 2022). Even managers have recognised that it has become part of the workplace fabric (Shah et al. 2024).

Remote work is frequently described as a workplace amenity, and prior work has shown that employees value it (Cullen et al. 2025). For many employees, the option to work from home carries intrinsic value, much like flexible scheduling, generous leave policies, or autonomy in task execution. For firms, remote work has become a strategic lever tied to recruitment, retention, real estate costs, and productivity (Barrero et al. 2021).

Yet despite intense attention from researchers and practitioners, the empirical record remains unsettled. Some studies report higher satisfaction and stable productivity; others document coordination challenges, weaker mentoring, or diminished advancement. In policy debates and corporate boardrooms, remote work is alternately framed as a long-term competitive advantage or as a threat to organisational cohesion.

In our recent paper (Makridis and Schloetzer 2026), we argue that the debate has been framed too narrowly. Remote work is rarely implemented in isolation. Instead, it is a practice that gets embedded in a broader bundle of job characteristics that include pay, job attributes, managerial quality, communication practices, and organisational culture. If remote work is correlated with these other amenities, then simple comparisons between remote and on-site workers will produce biased or unstable estimates. To understand the true relationship between remote work and employee outcomes, one must examine how it is sorted across jobs and how it interacts with deeper workplace features.

Data and measurement

Our empirical analysis rests on two complementary data sources.

The first is a large-scale dataset from PayScale, which provides detailed information on individual compensation, occupation, industry, and work arrangement, together with measures of job satisfaction. PayScale’s data are especially valuable because they allow for granular measurement of income and job characteristics across individuals in a wide range of roles.

This granularity is essential. Remote work opportunities are not evenly distributed across the labour market – high-skill, professional, and technical occupations are far more likely to offer flexibility than frontline service, production, or operational roles. If remote workers earn more or are concentrated in certain occupations, then any observed satisfaction premium may reflect income or professional status rather than the intrinsic value of location flexibility.

The second dataset comes from Gallup, whose surveys provide rich measures of employee engagement, satisfaction, and workplace perceptions. Gallup data include validated measures of engagement, as well as comparable perceptions of the workplace to the Payscale data on a nationally representative sample. Crucially, the data are also longitudinal, making them a gold standard for analysing the relationship between remote work and employee outcomes.

The integration of PayScale and Gallup data allows us to do something that prior studies have been unable to do, namely, link detailed compensation and occupational information with validated measures of engagement and workplace quality. This combination enables a systematic test of whether remote work independently predicts employee outcomes once compensation and informal workplace characteristics are taken into account.

Baseline results from PayScale

Our PayScale data provide detailed information on compensation, occupation, industry, and work arrangement, together with measures of job satisfaction. In baseline specifications that compare workers by remote status alone, those who work remotely report higher job satisfaction than fully on-site employees. The magnitude of this difference is economically meaningful and consistent with the common view of remote work as a valued amenity.

However, these raw comparisons mask substantial selection effects in who works remotely. Remote arrangements are not randomly distributed across the labour market – they are concentrated in higher paid, professional, and technical occupations. Without controlling for these types of unobserved heterogeneity, the estimated ‘remote premium”’risks capturing differences in income and job type rather than the effect of location flexibility itself.

When we introduce detailed compensation controls using PayScale’s granular income measures, the estimated association between remote work and job satisfaction declines sharply. Remote workers earn more on average, and income is strongly associated with satisfaction. A meaningful portion of the initial premium reflects this compensation differential.

Adding occupation and industry fixed effects further reduces the estimated relationship. Professional and managerial roles are both more likely to offer remote flexibility and more likely to provide autonomy, advancement opportunities, and higher status. Comparing workers within narrowly defined occupational and industry categories attenuates the remote work coefficient.

In our preferred specifications, which jointly control for income, occupation, industry, and perceptions of workplace quality, the independent association between remote work and job satisfaction becomes small and statistically indistinguishable from zero. A similar attenuation appears for intent to leave. In fact, once workplace characteristics are accounted for, workers who report remote arrangements do not exhibit lower turnover intentions, and in some cases report slightly higher intent to leave. Only those who sometimes work remotely display a modest positive association with satisfaction, but even they report higher turnover intentions.

What initially appears to be a sizeable remote work premium is largely explained by its correlation with higher wages, professional occupations, and stronger workplace environments. Location flexibility, once separated from these correlated attributes, carries considerably less explanatory power. In this sense, remote work is embedded within a broader bundle of compensation, occupational sorting, and workplace amenities. Failing to account for these complementarities overstates its standalone association with job satisfaction and retention.

Figure 1 Relationship between remote work intensity and job satisfaction

Figure 2 Relationship between remote work intensity and intent to leave

Validation through Gallup’s engagement measures

Gallup has long documented that engagement is driven by specific workplace factors: clarity of expectations, recognition, opportunities to learn and grow, and strong manager relationships. These dimensions are measured through validated survey instruments with established predictive validity for turnover, productivity, and wellbeing.

Consistent with our earlier results, these Gallup measures strongly predict overall engagement and satisfaction. Importantly, they also correlate with remote work status. Remote workers often report higher scores on certain workplace attributes, particularly in professional contexts.

However, once workplace practices are controlled for, the remote work premium is greatly reduced again, consistent with the view that what appears to be a remote work premium is largely a reflection of better workplace quality in jobs that also happen to be remote. This validation is important because it is conducted on a nationally representative and longitudinal sample, allowing us to follow the same individual over time, thereby controlling for differences in preferences and ability and instead exploit differences in perceptions at different firms.

Heterogeneity across task structure and manager quality

A second major contribution of our paper lies in its exploration of heterogeneity.

Not all jobs are equally suited to remote work. Using task information from Payscale and related occupational characteristics, we distinguish between roles that require high coordination intensity and those that are more independent. In low coordination roles with substantial autonomy, remote work is more likely to be associated with positive outcomes. These jobs depend less on synchronous interaction and informal knowledge transfer.

In high coordination roles, the association weakens and, in some cases, turns negative. Physical proximity may facilitate communication, mentoring, and team cohesion in such environments.

Manager quality further moderates these relationships. Gallup data show that employees who report strong manager relationships tend to maintain higher engagement under remote or hybrid arrangements. Where managerial support is weak, remote work may amplify disengagement.

These heterogeneous effects help reconcile the conflicting results in the broader literature. Studies focusing on software engineers or professional analysts may find positive productivity and satisfaction effects. Studies examining coordination-heavy teams may observe declines. The average effect conceals meaningful variation across task types and managerial contexts.

Reconciling the conflicting evidence

Our results provide a coherent explanation for why the remote work debate has produced divergent findings.

First, different studies examine different segments of the labour market. Because remote work is concentrated in higher paying professional occupations, studies with limited controls may attribute the advantages of those occupations to remote work itself.

Second, organisational quality, and the resulting consequences on employee perceptions, varies across firms. Firms that are capable of implementing effective remote systems often also invest in communication, development, and managerial training. In such contexts, remote work appears successful. In weaker organisational environments, remote work may struggle, relating with Boeri et al. (2026), who show how firm capabilities moderate remote work transitions.

Third, the impact of remote work depends on task structure. Without accounting for coordination intensity and autonomy, aggregate estimates will mix positive and negative effects.

By systematically controlling for compensation, occupation, and validated workplace attributes, and by examining heterogeneity, we clarify that the independent contribution of remote work is smaller and more conditional than headline comparisons suggest.

Conclusion

Remote work has become a defining feature of the contemporary labour market. Yet its effects cannot be understood in isolation of the suite of other workplace features.

Using detailed compensation and occupational data from Payscale and validated engagement measures from Gallup, we show that much of the apparent remote work premium reflects its correlation with higher wages, professional occupations, and stronger workplace environments. Once those factors are taken into account, the independent association between remote work and engagement is considerably smaller.

Nevertheless, remote work’s effectiveness depends on task structure and managerial quality. In low coordination roles with capable managers, it can function well as an amenity. In high coordination environments with weaker managerial support, it may introduce costs.

Location flexibility is one component of a complex organisational system. Engagement and satisfaction are driven less by where employees sit and more by how their work is designed, managed, and rewarded. By reframing remote work within a comprehensive model of compensation, task structure, and workplace quality, our paper helps reconcile conflicting empirical findings and provides a clearer foundation for managerial decision making.

References

Barrero, J M, N Bloom and S J Davis (2021), “Why working from home will stick”, NBER Working Paper No. 28731.

Cullen, Z, B Pakzad-Hurson and R Perez-Truglia (2025), “Home sweet home: The value of remote work”, VoxEU.org, 18 February.

Makridis, C A and J Schloetzer (2026), “How Do Different Remote Work Arrangements”, Management Science.

Ramani, A and N Bloom (2021), “The doughnut effect of COVID-19 on cities”, VoxEU.org, 28 January.

Shah, K, N Bloom, P Bunn, P Mizen, G Thwaites and I Yotzov (2024), “Managers say working from home is here to stay”, VoxEU.org, 18 February.

Zarate, P, M Dolls, S Davis, N Bloom, J M Barrero and C G Aksoy (2022), “Working from home around the world”, VoxEU.org, 8 October.