The Covid-19 pandemic and the related contention measures represented important changes in the organization of work and family life. The adoption of teleworking or working-from-home represented one of the most important changes (and challenges) in the organization of the daily routines of many workers during the lockdown. Three years after the first lockdown in March 2020, teleworking or partial teleworking is still a common practice for a significant fraction of the workforce in Spain. According to the official statistics collected by INE, in 2022 a 15% of the total number of workers worked from home at least occasionally.
The aim of this article is to summarize some stylized facts about workers’ perceptions regarding the possibility to work from home. To this end, in May 2022, we conducted a survey on a sample of 4,000 individuals aged 25 to 50 years old representative of the Spanish population. The survey allows us to provide insights into workers’ experiences regarding teleworking.

In our sample, 36% of the respondents who work, worked from home at least one day per week in May 2022. Our results show that teleworking is not equally distributed in entire working population. It has a higher incidence among high-skilled workers (54%) and those living with a partner (38%) and with dependent children (38%).

Men and women provide a high valuation of working from home, even if women’s value is slightly higher. They both agree or strongly agree that the most valuable attribute of having the possibility to work from home is the improvement in the balance between their personal and family life (88% of women and 86% of men). In addition, both genders also agree that teleworking saves time and money, improves emotional well-being and allows them to be more productive and have more time for hobbies.

We also designed a discrete choice experiment where survey respondents are faced with two job offers. One allows for the possibility to work from home while the alternative one is a full on-site job that provides an equal or higher wage. The result of our experiment indicates that workers are willing to forgo part of their wage, on average 11%, to have the possibility to work from home. Distance to the workplace is also an important determinant of individual choices. Commutes equal or longer than 30 minutes significantly reduce the probability of accepting a full on-site job. 

While teleworking is slightly more valued by women than men, presumably because it eases the work-family balance, our data reveals that men who have the possibility to telework are more involved in home production, both in household chores and childcare.
The analysis is conducted as follows. In the first section, we provide details of our sample. Secondly, we explain the main characteristics of individuals who telework. Then, the analysis looks at the workers’ valuation of telework. After that, we provide an analysis of the relation between telework and housework distribution across genders. We end up with a conclusion section. 

The data in our study was collected through an online questionnaire designed by the authors and carried out by a market research company (IPSOS) in May 2022. We collected information on 4,000 individuals representative of the Spanish population aged 24 and 50 years old. The duration of the survey was about 15 minutes. 

Table 1 presents the descriptive statistics of our sample. About 50% of the individuals are females. The average age of the respondents is 38 years old, 25% of them have a college degree, 73% live with a partner and 60% have children. On average, respondents in our sample have 1.93 children. 79% of our sample is employed at the time of the survey and among those who work, 36% work from home at least 1 day per week. Our survey also contains information on the number of hours worked and the number of hours worked from home. According to the descriptive statistics provided in Table 1, on average, the share of hours worked from home amounts to 20%.   

Figure 1 shows the incidence of telework in our sample. We restrict the analysis to individuals who work at the time of the survey (i.e. 3,101 respondents). To conduct our quantitative analysis, we define telework as a variable that takes value 1 if the respondent reports to work from home at least one day per week and 0 otherwise.

Following this definition, 36% of the workers in the sample declare to work from home at least one day per week. As we can see from Figure 1, the incidence of teleworking does not vary significantly by gender.

However, working from home is more common among those living with a partner (37% vs. 34% for men and 38% vs. 32% for women) and those who have children (38% vs. 34% for men and 39% vs. 34% for women). Figure 1 also reveals an important gap in the probability to work from home between those with and without a college degree (54% vs. 30% for men and 53% vs 29% for women). 

Table A1 in the Appendix reports the estimation results of a linear model for the probability of teleworking. The estimates in the table confirm that the level of education and the composition of the household are important determinants of the probability of working from home. 

Our data also enables us to explore the incidence of telework across socioeconomic groups using alternative measures. Accordingly, Figure 2A reports the share of hours that men and women work from home while Figure 2B focuses on the number of days of telework. We focus on individuals with and without family responsibilities and with and without a college degree.

Both figures suggest that the largest differences occur across education subgroups. Accordingly, individuals with a college degree report a higher share of hours of telework relative to those without a college degree (29% vs. 14% for men and 29% vs. 17% for women) and are more likely to work from home more than 3 days per week (18% vs. 9%). Among those with children, we do not observe important differences in the share of hours worked from home relative to those without children.

When it comes to the number of days worked from home, those with children are more likely to work 1 to 3 days from home (21% vs. 14%), while those without children are more likely to work more than 3 days from home (13% vs. 11%). 

The companion regressions for these alternative measures of telework are reported in columns 2 (share of hours teleworked) and 3 (number of days per week worked from home) of Table A1. From these results, we can confirm that the level of education is the most important determinant for the probability to work from home. 

The results from this section allow us to conclude that the incidence of telework varies across workers’ characteristics. Family composition and the level of education are the strongest predictors of the probability of working from home. Namely, having children or a college degree increase the probability to work from home by 5 and 24 percentage points, respectively. 

Working from home was a rare event before the pandemic (less than 10% worked from home at least occasionally in 2019 according to INE’s official statistics). With the outbreak of the Pandemic in March 2020, many workers were forced to work from home and the share of those teleworking increased to more than 15% in 2020.

During the severe lockdown, working from home was disrupted by the presence of dependent children due to school closures and the lack of a convenient work environment resulting from the unplanned shift from on-site to home-office work. 

Three years after the first lockdown the rate of telework remained quite stable and many workers still do some type of home-office work. As we have seen in the previous section, in our sample, 36% of the employed work from home at least one day per week and among those with a college degree 18% work from home more than 3 days per week. Teleworking conditions may now be more favorable than during the lockdown as workers and firms have had the time to adapt to this “new normal” organizational mode of work. 

Figure 3 reports workers’ subjective valuations of different attributes related to the possibility to work from home. The figure reveals that the ease of personal and family balance is the most valued attribute of teleworking. Accordingly, 88% of women and 86% of men agree or strongly agree with the statement that working from home improves the balance between personal and family responsibilities. 

Workers also value the possibility to save costs when working from home. In the sample, 84% of men and 83% of female respondents agree or strongly agree with the statement that telework saves money and time. Another characteristic that has often been attributed to teleworking is that it reduces stress, and hence improves workers’ well-being. In our sample, 83% of women and 79% of men agree or strongly agree that having the possibility to work from home improves their emotional well-being. Finally, workers also agree or strongly agree with the statement that telework allows them to be more productive at work (76 % of women and 72% of men) and that it releases some time for hobbies (75% of women and 74% of men).   

An important aspect of teleworking is its impact on productivity. Our data does not contain objective measures to quantify the productivity gains or losses of working from home. Previous evidence based on randomized experiments is mixed and suggests that routine and non-routine jobs may differently benefit from workplace flexibility. For example, Angelici and Profeta (2023) show that working from home one day per week increases productivity and reduces leave absences among white-collar workers. In contrast, Emanuel and Harrington (2022) and Battiston et al. (2017) document negative effects of remote work on more routine tasks such as call centers or emergency phone operators.   

The theory of compensating wage differentials in labor economics predicts that workers may be willing to accept a lower wage to have the possibility to work from home if this job attribute is considered an amenity. From observational data on wages and work practices, it is difficult to test this prediction as workers sort into jobs based on unobservable characteristics. Thus, a simple correlation between wages and work practices does not allow to derive accurate conclusions on workers’ monetary valuation or willingness to pay for the possibility to work from home. 

Following Mas and Pallais (2007), our survey includes a discrete choice experiment to gain some insights into workers’ monetary valuation of telework. During the questionnaire, respondents are asked to choose between two hypothetical job offers. One is a full on-site job that does not allow for the possibility to work from home. The other has the possibility of home-office, but it also pays an equal or lower wage. The wage penalty associated with the possibility of teleworking varies randomly across survey respondents and takes values of 2%, 5%, 10%, 15%, 25% and 35%. The distance from home to the workplace in the on-site job also varies randomly and takes values of 10, 20, 30 or 40 minutes. 

To investigate the effect of money and time on workers’ decisions to telework, we estimate a linear probability model where the dependent variable is an indicator that takes value 1 if the worker accepts the offer that allows for telework and pays an equal or lower wage, and 0 if the worker opts for the full on-site job. As control variables, the model includes an indicator for the wage premium assigned to the full on-site job, the distance from home to the workplace and some socioeconomic controls such as gender, age, and indicators for having a college degree, living with a partner, having children and a variable that measures the number of children.   

The estimated coefficients and the confidence intervals (at 95%) are reported in Figure 5. The estimates indicate that college-educated workers are more likely to accept the job that offers the possibility to work from home.

These workers may have a higher income and be willing to give up some of the wage premium associated with the on-site job. It could also be that higher-skilled workers have more suited spaces to work from home or that they value it more. The figure also suggests that having children slightly decreases the probability of accepting a work-from-home job. This result may be related to the increasing difficulties to telework while having children at home.

The estimates also suggest that workers economically value the possibility to work from home and are willing to forgo a fraction of their wage to have this option.

However, the probability to accept the work-from-home job decrease monotonically with the size of the wage penalty. When the wage cut is about 2 to 5 percent, the probability to accept the work-from-home job decrease by about 20 percentage points compared to the situation where there is no wage reduction. This probability drops by almost 40 percentage points when the wage penalty associated with the possibility to work from home increases to 35 percent.  

The estimates in Figure 4 also reveal that distance is an important factor for workers’ decisions. According to our estimates, the probability to accept a work-from-home job increases with the distance to the workplace. This probability increases by 8 and 10 percentage points when the workplace is 30 or 40 minutes away from home.  

As we have seen in the previous section, the possibility to improve the personal-family balance is the most valuable attribute of teleworking. This section explores the implications of teleworking for the distribution of household chores and childrearing activities across genders. As observed in Figure 5A, women’s contribution to childcare is very similar regardless of whether they telework or not. Among those who telework, 48% declare to contriute more than their partners to childrearing activities and 47% report to contribute by a similar amount. Among females who do not have the option to work from home, these numbers are 49% and 48% respectively. 

However, we observe a larger variation among men. Those who work from home at least one day per week are more involved in childcare activities: 24% declare to do more than their partners and 59% report to contribute by a similar amount. In contrast, among those who do not telework, the share of men who contribute more than their partners is only 12%, while the share reporting a similar participation raises to 66%.

Consequently, the share of men doing less than their partners increases by 5 percentage points when they do not telework. 

A similar pattern is observed when we focus on domestic chores, plotted in Figure 5B. Women’s contribution is larger than that of men and unaffected by their working practices while men who work from home at least one day per week are more involved in home production. In this group, the share of men doing more than their partners is 12 percentage points higher than among the group of men who do not telework. 

The evidence in these figures does not permit making causal claims about the effects of teleworking on home production or the distribution of unpaid work within households. However, it suggests that men who telework are more engaged in domestic and childrearing activities than those who do not. 

A final piece of evidence that telework may result in a more equal distribution of domestic production is reported in Figures 6A and 6B. These figures focus on heterosexual couples (70% of the couples in our sample) and report the share of childcare activities and domestic chores done by women separately by those whose partner teleworks and does not telework. As observed in these figures, households where the male partner teleworks have a slightly more egalitarian distribution of home production. The share of females who do more childrearing than their partners is 54% when the partner does not telework and it drops to 48% when he teleworks. These numbers are 57% and 47% for domestic chores. 

The main goal of this piece of work is to provide some facts about the incidence of telework and workers’ perception of this “new normal” organizational mode of work. We use survey data for a sample of individuals aged 25 to 50 years old representative of the Spanish population in May 2022.

Our analysis indicates that telework is not equally distributed across survey respondents. We observe a higher incidence among high-skilled workers and those living with a partner and with dependent children.

Regarding the value of telework, results show that while men and women highly value telework, women’s valuation is slightly larger. For both, the most valuable attribute is the improvement in the balance between their personal and family life.  We conduct a discrete choice experiment to determine the economic value of telework based on wages and distance. Our findings indicate that our survey respondents are willing to forgo part of their wage to have the possibility to work from home. Moreover, long distance to the workplace also significantly reduces the probability of accepting a full-on site job. Finally, our results also suggest that men who telework are more involved in home production, both in doing household chores and childcare.

In conclusion, having the possibility to work from home is likely to be a valuable attribute of jobs as it improves the personal and family life of workers and other aspects of their daily routines. Thus, employers should consider workers’ preferences for telework in designing their job offers and organizational practices. 

ANGELICI, MARTA, and PAOLA PROFETA (2023): "Smart working: work flexibility without constraints." Management Science.

BATTISTON, DIEGO, JORDI BLANES & VIDAL, and TOM KIRCHMAIER (2017): "Is distance dead? Face-to-face communication and productivity in teams." 

EMANUEL, NATALIA, and EMMA HARRINGTON (2023): "Working remotely? Selection, treatment, and the market for remote work." Selection, Treatment, and the Market for Remote Work (June 2023). FRB of New York Staff Report 1061.

MAS, ALEXANDRE, and AMANDA PALLAIS (2020): "Alternative work arrangements." Annual Review of Economics 12.