The pandemic has led to a big shift to working from home among people in occupations where it is possible for the jobs or some part of them to be done remotely. There has been much debate about the extent to which such forms of working (or a hybrid model, with some hours done at home and some on business premises) will continue over the longer term. Key questions include the potential impact on employees’ productivity and their job satisfaction, and whether the career trajectories of women and men may be affected differently by a substantial increase in working from home.
In November 2021, we invited our US and European panels to express their views on these issues. We asked the experts whether they agreed or disagreed with the following statements, and, if so, how strongly and with what degree of confidence:
Not all occupations can be done even in part from home. For those that can:
a) Employees who spend two of their days each week working from home are, on average, likely to be more productive over the longer term.
b) Employees who spend two of their days each week working from home are, on average, likely to report higher levels of job satisfaction over the longer term.
c) Having the opportunity to work two to three days a week from home is, on average, likely to be more beneficial for women’s career progression than for that of their male colleagues.
Of our 43 US experts, 41 participated in this survey; of our 48 European experts, 41 participated – for a total of 82 expert reactions.
The impact of working from home on productivity
On the first statement about whether employees spending two of their days each week working from home are, on average, likely to be more productive over the longer term, a majority of panelists say that they are uncertain. Weighted by each expert’s confidence in their response, 5% of the panels strongly agree, 27% agree, 56% are uncertain, 12% disagree, and 0% strongly disagree.
The experts are able to include short comments in their responses, and among only two who strongly agree about a positive impact on productivity, one is Nicholas Bloom at Stanford, who has done a considerable amount of research in this area and whose work was cited by several panelists. He comments: ‘A growing body of research, natural experiments and RCTs [randomized controlled trials], finds moderate levels of WFH [working from home] increase employee productivity.’
Among others who agree, Franklin Allen at Imperial College London says: ‘Depends on the person’s level but provided this is sufficiently high, the reduction in time and effort from commuting is likely a benefit.’ Jan Pieter Krahnen at Goethe University Frankfurt states: ‘I think that over the longer term the productivity is more or less the same, as the innovation effect fades out.’ And Ricardo Reis at the London School of Economics (LSE) directs us to some evidence from Italy that the introduction of ‘smart working’ can have a positive effect on productivity, wellbeing and work-life balance.
Among the majority of panelists who say that they are uncertain, several share the view of Judith Chevalier at Yale: ‘I am pretty certain that there is not enough evidence to have a certain opinion.’ Others point out the difficulty of knowing what the average effect will be. Barry Eichengreen at Berkeley remarks: ‘All we know is some employees yes, other employees (where interpersonal contact is important) no. Average is still uncertain.’ And Richard Schmalensee at MIT responds: ‘Surely depends on the nature of the job and the management style of the firm. I’ve seen no evidence on the overall average.’
Several other experts note the importance of the nature of the work and an individual’s particular position. Christopher Pissarides at the LSE observes: ‘It depends on the nature of work and on the ability of the employee to concentrate on work when there are many alternative uses of home time.’ Robert Shimer at Chicago adds: ‘This will be very dependent on the particular position. The claim is more likely to be true for knowledge workers.’ John Van Reenen at the LSE suggests: ‘It will depend on the type of job one is doing. For creative occupations, social interaction is key so unclear if 40% at home will be better.’
Others comment on the impact on organizational interactions. Daron Acemoglu at MIT remarks: ‘We do not yet know long-run consequences on trust, collaboration and coordination in organizations.’ Anil Kashyap at Chicago notes: ‘There are competing forces, organizational capital depends on some in-person contact; can’t tell if three days in office is enough.’ His colleague Christian Leuz adds: ‘Short-run likely positive, but long-run unclear: effect on innovation, information sharing? Likely also depends on job, lots of heterogeneity.’
Among the panelists who disagree about a positive impact of working from home on productivity, Joseph Altonji at Yale suggests: ‘Maybe less productive on average, but it really depends on the job.’ Kenneth Judd at Stanford adds: ‘Most jobs require collaboration. The formal communication channels of email and Zoom lack the value of spontaneous meetings.’ Daniel Sturm at the LSE concurs: ‘Positive productivity spillovers in the office can likely not be fully replaced by Zoom and other measures.’
The impact of working from home on job satisfaction
On the second statement about whether employees who spend two of their days each week working from home are, on average, likely to report higher levels of job satisfaction over the longer term, a majority of panelists say that they agree. Weighted by each expert’s confidence in their response, 9% of the panels strongly agree, 60% agree, 31% are uncertain, and 0% disagree or strongly disagree.
Among those who agree or strongly agree, several point to evidence. David Autor at MIT says: ‘Lots of evidence that people prefer this arrangement. I suspect that in steady state, it will increase satisfaction.’ Nicholas Bloom adds: ‘Attrition rates from WFH jobs in RCTs are about half those of in-person jobs, and survey data shows a preference to WFH two days a week.’ And Christian Leuz notes: ‘Assuming they get to choose, self-selection suggests outcome. WFH in lockdown was mixed. But with choice, people likely value flexibility’, linking to the Bloom and colleagues’ evidence from an experiment in China.
Others express further caveats about choice and selection, as well as the possibility of longer-term regrets among those who prefer home working. Daron Acemoglu mentions: ‘Provided that it is a choice, not imposition.’ Barry Eichengreen adds: ‘Agree, but these may simply be people who work in occupations offering more flexibility and personal control (pattern is driven by selectivity).’ Anil Kashyap says: ‘Though if promotions lag eventually from insufficient organizational investment, they may have regrets.’ And Christopher Pissarides remarks: ‘In the short term yes, less travel, more time to oneself. In the longer term, maybe not if it delays promotion or pay rise.’
Several panelists comment on why job satisfaction might rise from higher levels of working from home. Darrell Duffie at Stanford observes: ‘Depends on their working conditions at home but if these are good then they should be better off.’ Richard Schmalensee agrees: ‘That’s what most seem to want, and it will likely make them happier. And John Van Reenen notes: ‘Will reduce commutes, which are most stressful parts of day for most people.’
Among the panelists who say that they are uncertain, Austan Goolsbee at Chicago asks: ‘Will employers expect them always to be on call?’ His colleague Lubos Pastor adds: ‘Likely to vary with seniority’; and another Chicago colleague Richard Thaler notes: ‘Again, no way to generalize. At Chicago Booth, there was a strong culture to come to the office. That can easily unravel, which would be bad.’
Kenneth Judd agrees: ‘The advantage of less commuting may balance the lower level of social interaction. And Robert Shimer concludes: ‘Again, a lot of heterogeneity. Revealed preference suggests many workers want to work from home, but this will decrease job attachment.’
The impact of working from home on women’s career progression
On the third statement about whether having the opportunity to work two to three days a week from home is, on average, like to be more beneficial for women’s career progression than for that of their male colleagues. Weighted by each expert’s confidence in their response, 1% of the panels strongly agree, 25% agree, 55% are uncertain, 17% disagree, and 2% strongly disagree.
Among those who agree about the positive effect, several comment on the value of job flexibility for women’s careers. Pinelopi Goldberg at Yale says: ‘More flexibility is good for the careers of women, especially when they have young children.’ David Autor notes that: ‘Claudia Goldin has documented that the economic costs of job inflexibility fall at present most heavily on women.’
Several who agree comment on social expectations of women’s role at home. Barry Eichengreen states: ‘Likely to be true insofar as women, for reasons of tradition and gender roles, are still responsible for more household/childcare duties.’ Darrell Duffie notes: ‘Many women have, historically at least, taken relatively greater responsibility for child rearing. This may provide added flexibility.’
Karl Whelan at University College Dublin declares: ‘I wish this wasn’t true – ideally child-minding duties would be shared equally – but in practice this probably is the case.’ Daron Acemoglu adds a caveat: ‘Provided that there are not perverse incentives, where women will be expected to work even harder to signal work-ahead-of-family commitment.’
Among those who say that they are uncertain, Nicholas Bloom states: ‘This might be correct, but I am not aware of robust evidence on this – indeed more research on this would be invaluable.’ And Larry Samuelson at Yale notes: ‘These key questions will play a role in shaping future working conditions and warrant careful study.’
Others explain how the outcome could go either way. Nicola Fuchs-Schündeln at Goethe University Frankfurt says: ‘On the positive side, it might increase hours worked for women. On the negative side, they will be less engaged in the office networks.’ Christopher Pissarides replies: ‘Pros: easier to combine office work with home production. Cons: more difficult to escape home production than in an office.’
Daniel Sturm, who is also uncertain, notes: ‘While remote working may well hold back the careers of both women and men equally, women will likely make more use of this option.’ John Van Reenen, who disagrees that working from home will benefit women, explains why that might be a problem: ‘Although good for participation, women may end up being less visible in workplace.’ Robert Shimer adds: Work-from-home likely means child-care-during-work for many people, particularly women.’
Others who disagree link to evidence. Christian Leuz comments: ‘Lots of evidence (e.g., COVID) that traditional gender roles in childcare persist, which can lead to gender differences in WFH that hurt [women’s] career progression.’ And finally, Ricardo Reis points to US evidence on alternative work arrangements.
All comments made by the experts are in the full survey results.
Romesh Vaitilingam
@econromesh
January 2022
Question A:
Employees who spend two of their days each week working from home are, on average, likely to be more productive over the longer term.
Responses
Responses weighted by each expert's confidence
Question B:
Employees who spend two of their days each week working from home are, on average, likely to report higher levels of job satisfaction over the longer term.
Responses
Responses weighted by each expert's confidence
Question C:
Having the opportunity to work two to three days a week from home is, on average, like to be more beneficial for women’s career progression than for that of their male colleagues.
Responses
Responses weighted by each expert's confidence
Question A Participant Responses
Participant | University | Vote | Confidence | Bio/Vote History |
---|---|---|---|---|
Daron Acemoglu |
MIT | Bio/Vote History | ||
We do not yet know long-run consequences on trust, collaboration and coordination in organizations.
|
||||
Joseph Altonji |
Yale | Bio/Vote History | ||
Maybe less productive on average, but it really depends on the job.
|
||||
Alan Auerbach |
Berkeley | Bio/Vote History | ||
|
||||
David Autor |
MIT | Bio/Vote History | ||
No data
|
||||
Katherine Baicker |
University of Chicago | Bio/Vote History | ||
|
||||
Abhijit Banerjee |
MIT | Bio/Vote History | ||
|
||||
Marianne Bertrand |
Chicago | Bio/Vote History | ||
|
||||
Markus Brunnermeier |
Princeton | Bio/Vote History | ||
|
||||
Raj Chetty |
Harvard | Did Not Answer | Bio/Vote History | |
|
||||
Judith Chevalier |
Yale | Bio/Vote History | ||
I am pretty certain that there is not enough evidence to have a certain opinion.
|
||||
David Cutler |
Harvard | Bio/Vote History | ||
|
||||
Angus Deaton |
Princeton | Bio/Vote History | ||
|
||||
Darrell Duffie |
Stanford | Bio/Vote History | ||
No evidence. Just a guess based on introspection.
|
||||
Aaron Edlin |
Berkeley | Bio/Vote History | ||
|
||||
Barry Eichengreen |
Berkeley | Bio/Vote History | ||
All we know is some employees yes, other employees (where interpersonal contact is important)no. Average is still uncertain.
|
||||
Liran Einav |
Stanford | Bio/Vote History | ||
|
||||
Ray Fair |
Yale | Bio/Vote History | ||
|
||||
Amy Finkelstein |
MIT | Bio/Vote History | ||
|
||||
Pinelopi Goldberg |
Yale | Bio/Vote History | ||
|
||||
Austan Goolsbee |
Chicago | Bio/Vote History | ||
|
||||
Michael Greenstone |
University of Chicago | Bio/Vote History | ||
|
||||
Robert Hall |
Stanford | Bio/Vote History | ||
|
||||
Oliver Hart |
Harvard | Bio/Vote History | ||
|
||||
Bengt Holmström |
MIT | Bio/Vote History | ||
|
||||
Caroline Hoxby |
Stanford | Did Not Answer | Bio/Vote History | |
|
||||
Hilary Hoynes |
Berkeley | Bio/Vote History | ||
|
||||
Kenneth Judd |
Stanford | Bio/Vote History | ||
Most jobs require collaboration. The formal communication channels of email and Zoom lack the value of spontaneous meetings.
|
||||
Steven Kaplan |
Chicago Booth | Bio/Vote History | ||
|
||||
Anil Kashyap |
Chicago Booth | Bio/Vote History | ||
there are competing forces, organizational capital depends on some in person contact can't tell if 3 days in office is enough
|
||||
Pete Klenow |
Stanford | Bio/Vote History | ||
Jonathan Levin |
Stanford | Bio/Vote History | ||
|
||||
Eric Maskin |
Harvard | Bio/Vote History | ||
|
||||
William Nordhaus |
Yale | Bio/Vote History | ||
Waiting for evidence. Probably small after adaptation.
|
||||
Maurice Obstfeld |
Berkeley | Bio/Vote History | ||
|
||||
Emmanuel Saez |
Berkeley | Bio/Vote History | ||
|
||||
Larry Samuelson |
Yale | Bio/Vote History | ||
|
||||
José Scheinkman |
Columbia University | Bio/Vote History | ||
|
||||
Richard Schmalensee |
MIT | Bio/Vote History | ||
Surely depends on the nature of the job and the management style of the firm. I've seen no evidence on the overall average.
|
||||
Carl Shapiro |
Berkeley | Bio/Vote History | ||
|
||||
Robert Shimer |
University of Chicago | Bio/Vote History | ||
This will be very dependent on the particular position. The claim is more likely to be true for knowledge workers
|
||||
James Stock |
Harvard | Bio/Vote History | ||
|
||||
Richard Thaler |
Chicago Booth | Bio/Vote History | ||
How is anyone supposed to know the answer to this?
|
||||
Christopher Udry |
Northwestern | Bio/Vote History | ||
Seems unlikely in equilibrium ...
|
Question B Participant Responses
Participant | University | Vote | Confidence | Bio/Vote History |
---|---|---|---|---|
Daron Acemoglu |
MIT | Bio/Vote History | ||
Provided that it is a choice, not imposition.
|
||||
Joseph Altonji |
Yale | Bio/Vote History | ||
|
||||
Alan Auerbach |
Berkeley | Bio/Vote History | ||
|
||||
David Autor |
MIT | Bio/Vote History | ||
Lots of evidence that people prefer this arrangement. I suspect that in steady state, it will increase satisfaction
|
||||
Katherine Baicker |
University of Chicago | Bio/Vote History | ||
|
||||
Abhijit Banerjee |
MIT | Bio/Vote History | ||
|
||||
Marianne Bertrand |
Chicago | Bio/Vote History | ||
|
||||
Markus Brunnermeier |
Princeton | Bio/Vote History | ||
|
||||
Raj Chetty |
Harvard | Did Not Answer | Bio/Vote History | |
|
||||
Judith Chevalier |
Yale | Bio/Vote History | ||
|
||||
David Cutler |
Harvard | Bio/Vote History | ||
|
||||
Angus Deaton |
Princeton | Bio/Vote History | ||
|
||||
Darrell Duffie |
Stanford | Bio/Vote History | ||
|
||||
Aaron Edlin |
Berkeley | Bio/Vote History | ||
|
||||
Barry Eichengreen |
Berkeley | Bio/Vote History | ||
Agree, but these may simply be people who work in occupations offering more flexibility and personal control (pattern is driven by selectivity).
|
||||
Liran Einav |
Stanford | Bio/Vote History | ||
|
||||
Ray Fair |
Yale | Bio/Vote History | ||
|
||||
Amy Finkelstein |
MIT | Bio/Vote History | ||
|
||||
Pinelopi Goldberg |
Yale | Bio/Vote History | ||
|
||||
Austan Goolsbee |
Chicago | Bio/Vote History | ||
Will employers expect them always to be on call?
-see background information here |
||||
Michael Greenstone |
University of Chicago | Bio/Vote History | ||
|
||||
Robert Hall |
Stanford | Bio/Vote History | ||
|
||||
Oliver Hart |
Harvard | Bio/Vote History | ||
|
||||
Bengt Holmström |
MIT | Bio/Vote History | ||
|
||||
Caroline Hoxby |
Stanford | Did Not Answer | Bio/Vote History | |
|
||||
Hilary Hoynes |
Berkeley | Bio/Vote History | ||
|
||||
Kenneth Judd |
Stanford | Bio/Vote History | ||
The advantage of less commuting may balance the lower level of social interaction.
|
||||
Steven Kaplan |
Chicago Booth | Bio/Vote History | ||
|
||||
Anil Kashyap |
Chicago Booth | Bio/Vote History | ||
Though if promotions lag eventually from insufficient organizational investment they may have regrets
|
||||
Pete Klenow |
Stanford | Bio/Vote History | ||
Jonathan Levin |
Stanford | Bio/Vote History | ||
|
||||
Eric Maskin |
Harvard | Bio/Vote History | ||
|
||||
William Nordhaus |
Yale | Bio/Vote History | ||
Harder to measure
|
||||
Maurice Obstfeld |
Berkeley | Bio/Vote History | ||
|
||||
Emmanuel Saez |
Berkeley | Bio/Vote History | ||
|
||||
Larry Samuelson |
Yale | Bio/Vote History | ||
|
||||
José Scheinkman |
Columbia University | Bio/Vote History | ||
|
||||
Richard Schmalensee |
MIT | Bio/Vote History | ||
That's what most seem to want, and it will likely make them happier.
|
||||
Carl Shapiro |
Berkeley | Bio/Vote History | ||
|
||||
Robert Shimer |
University of Chicago | Bio/Vote History | ||
Again, a lot of heterogeneity. Revealed preference suggests many workers want to work from home, but this will decrease job attachment
|
||||
James Stock |
Harvard | Bio/Vote History | ||
|
||||
Richard Thaler |
Chicago Booth | Bio/Vote History | ||
Again, no way to generalize. At Chicago Booth there was a strong culture to come to the office. That can easily unravel, which would be bad.
|
||||
Christopher Udry |
Northwestern | Bio/Vote History | ||
Same as above. There will be (lots) of exceptions, though.
|
Question C Participant Responses
Participant | University | Vote | Confidence | Bio/Vote History |
---|---|---|---|---|
Daron Acemoglu |
MIT | Bio/Vote History | ||
Provided that there are not perverse incentives, where women will be expected to work even harder to signal work-ahead-of-family commitment
|
||||
Joseph Altonji |
Yale | Bio/Vote History | ||
|
||||
Alan Auerbach |
Berkeley | Bio/Vote History | ||
|
||||
David Autor |
MIT | Bio/Vote History | ||
Claudia Goldin has documented that the economic costs of job inflexibility fall at present most heavily on women
|
||||
Katherine Baicker |
University of Chicago | Bio/Vote History | ||
|
||||
Abhijit Banerjee |
MIT | Bio/Vote History | ||
|
||||
Marianne Bertrand |
Chicago | Bio/Vote History | ||
|
||||
Markus Brunnermeier |
Princeton | Bio/Vote History | ||
|
||||
Raj Chetty |
Harvard | Did Not Answer | Bio/Vote History | |
|
||||
Judith Chevalier |
Yale | Bio/Vote History | ||
|
||||
David Cutler |
Harvard | Bio/Vote History | ||
|
||||
Angus Deaton |
Princeton | Bio/Vote History | ||
|
||||
Darrell Duffie |
Stanford | Bio/Vote History | ||
Many women have, historically at least, taken relatively greater responsibility for child rearing. This may provide added flexibility.
|
||||
Aaron Edlin |
Berkeley | Bio/Vote History | ||
Yes is a reasonable guess but only a guess.
|
||||
Barry Eichengreen |
Berkeley | Bio/Vote History | ||
Likely to be true insofar as women, for reasons of tradition and gender roles, are still responsible for more household/childcare duties.
|
||||
Liran Einav |
Stanford | Bio/Vote History | ||
|
||||
Ray Fair |
Yale | Bio/Vote History | ||
|
||||
Amy Finkelstein |
MIT | Bio/Vote History | ||
|
||||
Pinelopi Goldberg |
Yale | Bio/Vote History | ||
More flexibility is good for the careers of women, esp. when they have young children.
|
||||
Austan Goolsbee |
Chicago | Bio/Vote History | ||
|
||||
Michael Greenstone |
University of Chicago | Bio/Vote History | ||
|
||||
Robert Hall |
Stanford | Bio/Vote History | ||
|
||||
Oliver Hart |
Harvard | Bio/Vote History | ||
|
||||
Bengt Holmström |
MIT | Bio/Vote History | ||
|
||||
Caroline Hoxby |
Stanford | Did Not Answer | Bio/Vote History | |
|
||||
Hilary Hoynes |
Berkeley | Bio/Vote History | ||
|
||||
Kenneth Judd |
Stanford | Bio/Vote History | ||
NO, IF we assume more women will work from home. Working at home will hurt career advancement compared to those who go to the office.
|
||||
Steven Kaplan |
Chicago Booth | Bio/Vote History | ||
|
||||
Anil Kashyap |
Chicago Booth | Bio/Vote History | ||
See prior comment, same concern applies
|
||||
Pete Klenow |
Stanford | Bio/Vote History | ||
|
||||
Jonathan Levin |
Stanford | Bio/Vote History | ||
|
||||
Eric Maskin |
Harvard | Bio/Vote History | ||
|
||||
William Nordhaus |
Yale | Bio/Vote History | ||
|
||||
Maurice Obstfeld |
Berkeley | Bio/Vote History | ||
|
||||
Emmanuel Saez |
Berkeley | Bio/Vote History | ||
|
||||
Larry Samuelson |
Yale | Bio/Vote History | ||
These key questions will play a role in shaping future working conditions and warrant careful study.
|
||||
José Scheinkman |
Columbia University | Bio/Vote History | ||
|
||||
Richard Schmalensee |
MIT | Bio/Vote History | ||
I can tell stories both ways. Evidence would be nice.
|
||||
Carl Shapiro |
Berkeley | Bio/Vote History | ||
|
||||
Robert Shimer |
University of Chicago | Bio/Vote History | ||
Work-from-home likely means child-care-during-work for many people, particularly women
|
||||
James Stock |
Harvard | Bio/Vote History | ||
|
||||
Richard Thaler |
Chicago Booth | Bio/Vote History | ||
My crystal ball is broken here too. Yes, more flexibility about job design would help women,, especially mothers, but otherwise, who knows?
|
||||
Christopher Udry |
Northwestern | Bio/Vote History | ||
|