Three letters alone can describe the last 12 months for millions of people: W-F-H. Working from home is no longer a niche arrangement. Overnight, whole sectors of the economy abandoned the office in favour of the kitchen table, sofa or garage.
There are, of course, pros and cons to WFH. One size does not fit all.
On one hand, WFH removes the commute, grants people more time with their families, and allows employers to reduce office space.
On the other hand, some us do not have home environments that are conducive to productive working, and some organisations have found that it is harder to shape culture remotely. We may all be in the same storm, but we are not all in the same boat.
In this context, a large UK-based charity asked Oxera to advise on the use of behavioural science to encourage its employees to continue to WFH (while continuing to offer staff the choice of location).
This is not so much about ‘manipulating’ peoples’ choices. There is no such thing as ‘neutral’ way to present information. The alternative to designing new messaging is for the previous (outdated) messaging to continue to influence behaviour. Without updating the information provided to employees, employees might not realise that their perceptions of employer expectations may need updating.
Behavioural factors at play
Behavioural economics tells us that we all make decisions subject to ‘cognitive limitations’, often using mental shortcuts to conserve cognitive effort. The result of these mental shortcuts is that the framing of a decision—the so-called choice architecture—really matters.
There are three behavioural factors that are especially relevant to the decision to WFH.1
- Perceptions of employer expectations. For example, perceived expectations over hours worked can lead to ‘presenteeism’. This is not necessarily ‘irrational’—the evidence shows that in an environment where only some employees WFH, working from home can limit salary growth.2
- Social norms. People are influenced by how other people behave. If the social norm is perceived to be slow to revert to pre-pandemic behaviour, people will be slower to revert to their previous behaviour (and vice versa).
- The default option. People are more likely to choose the ‘default’ option, as this takes less cognitive effort (and the default option can communicate recommendations or expectations). People naturally conform to their habits, and our habits are likely to have changed during the pandemic as we adapted our behaviour (e.g. to WFH).
There is likely to be a diverse range of employee preferences over the choice to WFH or not. Employees’ views over ‘working from anywhere’ are likely to remain the key determinant of whether they choose to work from the office or another location. The use of nudges may have an effect at the margin, but is unlikely to result in a radically different outcome.
Moreover, the reaction may not be positive if employees perceive that their employer is trying to nudge them in a direction that is counter to their (strongly-held) preferences. Therefore, an important first step (before implementing any nudges) would be to gather employee views. One way of doing this would be through an employee survey. For example, Nationwide Building Society found that 57% of employees would like to WFH fulltime, 6% would like to work from the office fulltime, and 36% would like a blend of both.3
Nudges: carrot and stick
A nudge is an intervention designed to change the choice architecture without forcing people to make a certain choice.
There are two types of nudge:
- the ‘carrot’—smoothing the decision-making process towards a desired option (i.e. decreasing the friction involved in choosing a desired option);
- the ‘stick’—adding hurdles in the decision-making process towards an undesired option (i.e. increasing the friction involved in choosing an undesired option).
The ‘sticks’ (nudges that increase friction) are sometimes known as ‘sludges’ if they result in good decisions being harder to make.4
However, it is possible to increase friction with the aim of making good decisions more likely. Indeed, increasing the friction involved in making a poor decision is arguably comparable to decreasing the friction involved in making a good decision.5 Thus, we should not rule out the ‘stick’ without considering it more carefully.
Nonetheless, it is true that using the ‘stick’ risks greater unintended consequences (see below) than the ‘carrot’, and so should be carefully considered before implementation.
The ‘carrot’ involves smoothing the decision-making process towards WFH. This could work in a number of ways, such as the following.
- The perceptions of employer expectations are a vital determinant of employee behaviour. The tone from the top is crucial. If senior management work from the office, then we might expect the people who work for them to also wish to attend to office, and so on.6
- Social norms are likely to be important in influencing whether employees WFH. If (and only if) a social norm is helpful, then it could be communicated (e.g. ‘97% of us WFH last week’).
- Making WFH the ‘default option’. This could be through the ‘tone’ of internal communications, emphasising that WFH is the cultural norm.
- Loss aversion could be harnessed to re-enforce this message. Language that emphasises, for instance, ‘Travelling to the office’ will encourage greater WFH than language that emphasises ‘Coming back to the office’.
The ‘stick’ involves introducing more friction in the decision-making process towards working in the office. In this case, more cognitive effort is required (than before) when working from the office. For example, the employer could move towards hot-desking so that employees select a new desk each day. This removes one element of habit (not the same familiar desk every day), and introduces a new cognitive task when entering the office.7
Ultimately, a ‘carrot and stick’ approach is more likely to be effective than the ‘carrot’ or ‘stick’ alone. Using the ‘carrot’ alone would minimise the risk of unintended consequences (see below).
First, WFH is likely to have disadvantages as well as advantages. For example, company culture may be dissipated or diluted, and some employees may not be able to WFH. It is highly unlikely that ‘one size fits all’.
Second, nudges that increase the frictions from working in the office (the ‘stick’) risk causing unintended consequences. Such interventions could reduce the productivity of office working, or even resonate poorly with employees, especially if employees perceive that their employer is trying to nudge them in a direction that is counter to their (strongly-held) preferences.
Top-down or bottom-up?
Conducting an employee survey into WFH preferences would be an insightful first step for any organisation. Careful design of the survey choice architecture would ensure that it accurately reflected employee preferences. After all, listening to the diverse range of employee voices will always be constructive and enlightening.
Indeed, putting employee engagement and ownership at the heart of WFH decisions (and flexible working arrangements more generally) will help foster a positive, diverse, and inclusive culture.
 See also Oxera (2020), ‘How quickly will customers revert to pre-pandemic behaviour?’, Today’s Agenda, May. https://www.oxera.com/insights/agenda/articles/how-quickly-will-customers-revert-to-pre-pandemic-behaviour/
 Golden, T. D. (2020), ‘Is there a price telecommuters pay? Examining the relationship between telecommuting and objective career success’, 116, Part A.
 Nationwide (2021), ‘People power: 13,000 employees put in control as nationwide rolls out “work anywhere” approach’, 25 March. https://www.nationwidemediacentre.co.uk/news/people-power-13-000-employees-put-in-control-as-nationwide-rolls-out-work-anywhere-approach
 Thaler, R. H. (2018), ‘Nudge, not sludge’, Science, 361:6401. https://science.sciencemag.org/content/361/6401/431.summary
 Mills, S. (2020), ‘Nudge/sludge symmetry: on the relationship between nudge and sludge and the resulting ontological, normative and transparency implications’, Behavioural Public Policy, pp. 1–24. https://www.cambridge.org/core/journals/behavioural-public-policy/article/nudgesludge-symmetry-on-the-relationship-between-nudge-and-sludge-and-the-resulting-ontological-normative-and-transparency-implications/33F75B873929827F6DDCE709C7315F68
 Burt, E. (2019), ‘Nudge theory can help change your employees’ behaviour (without them even realising)’, People Management, 28 March.
 Although hot-desking may not be compliant with social distancing if the desks are not cleaned regularly enough.