Remote Work vs. Back to the Office: What can we learn from the Battle of the Sexes?

Goldman Sachs announced earlier this week that they would require all of their employees to be back in the office starting June. 

"We know from experience that our culture of collaboration, innovation, and apprenticeship thrives when our people come together," the Goldman Sachs executives wrote.

Similarly, Jamie Dimon, the CEO of JPMorgan said that “I’m about to cancel all my Zoom meetings. I’m done with it.” requiring the employees to be back by July. 

At the same time, Alphabet/Google announced that it would allow for a significant level of flexibility. Employees can choose where they want to work: their “original” office, a new relocated office, or continue to work remotely as long as their role allows it. Forever. Sundar Pichai, the CEO, wrote his employees:

"The future of work is flexibility...The changes above are a starting point to help us do our very best work and have fun doing it.”

This came on the heels of a recent study that shows that Google saved on the order of a Billion dollars from allowing people to work from home. Furthermore, a recent study showed that “Work From Home” will lift productivity post-pandemic significantly:

“The work-from-home boom will lift productivity in the U.S. economy by 5%, mostly because of savings in commuting time, the study says. The findings suggest the rapid adoption of new technology amid the pandemic will offer lasting economic gains, helping to boost sluggish productivity that has long weighed on global growth.”

Around 5% of the US workforce worked from home pre-pandemic. The expectation is that post-pandemic, the numbers are going to be closer to 20%. 

My take: be ready to go back to the office. 

It’s not what’s optimal for you or your employer. 

It’s the equilibrium.  

The Case of Online vs. Hybrid Education

A lot can be learned from the attempts of colleges to teach in hybrid models during COVID.  When COVID started, we all moved to teach online. Once things settled a bit, there was a lot of pressure on schools to allow students to have some in-person experience. 

The two main issues were:

(i) some students could not go back to be in person. Some due to health restrictions. Some due to the fact the announcements were made late, and they couldn’t find proper housing. And some just liked the flexibility of studying from another location (say, Vail or Hawaii). 

(ii) The second issue was the requirement to have social distancing in classrooms, which allowed, in most cases, only around 24 people in an 80 person classroom. 

While most faculty didn't mind the idea of teaching online, the pressure from students to allow in-person teaching only grew over time. In a survey ran by Wharton, more than 80% of the MBA students expressed interest in having in-person classes.

What happened next will shock you… 

Students just didn't show up in person. 

Many reasons were given: It was very inconvenient. You had to wear a mask. You had to quarantine and couldn't travel between classes (cue “the scream” emoji). But some of the issues ran deeper. Since most faculty continued to teach online, not all classes were taught in person, and students had to juggle between different modes and locations. Since not all students came to class, the faculty that did teach in person had to teach simultaneously in person and via zoom. The audio, video and chat options were all compromised in this hybrid model. 

The experience was not good for those at home or those in the classroom. Several of my colleagues taught with 2 or 3  students in the classroom, while all the rest were sitting at home, sipping their hot chocolates or drinking coffee.

As you may guess, many of these converged into online classes after a short while.  

This is the equilibrium the system converged to. 

Clearly, a sub-optimal equilibrium, when compared to pure in-person, but it is not clear that this is sub-optimal when compared to the hybrid model. In other words, we mostly prefer to teach and study in person when it comes to education. We also can agree that online has some advantages (the multimodality of the chat, the ability to manage breakout sessions) and, if done well, can be quite good.  But we all agree that a hybrid situation is hard to execute well. Not impossible, but it does require significant planning and commitment from all sides. 

This is very predictive of how things will look for firms trying to understand the viability of a remote/office hybrid model.   

The issue is quite similar in the following sense: some people prefer to work from the office. It allows for more collaborations, serendipity in meeting people, the ability to have informal communication, and potentially develop social life and culture.  Some people prefer working from home: less commuting, the ability to choose your own location, and the fact that you don’t need to mix social life and professional ones. After a year of working from home, even those who like one model can agree that the other option can be done well (just like education). 

We can also all agree that in this case, hybrid will be worst than either model.  For example, meetings, where some people sit in the conference room while others call remotely, are much less efficient: It’s hard to hear and know who said what, and not all communication is shared between everyone. For example, side comments made in the conference room cannot be heard by those working remotely.  Every activity requires much more coordination: for example, who will be in the office on what days for what activity.  When everyone was on the Zoom, everything was much simpler. You knew where to find people. Of course, one can join a Zoom call from the office, but then, why am I in the office, and where can I find a quiet place.

In other words, a mess. 

Battle of the Sexes 

In analyzing situations like that, it is useful to use notions from game theory; a theory developed to study how things evolve when people make their own choices.  And the best game to describe it is called Battle of the Sexes.

Let me illustrate the game. Say, my wife loves to watch Avengers Movies (a factually correct statement), and I like to watch French New Wave romantic drama (which is true as well). But we also like to watch movies together. So if both watch a french romantic drama, I get a utility of 3 (arbitrary for the sake of discussion), and she gets a utility of 2. If both watch an Avenger’s movie, she gets a utility of 3, and I get 2. If we both watch a different movie on a different TV, we both get a utility of zero. How can you enjoy a movie without your better half? 

The following diagram illustrates it:

This is the most classical of coordination games. To predict behavior, we look for Nash equilibria. If you watched the movie “A Beautiful Mind,” you may have the wrong idea of what equilibrium means, so I will simplify things here: An equilibrium is a point where if both of you are in, neither of you can do better by unilaterally deviating. You are “stuck”. If we both agreed to watch an Avengers Movie, if I unilaterally leave and watch my favorite french movie, I will do worse than before, since now I get a utility of 0. 

Thus, the two Nash equilibria here are both watching a french movie or watching an Avengers movie. There is a mixed strategy one wherewith ⅗ of the time we play our preferred strategy (I will not get to the technical details here).  Taken from the Wikipedia page of this game:

“This presents an interesting case for game theory since each of the Nash equilibria is deficient in some way. The two pure strategies Nash equilibria are unfair; one player consistently does better than the other. The mixed strategy Nash equilibrium (when it exists) is inefficient. The players will miscoordinate with probability 13/25, leaving each player with an expected return of 6/5 (less than the return one would receive from constantly going to one's less favored event).”

Disregard the math.

The idea is quite simple: The solutions that emerge are either unfair or inefficient.  

Implications for Office vs. Remote vs. Hybrid

Similar miscoordination is going to happen in a hybrid model of working remotely and working from the office. Similar issues of unfairness are going to arise if we allow for full flexibility and choice. It’s not that things cannot be coordinated, but coordination is expensive. It requires adding processes that are unnecessary otherwise. It requires making unfair choices. For example, if we say that everyone should be in the office on Wednesdays, some will surely oppose it. Same for any other day. 

So why am I predicting that most firms will converge to going back to the office? One big difference with the game or education is that whatever equilibrium emerges, one where everyone is working from home, will not be a viable option for most firms.  It is viable for some firms, such as Automattic (creator of WordPress), but even there, when they acquired Tumblr, they realized that it’s much easier to start as a remote firm than to become one. 

Now, don’t get me wrong: Do I like working from home on some days? Absolutely. Will I continue to do it? I hope so. Will I allow my team members to do so after Covid? Yes. I allowed them, within the Wharton rules to do it even before Covid and had team members who worked remotely on some days consistently. 

The issue is that I do not see it as sustainable in equilibrium for most firms, at scale. 

Flexibility and choice always sound like good ideas.

But when you talk about flexibility, you have to solve for the equilibrium, and in this case, the equilibrium is neither fair nor efficient.