The costs of commuting

Something that’s happened to me in the last year, and one of the reasons why I’m writing less, is that my weekdays have got three hours shorter.

This is, of course, not literally true. What’s happened is that I’ve started an in-office job that has a 90 minute commute.

But it feels true. Actually it feels like an underestimate - sure, I’m only spending 3 hours commuting, but there’s also transition time on leaving and arriving, and energy cost. The office is also less convenient for e.g. food. I have to either prepare a packed lunch or take time out to go get food, which in the most optimised case is probably costing me 15 minutes.

Then I get home and I’m exhausted from all of this - being in an office is much more draining than the corresponding work from home, even not counting the drainingness of the commute. It also means I’m starting my evening much later than I would otherwise - I’m trying to leave the office by 5, but that means the earliest I’m home is 6:30, and often later than that - that 5:00 departure requires careful policing of myself and others to maintain.

Of course, the hours spent on commuting don’t literally disappear. Some of them are spent on walking, which is a pleasant enough activity. The time on the train is time I could in theory use productively. I do use it to listen to fiction or podcasts, and sometimes read books or papers. If I get a seat and the train wifi works. I might be able to work. That’s about 50 minutes of work, sometimes, and typically only viable if I get one of the later trains in (which of course costs much more useful in-office work time) as the earlier ones rarely have a seat, or at least one spacious enough to open my laptop on.

It very much feels like I’ve gone from having a weekday to having a workday and some dead time, and also that the workday is shorter and less productive.

This makes it sound like I hate going into the office. I actually don’t. If you gave me a teleporter, I’d probably choose to go into the office… at least 3 days a week. Maybe even 4 most weeks, and 5 some weeks. It’s easy to say “I get so much more done at home” and it’s true, but it’s also harder to coordinate with others and less motivating. When I was working from home by default, I had a lot more days when I got nothing done. Those still happen, but not nearly as often.

I still wouldn’t choose to go into the office every day. It’s not a cost of commuting per se, but even just being in the office comes with a certain cost. It’s a much less focused space because of the number of things going on around me, and there’s also less flexibility in my schedule. For example, left to my own devices, I tend to do a bunch of work when I first wake up, get into a flow state for a few hours, and then shower. Working in my pyjamas is somewhat less viable in the office.

Writing all of this down, I do note two things that it used to be that everyone worked like this. Sure, 90 minutes is an unusually long commute, but most of the costs aren’t actually that dependent on it being 90 minutes. Many of them would go away if it were 5 minutes, but I think any commute over about 30 minutes would carry most of these costs and a typical London commute is more like an hour. Maybe this has got worse - commute lengths have definitely gone up as people have got priced out of London - but I don’t know.

Even today, most office workers probably work most days in an office. In London it’s about 3 days per week and that’s unusually low.

Certainly I feel stuck between a rock and a hard place on changing this. We’ve got some plans to move a little closer to London, but it would actually only cut about 10 minutes off my commute. Anything closer becomes much costlier, and also we’re rather tied to a certain transport radius around my daughter’s school.

I often find myself a bit “housing theory of everything” about social problems, and I think that’s a big part of this, but I think there’s also a “transport theory of everything”. The two problems are pretty tightly entangled, but not synonymous. I don’t think you could fix my commuting problem much with better transport, but mine is anomalously (though by no means uniquely) bad.

Reliability aside, I’m probably in one of the versions of this with some of the best available transport infrastructure in the world for convenience, and it still sortof sucks. The point isn’t really that commuting means that you need to improve transport infrastructure (though in many places it does), but that transport defines a sort of “basin of viability” for in person work.

I have no satisfying conclusion here other than that hybrid work is obviously the best choice for almost everyone who is capable of being in the office at all. I also think people are missing out on some great opportunities by not enabling at least the possibility of fully remote, but sometimes it’s correct to miss out, and that’s a whole separate post.


  1. The astute observer may notice that if I’m leaving at 5, I must be arriving by 9 to get a full workday in. I am, of course, not arriving at 9. I generally arrive some time between 9:30 and 11:00, and do not vary my departure time with my start time. My stance that is that if you insist on in-office work, you get to bear some of the costs of that commute, so I work a shorter work day. TBH I probably do enough “out of hours” work that it makes up for it, but it’s extremely noticeable how much more I get done on days I can work from home.

    ↩︎
  2. Specifically the commute involves about half an hour of walking, so about an hour per day in total. Which is all well and good - I generally tried to do an hour walk most days when working from home. But four fifteen minute walks is a much less pleasant experience than an hour walk, and has few of the same benefits.

    ↩︎
  3. Indeed I’m writing this on the train, in what is technically a problematic use of my work computer for personal activity.

    ↩︎
  4. My phone does not have reliable coverage across the line, even outside tunnels

    ↩︎
  5. Especially if you don’t count meetings as real work. I do count meetings as real work, but they also come at a significant cost to my other work.

    ↩︎
  6. Though I obviously would not say no to reclaiming two hours of my day.

    ↩︎
  7. And of course non-office workers will typically work on site and are probably going to grouse at me to quit my bitching about my cushy job. Sorry, shan’t.

    ↩︎
  8. The real benefit will actually be frequency and reliablity of trains, which is a whole separate cost to being out of London. I’ve got a reasonably good train line for my commute… but I only have the one. It runs every half an hour and is often disrupted. Moving would allow us to be on multiple lines.

    ↩︎
  9. Whose commute is also very costly to her. We need to fix that rather more urgently than we need to fix mine, but it’s equally hard.

    ↩︎
  10. A few things could optimise it a bit. A high speed rail line along the same route could maybe save half an hour, especially if you cut out a stop or two. Though TBH if you were to cut out an obvious stop on the train I get, it would be City Thameslink, but if you were to cut out two then my town would be the second one to cut. Ultimately though without new stations, a teleporter from my home station to my work station would still only get the commute down to a bit under half an hour.

    ↩︎