Derek Thompson has an interesting piece in the Atlantic today looking at recent trends in America’s gradual return to normal. The good news is that lots of activities—sporting events, dining out, travel, even going to the movies—appear to be headed for a recovery. The chart above shows some of this.

But that red line along the bottom of the chart is a measure of people’s return to office buildings in cities around the country based on their key card swipes. As you can see, that curve looks a little different. Thompson argues that white collar office work probably isn’t going to return to pre-COVID normal because a lot of people don’t want it to. We may have reached what he calls “peak office.”

“I talk to hundreds of companies about remote work, and 95 percent of them now say they’re going hybrid, while the other 5 percent are going full remote,” Nick Bloom, an economics professor at Stanford University, told me. The exceptions to the rule, such as Goldman Sachs, are scarce.

“The number of person-days in the office is never going back to pre-pandemic average, ever,” Bloom told me. After two years of working from home, he said, employees don’t just prefer it. They also feel like they’re getting better at it. Despite widespread reports of burnout, self-reported productivity has increased steadily in the past year, according to his research.

In the next decade, U.S. workers will spend about 25 percent of their time working from home, Bloom says. That’s 20 percentage points higher than the pre-pandemic figure, leaving companies with an important choice: sign for significantly less office space, or accept that significantly more of your space will go unused on a given day.

Bloom is betting strongly on the latter. “Office occupancy has plummeted, but corporate demand for office space is down only about 1 percent,” he said.

Here I’m going to disagree with the Stanford economics professor a bit. A friend of mine who works for a major engineering firm is now working from home 2-3 days a week even though offices have been open for months. That sort of schedule has become pretty common for people at the company and at rival companies during the pandemic. The higher ups in the company’s corporate offices have realized that the 2nd biggest cost on their balance sheet, the price of retail office space, can be reduced substantially if you assume workers won’t all be in the office everyday. The potential savings at my friend’s company could be millions of dollars per year because they lease office space in cities across the country.

However, office leases are multi-year agreements. So if my friend’s company, or any company, decided sometime last summer that downsizing space was a good idea, that may not be reflected very widely yet simply because existing leases haven’t allowed for it in most cases. So, I think it’s very possible the corporate demand for space will continue to shrink to match the new normal of lower office occupancy but getting there could take a while.

Perhaps if white collar employees were clamoring to return to a regular 5-day work week things would be different but it looks instead like a lot of people have come to appreciate not having to commute and being home for lunch and when their kids get out of school. The article suggests this could become a perk companies may begin to offer, i.e. if you join our team you’ll only be in the office 3 days a week. But this change could have all sorts of other implications for cities who are built around downtown office space.

If office occupancy never recovers, downtown areas will experience an extended ice age. Emptier offices will mean fewer lunches at downtown restaurants, fewer happy hours, fewer window shoppers, fewer subway and bus trips, and less work for cleaning, security, and maintenance services. This means weaker downtown economies and less taxable income for cities.

This makes sense and yet I don’t think people working from home two days a week are necessarily going to stop ordering lunch out. The result may be less demand downtown and more demand for delivery in residential neighborhoods. Delivery services had a big year in 2020 and 2021. Maybe those levels won’t be sustained but the use of those services could still be up relative to the pre-COVID world.

The more significant change might be to public transportation. During the early height of the pandemic in New York City, subway ridership dropped more than 90% as lockdown orders kept people at home. As of today, those figures have recovered somewhat but according to recent figures from MTA, they are still at only about 60% of pre-COVID levels:

And the MTA is doing pretty well compared to BART ridership in San Francisco which is about 1/4 to 1/3 of what it was pre-pandemic during the work week (the higher totals are weekends):

Just two weeks ago BARTs board of directors described a very uncertain future for the system thanks in part to telework:

New projections shared at a BART Board of Directors meeting Thursday underscore the challenge that the region’s public transit agencies continue to face as telework and low office occupancy have contributed to one of nation’s slowest transit recoveries.

The agency expects to exhaust the pandemic federal aid that’s helped sustain BART operations by fiscal 2024, when it will start incurring a budget deficit of $48 million, according to a “base case” projection. Under this scenario, the deficit is projected to balloon to $1.2 billion by fiscal 2032, absent new sources of revenue.

Fare revenues, which accounted for about 65% of the agency’s budget before the pandemic, are not a likely source of relief. Revised projections by BART expect the agency to recover 70% of its average pre-COVID weekday ridership in the next 10 years.

That’s a big problem for a couple of reasons. First, even workers who no longer rely on the system during the week may still want it to be there on the weekends for shopping or a night out. That ability to travel around downtown easily could keep some businesses afloat, but obviously if the system is deep in the red and can’t sustain itself then the entire downtown economy becomes increasingly isolated from the consumers required for it to survive.

Second, and this is related to the first problem, the people who still use public transit to get to work at jobs where telecommuting isn’t an option (nurses, police officers, restaurant workers, teachers, etc.) need the system to get to work. But if BART in San Francisco collapses for lack of ridership then those workers have few good options and may need to seek other jobs. That would also create another big stress on the survival of downtown businesses.

Working from home is appealing to a lot of white collar workers and to the companies that could save money by leasing less office space. That seems to be where things are headed, but the impact those changes will have on cities built around high-volume public transportation could be pretty dramatic.