The Office Is Full and the Collaboration Is Gone: The New York Fed Survey That Undercuts the Office Recovery

The survey is the first hard measurement of a distinction that matters for every office asset in America: people come to the office, but the office no longer reliably produces face time. Organizational sprawl, not absenteeism, explains about 48% of the lost face time, and a further37% comes from colleagues who simply did not show up that day. That is why tighter mandates have not fixed the problem, and why attendance statistics and collaboration statistics have come apart.

Policy is being set on inflation, not on payrolls. The committee held the target range at 3.75% to 4.00% on Oct. 28, 2026 in a 10–2 vote in which Lisa Cook and Beth Hammack preferred a quarter-point points to a Fed already reading the labor market through productivity, with the statement describing productivity growth as strong and job gains as keeping pace with the workforce. With CPI at 3.7% in September, the Nov. 6 employment report is the last soft input before December.

The practical consequence is that a soft payroll print on Nov. 6 is a weaker signal than it looks. If output is holding up while headcount does not, unit labor costs fall and the disinflation arrives without a recession, which would leave a committee that just held at 3.75% to 4.00% with two hike dissents looking less comfortable than the labor data alone implies. The reading that fits every New York Fed survey at once is not a weak economy but a differently organized one.

The office recovery is real in volume and thin in price. Leasing activity is up 6.4% over12 months with gateway markets up 13%, but supply has been withdrawn faster than demand has returned: only 23 million square feet is under construction after 45 million square feet left the inventory pool since the late-2023 peak. Landlords are filling buildings and conceding rate, so attendance has returned to the buildings without returning to the rent roll.

The commuting economy is not the casualty. Mandates preserve the trip to the office even when they do not preserve the meeting, so downtown demand has behaved like a normal consumer economy rather than like a casualty of remote work. Starbucks and Uber are the cleanest read-throughs, and both point to traffic-led growth with the commute intact.

Japan has run the hybrid experiment far longer, and its results are the uncomfortable precedent for U.S. landlords. Mitsui Fudosan's own indicator put the Tokyo metropolitan office vacancy rate at 1.6% at its March 2026 fiscal year-end, and the group grew revenue from operations 3.2% to ¥2,709.7 billion with business income up 11.6% to ¥445.1 billion.

The two horizons point in opposite directions. Over days to quarters, the market trades the Nov. 6 employment report and the December meeting using statistics that cannot see collaboration. Over one to three years, the question is whether the reallocation of work shows up as measured productivity gains, which the FOMC statement already assumes, or as a slower economy that eventually reaches rents and headcount.