Real Estate

The forecast peak against the actual

Moody’s projected peak vacancy versus what Q1 2026 actually printed, %.

← Back to US Office Vacancy

About this data

Moody’s Analytics research projected national office vacancy peaking near 24% in early 2026. The actual Q1 2026 print was 21% — three points below, and still creeping up rather than peaking. Distress forecasts in this sector have consistently run ahead of the realised numbers, which matters because those forecasts drive lender provisioning and valuation marks.

The forecast peak against the actual

The forecast peak against the actual

Moody’s projected peak vacancy versus what Q1 2026 actually printed, %.

View data & sources →

Data table

The forecast peak against the actual — forecast_vs_actual data table (US Office Vacancy)
pct basis series source_ref value_basis
24 Moody’s projected peak forecast_vs_actual moodys-q1-2026 Moody’s Analytics research projected national office vacancy peaking near 24% in early 2026 — FORECAST
21 Actual, Q1 2026 forecast_vs_actual moodys-q1-2026 Actual Q1 2026 office vacancy 21% — three points below the projected peak

Methodology & sources

Last updated: Aug 14, 2026

Methodology

Vacancy is the share of office space available and unoccupied across the markets Moody’s tracks (79 US markets as of the Q1 2026 reading). Vacancy is NOT the same as availability, which also counts space being marketed but still occupied, and is not the same as utilisation, which measures whether leased space is actually being used. Utilisation is materially worse than vacancy in the post-2020 market — a leased but empty floor counts as occupied here. The x-axis mixes an annual figure (2020) with quarterly readings, so the spacing is not uniform. Each point’s label states its period. The 20.1% print in Q2 2024 was the first time the series crossed 20% in roughly fifty years of Moody’s data. CAVEAT: the forecast-vs-actual chart compares a projection with an outcome and should not be read as a time series. Moody’s projected a ~24% peak in early 2026; the Q1 2026 actual was 21% and still rising slowly rather than peaking. Distress forecasts in this sector have run ahead of realised numbers, which matters because those forecasts feed lender provisioning and valuation marks. Read alongside datacenter-real-estate: same broad asset class, same quarter, one at a record high and the other at a record low. Re-verified 2026-08-11.

Comparisons are informative, not definitive. See each source for definitions and limits.

How relevant was this information?