Nikil's September 2026 Bay Area Corporate Housing Forecast: 7 Predictions, Plus a Change to How I Grade

I'll come back on October 1 and grade every call below.

I owe you a worse admission first. I can't grade August at all.

The August grade, such as it is

Seven calls. Seven incompletes.

I predicted that 85% of my September new-grad placements would be committed by August 21. That Mountain View and Sunnyvale furnished one-bedroom availability would climb back above 15%. That 60-day rates would land 5 to 9% below July. That budget and rate-card requests would run 2x July volume. That three assignments would extend past twelve months. That Palo Alto and Menlo Park volume for early September would beat early August. That pet-friendly availability would stay under 12%.

Every one of those required an analysis I don't run in the ordinary course of business. So at the end of August there was no number sitting anywhere to check any of them against, and I'm not going to reconstruct figures after the fact to make a scorecard look tidy. Reconstructed numbers are just invented numbers with extra steps.

Zero for seven, not because the calls were wrong but because they were unscoreable. That's on how I wrote them.

What changes now

Two rules for every call from here.

It either resolves against a public source that you can check yourself without taking my word for anything, or it resolves against one number I can read off a dashboard in under a minute. Anything needing a special study doesn't make the list.

The consequence is that these are more modest than August's. A forecast you can score is worth more than a bolder one you can't, and after a month like this I'd rather rebuild the habit than protect the drama.

The seven calls for September

1. The $100,000 H-1B fee is still not in force at month end.

The district court vacated the implementing policy, and on July 24 the First Circuit declined to stay that ruling while the appeal proceeds. Separately, Proclamation 10973's twelve-month restriction expires September 20 unless it's extended. Both are matters of public record.

The call: as of September 30, the $100,000 fee remains not in effect.

2. Dreamforce tightens San Francisco and leaves the Peninsula alone.

Dreamforce runs September 15-17 at Moscone. It reliably compresses San Francisco lodging for that week. My inventory sits south of it.

The call: I turn away no South Bay or Peninsula corporate housing request for September 14-18 due to lack of inventory.

3. The Stanford calendar pulls Palo Alto and Menlo Park demand into the first half of the month.

New undergraduates arrive September 15, returning students September 17, and instruction begins September 22. Visiting faculty, medical fellows, and relocating families track that calendar closely.

The call: my Palo Alto and Menlo Park furnished inquiry count for September 1-15 exceeds my count for September 16-30.

4. Q4 and 2027 budget season starts in earnest.

September is when mobility teams stop planning this year and start pricing next year. It's the single most reliable seasonal pattern in this business.

The call: I receive at least three separate 2027 budget, benchmark, or rate-card requests during September.

5. Sunnyvale demand holds or grows.

The AI expansion has concentrated in Sunnyvale and Mountain View, with Databricks alone assembling roughly 635,000 square feet across three Sunnyvale buildings. I've written about what that does to housing demand. This is the conservative version of that thesis.

The call: my September placement requests for Sunnyvale meet or exceed my August count.

6. Rates move sideways, not down.

Post-summer softening is the conventional expectation. I think this year's demand picture is strong enough to offset the usual seasonal give-back, but I'm not predicting an increase.

The call: my average 60-day rate across Mountain View, Sunnyvale, Palo Alto, and Santa Clara lands within 5% of the August figure, in either direction.

7. October gets committed before October arrives.

The single best predictor of a bad month is a client who hasn't booked it yet. This is the number I watch most closely and the easiest one to check.

The call: by September 30, at least half of my known October arrivals are committed rather than pending.

What could break these

An extension or reissue of Proclamation 10973 in a different legal form is the biggest single risk, and it would hit calls 1 and, with a lag, 5. The courts vacated the policy implementing the fee, not the proclamation, so there's room for another attempt on different footing.

A sharp move at the September 16 Fed meeting could cool hiring sentiment enough to soften call 6, though housing rates respond to that with a lag measured in quarters rather than weeks.

And if fire season turns in late September, everything about lead times gets less predictable, which mostly threatens call 7.

How to use this

If you're a mobility lead, call 7 is the one to copy. Look at your October arrivals today and count how many are committed versus pending. If it's under half, you have three weeks to fix something that otherwise becomes a scramble, and the flexible-hold structures I've described before are the tool for it.

Call 4 is the calendar reminder. If you haven't started your 2027 numbers, this is the month, and the budget framework is where I'd start.

I'd rather publish a zero-for-seven month honestly than quietly drop the scorecard and hope nobody remembers. October 1, with real numbers.

Planning Q4 or Q1 arrivals into the Peninsula or South Bay? Request a free consultation and we'll look at your inventory needs before the market decides for you.

Sources

• August 2026: $100,000 H-1B Fee Blocked Again — Klasko Immigration Law Partners
• Dreamforce 2026 — Salesforce
• Stanford Academic Calendar 2026-2027 — Stanford University
• Databricks Expands Again in Bay Area with 305K-SF Sunnyvale Lease — Connect CRE
• AI Boom Drives Office Leasing Surge in San Francisco Bay Area — CBRE
• FOMC calendar — Federal Reserve