The Bureau of Labor Statistics published its May 2025 metro-area wage estimates for San Jose-Sunnyvale-Santa Clara on July 9, 2026, and one number in it is worth pinning to the wall of every Series A engineering budget in the Bay Area: $221,710. That's the average annual wage the survey found for a software developer in that metro. The same job, counted the same way in the same survey nationally, pays $148,100. A senior engineering hire in Silicon Valley costs about 50% more than the identical title almost anywhere else in the country, and that's before a single share of equity is on the table.
What the BLS actually measured
The Occupational Employment and Wage Statistics (OEWS) program is BLS's employer survey, not a self-reported salary site. It draws on responses from roughly 1.1 million establishments, collected twice a year and combined into a rolling three-year model to produce estimates for about 830 occupations across the nation, every state, and roughly 530 metro areas. The May 2025 national estimates were released May 15, 2026 (release USDL-26-0725); the San Jose-Sunnyvale-Santa Clara metro estimates, covering the same May 2025 reference period, followed on July 9, 2026.
One methodology note matters more here than almost anywhere else in this data set: OEWS wages are "straight-time, gross pay, exclusive of premium pay." Base rate, cost-of-living allowances and production bonuses count. Overtime, severance and nonproduction bonuses do not, and neither does equity. That's a rounding error in most occupations the survey covers. In a market where a chunk of senior engineering comp arrives as options or RSUs, it means the $221,710 figure is a floor on total compensation, not an estimate of it.
The San Jose premium, by the numbers
Software developers were among the highest-paid detailed occupations in the San Jose metro, with 87,350 people working under that title, a mean hourly wage of $106.59 and a mean annual wage of $221,710. Nationally, the same detailed occupation (1,687,890 people) had a mean annual wage of $148,100 and a median hourly wage of $65.38. Divide one by the other and the Bay Area premium is almost exactly 50%.
BLS also reports a location quotient, a measure of how concentrated an occupation is in an area relative to the nation as a whole. San Jose's software developer LQ is 7.09, meaning the metro employs developers at more than seven times the national rate. That number explains the wage premium better than cost of living alone does: it's not just that the Bay Area is expensive to live in, it's that the local labor pool is doing the same job at seven times the national density, which means every open req is competing against a denser field of employers chasing the same people.
Software developers weren't the only computer occupation carrying a premium in the release. A few of the neighboring titles, for scale:
| Occupation | San Jose employment | Mean hourly wage | Mean annual wage | Location quotient |
|---|---|---|---|---|
| Software developers | 87,350 | $106.59 | $221,710 | 7.09 |
| Computer & information research scientists | 2,210 | $104.12 | $216,560 | 8.14 |
| Web & digital interface designers | 4,400 | $96.48 | $200,670 | 5.32 |
| Computer network architects | 2,730 | $92.62 | $192,650 | 2.08 |
| Database architects | 1,120 | $91.91 | $191,170 | 2.29 |
| Software QA analysts and testers | 6,480 | $76.94 | $160,040 | 4.76 |
Research scientists carry a slightly higher location quotient than developers, which tracks: that title concentrates around a small number of labs and frontier-model employers, an even narrower slice of the market than "software developer" covers.
What a location quotient of 7 actually tells a hiring manager
A wage premium is a budgeting problem. A location quotient of 7 is a sourcing problem, and it's the one VPs of Engineering underweight when they build a hiring plan around "we'll just pay more." Paying above the national mean gets a company into the local market. It doesn't shrink the market: San Jose's 87,350 developers are still being recruited by every other funded company in a 30-mile radius, and the same names surface in every search, because there simply aren't that many more of them to find. Nationally, the pool is 1.68 million people spread across a country. Locally, it's a much smaller number of people spread across an unusually dense concentration of employers who all want the same skill set at the same time.
Where the average understates the real number
The honest caveat, and the one worth saying on a call rather than burying in a footnote: $221,710 is a mean, not a ceiling, and it's cash only. A staff or principal-level hire, the person a Series B company actually needs to own a system rather than extend one, sits well above that average before a recruiter fee or a signing bonus enters the conversation, and well above it again once the equity component gets priced in. Companies that budget to the BLS mean and then get outbid in the final week of a search aren't reading the data wrong. They're reading only the part of it that's countable.
What this means for the staff-aug math
Picture a 22-person Series A product company in Mountain View with one senior backend req that's been open for nine weeks. Every week it stays open is a week the roadmap doesn't move and a week the team covering the gap is doing two jobs badly instead of one job well. The BLS number puts a real floor under what filling that req with a local, full-time hire costs in cash comp alone, before recruiting fees, before the multi-week search that a location quotient of 7 makes almost inevitable, before equity. None of that is an argument against hiring locally when the role calls for it. It's an argument for knowing what "just hire someone" actually costs before treating a staff-augmentation option as the expensive alternative, when for the right role it's the cheaper one.
Where this doesn't fit
A disclosed hybrid team is the wrong answer for the roles this data describes best: the ones with a location quotient of 7 for a reason, because they need someone embedded in the Bay Area's own ecosystem. A staff research scientist maintaining a relationship with a specific lab, a founding engineer whose equity is the point of the role, or anyone whose job is closing the deals that require being in the room, doesn't get cheaper or better by moving the seat somewhere else. The honest read of this data isn't "never hire locally," it's "know which reqs the San Jose premium is actually buying you something for, and which ones it's just a tax."
For the reqs where it's a tax, staying open nine weeks while the BLS mean keeps climbing is the worse bet. Our team builds disclosed hybrid engagements for exactly that gap, with delivery accountability sitting in Silicon Valley rather than offshore alone. If a senior req has been open long enough that this math sounds familiar, a scoped conversation is a smaller cost than the next month it stays open. Our piece on what actually moves in the first week with an embedded engineer covers what to have ready once you decide to fill it that way, and the wider hiring market data covers why senior roles specifically are the hardest part of that market right now.
Sources
- U.S. Bureau of Labor Statistics: Occupational Employment and Wage Statistics, May 2025 (national release, USDL-26-0725, published May 15, 2026)
- U.S. Bureau of Labor Statistics: OEWS Table 1, national employment and wages by occupation, May 2025
- U.S. Bureau of Labor Statistics, Western Information Office: Occupational Employment and Wages in San Jose-Sunnyvale-Santa Clara, May 2025 (published July 9, 2026)