A software engineer in San Francisco makes $168K. She moves to Boise, Idaho and keeps working remotely for the same company. Should her salary change?
Companies are split. Google, Meta, and Stripe implemented location-based pay adjustments early. Airbnb, Reddit, and Zillow went the other direction — pay the same regardless of location. Both approaches have data behind them.
Here's what the numbers actually look like.
The two arguments
The employer argument (cost-of-labor theory): We pay market rate for the location where you work. If the market rate for your role in Boise is $120K, that's what we pay — even if we'd pay $168K in SF. We're not underpaying you. We're paying the going rate where you live.
The employee argument (value-of-work theory): My output doesn't change based on where I sleep. The code I ship from Boise is identical to the code I'd ship from SOMA. If my work is worth $168K, it's worth $168K.
Both arguments have merit. But the financial reality for the individual is clearer than the philosophical debate.
The purchasing power math
Let's trace what happens to that $168K SF salary in three scenarios.
Scenario 1: Stay in SF at $168K
- Take-home (California): ~$9,800/month
- Rent (2BR): $2,600
- RPP: 127.3
- Disposable income: $7,200
- Purchasing power: $5,655/month
Scenario 2: Move to Boise, salary adjusted to $130K
- Take-home (Idaho): ~$8,020/month
- Rent (2BR): $1,250
- RPP: 97.1
- Disposable income: $6,770
- Purchasing power: $6,972/month
Scenario 3: Move to Boise, keep $168K
- Take-home (Idaho): ~$10,220/month
- Rent (2BR): $1,250
- RPP: 97.1
- Disposable income: $8,970
- Purchasing power: $9,237/month
The unadjusted Boise engineer has 63% more purchasing power than the San Francisco engineer earning the same salary. The adjusted Boise engineer still has 23% more purchasing power despite a $38K pay cut.
Every version of moving to Boise makes the engineer richer in real terms. The question is how much richer — and who captures the savings.
What companies actually do
The landscape has consolidated since the remote work explosion of 2020-2021. Most large tech companies now use one of three models:
Geographic pay bands. The country is divided into 3-5 tiers. Tier 1 (SF, NYC, Seattle) pays full rate. Tier 2 (Denver, Austin, Boston) pays 90-95%. Tier 3 (everywhere else) pays 80-85%. This is the most common model.
National flat rate. Same salary regardless of location. Less common but growing. Companies like Basecamp, Buffer, and Automattic have used this model for years. The argument: it simplifies everything, eliminates resentment, and attracts talent in lower-cost areas who might not consider you otherwise.
Office-local rate. If you work in the office, you get the local rate. If you go remote, you keep your office-local rate. This rewards presence but doesn't penalize remote workers — it just freezes them at whatever they earned when they left.
The hidden factor: state taxes
When companies adjust salary by location, they're thinking about cost of living. But the employee's take-home is also affected by state income tax, which varies wildly.
Moving from California (top rate 13.3%) to Texas (0%) is a 5-8% take-home bump on its own — before any salary adjustment. A company that cuts salary 10% for a move to Texas while the employee gains 6% in tax savings is creating a 4% net loss. Companies that are thoughtful about this factor in the tax windfall. Many don't.
For employees: how to evaluate a geo-adjusted offer
If your company adjusts pay by location, use this framework:
1. Take the adjusted salary
2. Calculate take-home in the new state
3. Subtract rent in the new city
4. Divide remaining by local RPP / 100
5. Compare to your current purchasing power
If the adjusted number gives you more purchasing power than your current situation, the move is a win — even with the pay cut. If it gives you less, the adjustment is too steep and you should negotiate or reconsider.
The AffordMap cost of living tool runs this calculation for you. Plug in both salaries and both cities and see the comparison.
For employers: the data-driven approach
If you're designing a compensation framework, the BLS Regional Price Parities are the cleanest measure of cost-of-living differences between metros. They're based on actual price surveys, not opinion or real estate listings.
An RPP-based adjustment scales salary in proportion to real price differences. SF (RPP 127) vs. Boise (RPP 97) is a 24% cost difference. Adjusting by that ratio — not an arbitrary "tier" — produces a salary that maintains equivalent purchasing power. The employee keeps the same standard of living. The company saves proportionally.
What the data shows about who adjusts
In practice, companies split roughly into three camps on location-based pay:
Full geographic adjustment (the "cost of labor" companies). Google, Meta, and most large tech companies adjust salary bands by metro tier. A San Francisco L5 engineer might have a band of $180K-$240K, while the same level in Austin is $155K-$210K and in rural Ohio it's $130K-$175K. The adjustments typically run 10-25% between the highest and lowest tiers.
No geographic adjustment (the "pay for the role" companies). Basecamp, GitLab (pre-change), and many smaller startups pay the same salary regardless of location. Their argument: you're doing the same work, producing the same value, and your location is your personal choice. This policy implicitly favors employees in low-cost areas, who get big-city pay with small-town expenses.
Hybrid approaches. Some companies set a national band and adjust within it, or set a floor based on the lowest reasonable cost of living and let high-cost employees negotiate upward. Others use a "base + COL premium" model where everyone gets the same base and high-cost residents get a stipend.
The trend since 2023 has been toward geographic adjustment. As remote work matured and companies had more data, most found that paying San Francisco rates to someone in Des Moines created internal equity problems — the Des Moines employee had dramatically more purchasing power than their San Francisco peer at the same title and salary.
The employee's strategic calculation
If your employer uses geographic tiers, the decision is straightforward: multiply the salary for each tier by the purchasing power in that location and pick the one where the product is highest. Often, that's not the highest-salary tier.
A $180K salary in San Francisco (RPP 127, rent $2,600) produces less purchasing power than $155K in Austin (RPP 102, rent $1,480). The Austin engineer has about $800/month more in real spending capability on $25K less gross salary. Moving from SF to Austin looks like a $25K pay cut but functions as an $10K raise.
If your employer doesn't adjust for location, the math is even simpler: move to the cheapest reasonable place you'd enjoy living and keep the full salary. The purchasing power arbitrage can be worth $15K-$30K annually in effective compensation — the equivalent of a promotion without changing jobs.
The AffordMap cost of living comparison tool shows this side by side. Plug in two cities and your salary to see where the same paycheck goes further.
Salary data from BLS OES. RPP data from Bureau of Economic Analysis. Tax estimates use 2025 published federal and state brackets. Full methodology.