Geographic pay is back, and it's more granular than before

In October 2020, Reddit eliminated geographic compensation zones for its US employees, tying pay to high-cost-area ranges like San Francisco and New York regardless of where someone actually lived. Gumroad, Basecamp, Sourcegraph, and dozens of other companies made similar announcements within months of each other, part of a wave of employers publicly committing to pay the role, not the zip code, that Remotive later rounded up.

The reversal started almost immediately. WorldatWork's 2021 Geographic Pay Policies Study found that 67% of workers already expected their pay to reflect their location by April 2021, and 44% of organizations with existing geographic pay policies were actively reconsidering them, either expanding or consolidating how differentials applied. By September of that year, 46% of professionals said their company had already adjusted pay for employees who'd relocated outside their metro area, and another 24% expected it soon, in a survey of more than 4,000 verified professionals by Blind. Only 30% believed their employer would make no geography-based change at all.

Four years later, the reversal has widened into something far more specific than a blanket policy change. 81% of organizations now factor geographic pay differences into their salary structures, most applying a 5% to 10% band, per a survey of 158 organizations by the Economic Research Institute, and for a growing number of companies, even that band is too broad.

Pay is getting more precise, and less forgiving of shortcuts

26% of organizations set pay off one flat rate tied to their headquarters or the national median, per Payscale's 2026 Compensation Best Practices Report, and don't adjust for location at all. Among those that do, only 24% price by geographic pay zone, lumping similarly priced locations together; 60% price by individual city instead, according to Culpepper's own survey of North American compensation practices.

But "city" usually means a Metropolitan Statistical Area, and MSAs weren't built for pay analysis. In the California Bay Area, East and North Bay pay averages 109.7% of the national average, considerably lower than Silicon Valley and San Francisco specifically, even though a standard MSA groups all of it together, according to Culpepper's own analysis. Culpepper's own zone system erases real cost differences the same way an MSA does: it groups places with genuinely different labor costs into one priced-as-identical category. Its Pay Zone 3 puts Atlanta, the Research Triangle, Los Angeles, Chicago, Philadelphia, and Austin all in a single tier.

Only pricing down to something smaller catches gaps like these. ZIP-code pricing means working with a price point for every one of the 41,554 ZIP codes the US Postal Service counts. That precision has a real cost, though. Managing dozens of city-specific differentials already creates payroll setup errors, confusion among HR teams and recruiters, and inconsistent offers to candidates, according to SalaryCube's own compensation guidance, which recommends zones specifically to avoid that burden, even at the cost of blurring gaps as real as the Bay Area's.

Precision only helps if it stays current, and even the broadest differentials already can't keep up. San Francisco's sits at 26% above the national average. A $30,000 salary there needs a 35% premium just to match the national average, according to Mercer, a 9-point gap the 26% figure doesn't account for. That gap isn't unique to San Francisco. Local minimum wage increases raise pay at the low end of a salary range without touching the high end, according to ERI's own research, so a flat differential that was accurate when it was set can go stale at the bottom of the range even without a shift as dramatic as San Francisco's.

Austin used to cost less than the national average, which is why companies set a lower pay differential for employees there. That's no longer true. An influx of tech companies and high earners has pushed Austin's housing costs close to the national average, the same Mercer analysis found. A company still using Austin's old differential pays employees there as if Austin still costs what it did years ago.

What it costs the people in the middle

Cache Merrill, founder of Zibtek, a distributed software development company, had followed a location-blind pay policy since the pandemic, treating an employee's zip code as irrelevant to what they were paid. This year, that changed for four employees. "The problem wasn't in asking if location should play a role," he says. "The problem was realizing that people had been hired, had relocated, had bought houses, had set their lives down in the community, had invested themselves, all based on different assumptions."

The question Merrill kept hearing from employees was "if this changed, what else could change later?" "That stayed with me," Merrill says, "because it was not only about compensation. It was about whether people could rely on the expectations we had created. The financial costs of a compensation change are relatively straightforward to calculate. The costs in morale are harder to calculate, and far more enduring."

There's a name in organizational research for what employees were really asking: psychological contract breach, the violation of a promise an employer made, spoken outright or simply understood. A 2025 study in the European Management Journal found the damage lands on trust before it lands on the wallet — because trust is what the employment relationship is actually built on, not pay. "The biggest lesson for me was that compensation policies are much easier to write than to undo," Merrill says.

Thomas Ricotta, a partner at Ricotta & Marks, a Queens, New York employment law firm, represented a client who relocated for a new job after the employer promised that location wouldn't affect their pay. Months later, the employer introduced new regional pay bands, and the employee's compensation dropped anyway. "The employee was caught completely off guard," he says. The first thing he usually hears from clients in this position is "I trusted what the employer told me."

Tech employees are willing to forgo 25% of their total compensation just for the option to work remotely, based on real job offers accepted and rejected by 1,396 workers, according to a 2025 study by Cullen, Pakzad-Hurson, and Perez-Truglia published in AEA Papers and Proceedings. Many workers facing a location-based pay cut already gave something up once to work remotely in the first place.

"It will feel like a demotion when you take someone's pay away," says Jessica Hart, founder of the compensation consultancy Hart Consulting. "You compensated them for that work, and now you want the same work for different compensation." Explain the why and the when before the change lands rather than after, she says. Skip that step, and the employee finds someone else to explain it to them, "and that's usually a lawyer that will take 40% of the settlement."

Ricotta's client's case never reached a courtroom. "This dispute was resolved through a negotiated settlement that included a phased compensation adjustment and additional transition pay," he says, "rather than litigation."

The exposure isn't the pay cut

"The largest legal exposure from a pay cut is typically not the cut itself, but rather the promise made prior to it," says Nick Heimlich, founder of Nick Heimlich Law, who has spent more than two decades advising companies on contracts, litigation, and legal risk. An employee could bring a breach of contract claim, promissory estoppel, a wage law violation, misrepresentation, or unfair business practice, he says, as long as three things are true: "the employer made clear that the pay was going to be 'location blind,'" the employee relied on that promise by relocating or staying put, and there was "an insufficient amount of advance notice" before the company reversed it.

Written notice before implementation, no retroactive reductions, and enough time for someone to actually act on the news is the safer path regardless of state, according to Heimlich, since how much notice counts as "enough" varies too much by jurisdiction to give one number. California, New York, and Colorado enforce it most strictly, according to Ricotta. A signed lease, a sold home, a declined counteroffer, a relocated family, according to Heimlich, all count as more than a coincidence when it comes to proving an employee relied on the company's promise.

"There's a discrimination legal exposure when a remote worker's pay has been cut just because they moved to a less expensive location without showing proof that the salary reduction aligns with local cost of labor," Ricotta says. "Companies must prove the pay reduction is justified based on different factors like education, and labor market surveys."

"A company should have available documentation such as original policies, communications with respect to relocation, written offer terms and conditions, wage notices, documented business rationale for the change, the effective date of the change, a complete list of workers who were impacted, and documentation supporting that the wage changes are prospective only," Heimlich says. Larger, more established employers tend to have this in place before a decision is announced, Ricotta says. Smaller ones tend to build it after. "A question in litigation is whether the employer can produce contemporaneous documentation showing the decision was driven by compensation strategy rather than a desire to cut costs," he says.

The same failure, repeated across many employees, is what turns an individual mistake into a class action. "Class-wide exposure typically occurs due to the repeated use of a faulty process amongst multiple employees," Heimlich says. "Common examples include incorrect notifications, inaccurate wage statements, unpaid earned wages and miscalculations related to final pay." One mistake is a payroll error. The same mistake made against fifty employees is a lawsuit with fifty plaintiffs.

Even when handled well, some of these still end up in dispute. "Cases begin with a demand letter and negotiations rather than an immediate lawsuit," Ricotta says. "If litigation becomes necessary, the process can take anywhere from several months to multiple years depending on jurisdiction and the claims involved. Many disputes settle before trial because they are expensive and unpredictable for both sides. A nationwide salary survey can support business rationale," he says, "but it is less persuasive than role-specific compensation benchmarks for comparable positions in the relevant labor market."

"In jurisdictions outside the US, compensation changes can trigger collective bargaining or employee-representation requirements that do not exist for most private-sector US employers," Ricotta says. "An employer may need to notify and consult with a works council or union before implementing the change, which can take months and can delay implementation." France, Germany, and Brazil require that kind of union involvement specifically for demotions or policy reversals, he says. The UK, the Netherlands, Spain, and Canada carry their own versions of the same complication, tied to contractual change rules and provincial employment protections.

"Multinational employers underestimate consultation requirements and local employee-protection laws," Ricotta says. "A policy that is legally straightforward in the US can signal legal risk overseas if local employee representatives are not engaged before implementation. The larger the gap between what employees were told and what is implemented, the greater the litigation and employee-relations risk tends to become."