Things change. Periodic re-validations and updates of geographic pay differentials are essential. Huge differences can rapidly disappear or suddenly emerge.
Measure competitive pay, not personal spending. Cost of labor is not cost of living, or vice versa. Local pay variances are not the same as local living cost differences. What any individual earns from a single job is not the same as what your entire family spends. Labor costs surveys do not match household living expense reports. Competitive pay is a subject entirely different from family spending patterns, neither measured the same nor matching each other. Pay is an set by agreement between worker and boss while living expenses reflect unique independent family lifestyle choices beyond the control of any employer.
Focus appropriately. Only pay practices that affect your ability to attract, retain and engage competent workers should be relevant for geographic analysis. You wouldn't measure pay in France to set pay in Kentucky, would you? (Even though both feature cities named Paris.) Executive compensation patterns that are based on international or even global comparisons have virtually no effect on rank and file pay relationships. Likewise, unique State minimum wage practices in one State or Province usually have no influence on payrolls where almost everyone earns six-figure incomes. Letting bottom-level mandatory wage thresholds govern the budgets for your entire payroll can be a terrible (and expensive) mistake. Likewise, if you don't recruit from a location or have some internal equity relationship based on it, then exclude it from your RELATIVE comparison exercise.
Avoid binary ratios. Reality is not simple. Multiple polynomial formulae are much more accurate than single percentages. Patterns found at clerical levels are rarely correct for managerial jobs.
Any differential should reflect actual human talent cost variances between the relevant jobs in the respective appropriate labor market locations. No GeoDiff should ever be a fixed premium infinitely tied to a location without regard for actual competitive pay equity. If a law wipes out a prior competitive pay variance reflected in your geographic adjustment formula, you change it rather than ignore the leveling of a previously tilted playing field (or vice versa). Pay differentials are not absolute eternal entitlements but are short-term policy practices contingent on your unique compensation situation and your particular total reward strategy. Actually, they are lazy shortcuts to justify treating all employees in a particular class the same regardless of precise comparisons of current individual rates for specific job titles or competency clusters. Beware of assumptions that they are permanent rights.
When relative relationships change, methods and practices must follow and be adjusted accordingly rather than automatically repeated perpetually without any rational reason beyond just "staying consistent." If government imposed some new rule that increased your production costs or service delivery expenses for Product A by 20%, would you automatically spend 20% more on producing/delivering Products B through Z? Logic and purpose should determine methods and practices, especially those dealing with money invested in talent.
Be proactive. Revise or rework your geographic differentials to reflect competitive/internal reality. Don't force the entire organization to unnecessarily expend scarce payroll dollars like shackled minions serving a rigid HR/comp practice designed for expired circumstances. Sorry ... abandoning a prized familiar protocol is not easy, but change might be necessary for the future. Updates also allow you to demonstrate that HR/comp is a proactive business partner who adapts with change rather than thoughtlessly dictating outdated and currently inappropriate habits. Remember that resistance to a needed pay practice change usually means you are wastefully misdirecting funds into jobs where competent candidates are readily available for far less cost. That doesn't mean that continuing to pay a premium might not be important for your enterprise, but at least do it on purpose with open eyes rather than accidentally, based on misunderstandings.
No policy should be written in stone unless it is a basic fundamental strategic principle. Every prescription should follow a relevant diagnosis. But you already knew all this, didn't you? So now you have an outside expert's advice to back up your usual recommendations.
E. James (Jim) Brennan is an independent compensation advisor with extensive total rewards experience who conducted the initial research behind the first published study of this subject.
Population Cartogram image by Worldmapper.org, courtesy of Creative Commons
Metropolitan Statistical Areas by U.S. Census Bureau, via Wikimedia Commons