What every cost-of-living calculator gets wrong (and why the answer's usually off)

By Skyler Bissell · July 7, 2026 · 11 min read

Type two cities into a cost-of-living index and it hands you a clean verdict: "Oslo is 8% more expensive than Seattle." It feels like an answer. It's built from a real basket of prices, thousands of them, rent and groceries and a transit pass and a beer, and for a rough gut-check that basket is fine.

Here's the problem. That 8% is the answer to a question you didn't ask. You wanted to know what you'd keep after a move. The index told you what a shopping cart costs. Those are different numbers, and the gap between them is where every big relocation mistake lives.

Below are the five lines a price index can't see. Each one is enough to flip a real comparison, and the last section shows one flipping.

What a price index is actually good at

Start with credit where it's due, because this isn't a hit piece on Numbeo or Expatistan. Crowd-sourced indexes are genuinely useful for one job: getting a feel for daily prices in a place you've never lived. Wide coverage, granular line items, updated constantly by people on the ground. If your question is "roughly how pricey does a normal week feel in Lisbon versus Austin," a price index answers it well, and no other free tool covers as many cities.

The trouble starts when that percentage gets used to decide a move. A relocation is a take-home question wearing a cost-of-living costume. And the moment you cross a border, the two numbers, what things cost and what you keep, come apart, because taxes and the safety net change more than grocery prices do.

The five blind spots (each with a number)

Here they are, roughly in order of how hard they hit the answer. The figures come from cityparity's per-city engine, the same one behind every comparison page on the site.

1. Real progressive taxes

A price index applies no tax at all. It compares gross-to-gross prices and lets you assume the salary is the salary. But a $150,000 New York salary and its Berlin equivalent are taxed on completely different curves. Run the brackets and New York takes about 29% of it in income and payroll tax while Berlin takes about 38%. That is the whole reason a naive currency conversion misleads: $150,000 converts to roughly €138,000 today, but the salary that actually matches your New York take-home in Berlin is about €87,900. The €50,000 gap is tax and cost structure the index never touched.

2. Childcare net of subsidy

This is the biggest swing on the board for anyone with kids, and it's the line indexes handle worst. Most either skip childcare or quote an unsubsidized sticker price, which is the wrong number in every country that subsidizes care. A Seattle household with two kids in daycare pays about $30,000 a year, in after-tax dollars. Oslo caps the same care near NOK 66,000 through subsidy, roughly $6,700. That single line is a $23,000 swing toward Oslo, and it flows straight to net cash. A basket of grocery prices will never show it.

3. Healthcare you actually pay

In a price index, healthcare is a doctor's-visit price, if it appears at all. What you actually pay is a premium plus a deductible plus an out-of-pocket maximum, and it changes shape entirely across systems. For a single person, our engine puts New York around $4,000 a year against roughly €600 in Berlin. For that Seattle family of four the US exposure runs closer to $30,000 a year in premiums and out-of-pocket, a line that mostly vanishes under Oslo's universal system. The index charges you for neither version, so it can't tell you the difference.

4. The cash value of time

Statutory vacation and paid parental leave never appear in a price basket, because they aren't things you buy. They're pay you'd otherwise buy back. Oslo hands a worker 11 more guaranteed days off a year than Seattle and 37 more weeks of paid parental leave. Price a vacation day at your daily rate, salary over about 260 workdays, and at $150,000 those 11 days are worth roughly $6,300 every year. The leave, when you use it, is worth far more. None of it is on the offer letter, and none of it is in the index.

5. Expat tax regimes

For a few years, a special inbound-tax status can move the target number a lot, and no price index models it. The Netherlands 30% ruling is worth about €24,000 a year on a €150,000 base, for up to five years, and Spain's Beckham Law and Portugal's IFICI work similarly. They also expire, shrink, and sometimes do nothing for a normal local salary. We sorted which expat tax breaks are real and which are mirages, because "you'll qualify for the tax break" is doing a lot of unexamined work in most relocation pitches.

A worked flip: the family the index gets backwards

Put two of those blind spots on the same page and watch a verdict invert. Take a family of four, two working parents and two kids in daycare, weighing Seattle against Oslo. A price index reads this move as a small loss: Oslo's basket runs a little pricier, so the tool says stay.

Now add the two lines the index can't see.

Engine figures for a two-earner family of four, rounded. Childcare is net of Norway's subsidy; healthcare is US premiums plus out-of-pocket against Oslo's universal system.

Once those are on the table, the answer isn't close. Solve for equal net cash and a $280,000 Seattle household matches on about NOK 2,428,740 in Oslo. The family doesn't come out behind. They come out even on cash while pocketing a safety net Seattle would have charged them $50,000-plus a year to approximate: subsidized childcare, universal healthcare, 11 more paid days off, and 37 more weeks of parental leave. The "8% more expensive" headline had the sign backwards for this household, because the two biggest lines were the two it couldn't see. Run the whole thing on the Seattle vs Oslo family page.

None of this is a one-off. The bigger your family and the bigger your healthcare exposure, the more a price index misleads, always in the same direction: it flatters the low-tax, low-safety-net city because it's blind to the costs that city quietly leaves on you.

The honest method, and where it stops

The fix isn't a better basket. It's a different question. Instead of "what does a cart of goods cost over there," ask "what would I have to earn over there to keep the same net cash I keep now." That's one number, in money, and building it is mechanical: take your City A gross, subtract real income tax and the cost lines that move (housing, childcare net of subsidy, healthcare, transit, food, a trip home), and land on net cash. Then solve for the City B gross whose net cash matches, running B's own taxes and prices. The result is the equivalent salary, and it's the headline on every comparison here.

An honest tool also names its limits out loud, which a price index rarely does. A few of ours: for countries that tax spouses separately, we currently run household income through a single progressive ladder, which slightly overstates the bill for two-earner couples. Every figure is nominal, in the data's last-updated year, with no inflation adjustment. We don't price one-time moving costs, since those live in their own budget rather than your monthly run-rate. And a currency conversion still sits under the equivalence at a rate that drifts day to day. None of that makes the number a vibe. It makes it a strong estimate you should run with your own inputs, which is the point of putting the calculator in front of you instead of a percentage.

So which calculator should you use?

Both kinds, for different jobs. When you want a feel for whether daily life is pricier in a city you're curious about, a crowd-sourced price index is the right first look, and it's genuinely good at it. When you're holding an offer and the decision is real, you need the number that survives taxes and the safety net, which means a tool that computes take-home rather than a basket. The mistake isn't using a price index. It's letting a shopping-cart percentage answer a paycheck question.

If you're at the deciding stage, start here:

• Run your salary and family through the calculator → your real net-cash number instead of a percentage
• The equivalent salary you'd actually need: the method behind every figure here
• Seattle vs Oslo, for a family: the flip above, line by line
• New York vs Berlin: a high US gross against a deep safety net
• Expat tax breaks, decoded: the fifth blind spot, sized honestly

FAQ

What is the most accurate cost-of-living calculator?

It depends on the question. For a feel of daily prices, the crowd-sourced indexes like Numbeo and Expatistan are fine. For deciding a move, accuracy means computing take-home: each country's real progressive taxes, childcare net of subsidy, the healthcare you actually pay, and the cash value of leave. A price index computes none of those, so a tool that solves for net cash after them is more accurate for that job.

Why do cost-of-living calculators give different answers?

They measure different baskets of goods from different self-reported samples, and none of them compute what you keep after taxes. Small differences in which prices they sample produce small disagreements. The big disagreements with reality come from the lines none of them include: progressive income tax, childcare after subsidy, and out-of-pocket healthcare.

What do cost-of-living calculators leave out?

Five things, roughly in order of how much they move the answer: real progressive income and payroll tax on your actual salary, childcare net of government subsidy, the healthcare premium and out-of-pocket you actually pay, the cash value of statutory vacation and paid parental leave, and any expat tax regime you would qualify for. A basket of grocery and rent prices sees none of them.

Is a cost-of-living index enough to decide whether to move abroad?

No. A price index is useful for gauging whether daily life feels cheaper or pricier, but it can't see the lines that usually decide the move: taxes, childcare, healthcare, and time off. Those routinely swing the answer by tens of thousands, and for a family they can invert the verdict a price index gives.

Is Numbeo accurate?

For crowd-sourced prices, roughly, within the limits of self-reported data. Rent, groceries, transit, and a coffee are what it does well. It's incomplete for a relocation decision because it never computes income tax, childcare net of subsidy, healthcare out-of-pocket, or the value of paid leave. Use it for a price feel and pair it with a take-home calculation.

How much can childcare change a cost-of-living comparison?

For a family it's often the single biggest line. A Seattle household with two kids in daycare pays about $30,000 a year; Oslo caps the same care near NOK 66,000 through subsidy, roughly a $23,000 swing that flows straight to net cash. A price index either omits childcare or lists an unsubsidized sticker price, so it misses the line that most often flips a family's answer.

Does a higher salary always mean more money after a move?

No. The number that matters is net cash after taxes and the costs that change when you move. A higher gross in a city with high tax, expensive housing, unsubsidized childcare, and premium-plus-deductible healthcare can leave you with less than a lower gross somewhere the safety net covers those lines. Matching net cash is exactly what a price index can't do.

Stop comparing baskets. Run your own net-cash number →

Figures here come from cityparity's per-city engine and were current at publication; currency rates and tax rules move, so treat any single number as a strong estimate and run your own inputs. See the methodology.