The expensive-city penalty does not show up in the data
A familiar argument comes up early in conversations about financial independence. Housing is the biggest line in most budgets. Move somewhere cheaper, keep the same job if you can, and invest the difference.
It sounds hard to argue with. I went looking for that effect in hundreds of published millionaire profiles, each of which describes its cost of living as high, medium or low. I expected the cheap-area group to reach a million first.
They arrive last.
The headline comparison
People describing a high cost of living reached their first million at a median age of 37.5. Everyone else reached it at 40. That is two and a half years earlier for the group that is supposed to be at a disadvantage. It is not a rounding artefact either: a rank-based test puts it at p = 0.002.
The same pattern shows up in net worth. Median net worth is 3.00 million in the high-cost group against 2.40 million in the rest, at p = 0.003.
| Cost of living | Median net worth | Median age at first million | Median income | Median savings rate |
|---|---|---|---|---|
| High | 3.00M | 37.5 | 227k | 36% |
| Medium | 2.40M | 40 | 185k | 35% |
| Low | 2.35M | 41 | 150k | 27% |
Income and savings rate are where the difference shows up.
It is mostly income, and the savings rate barely moves
The high-cost group earns about 30 percent more: a median around 227,000 against about 175,000 for everyone else, at p = 0.011.
Their savings rate is 36 percent against 34. That is a difference of Two points, and a rank test gives p = 0.19, nowhere near significant. I would treat those as the same number.
So the expensive place appears in the income, not the savings rate. Housing may cost more, but pay is higher by more. A similar share of a bigger income reaches the portfolio each year.
One row in the table is stranger than the main result. The low-cost group saves 27 percent, while both other groups are in the mid thirties. If cheap living frees money for saving, this is where that effect should appear. Instead, the row points the other way.
I can think of two innocent explanations, but the data cannot separate them. Lower incomes may leave less room above fixed costs no matter what housing costs, so the percentage falls even when the intention is identical. Or this may be another problem with self-description: people calling their area low cost may be answering a different question from people calling theirs high cost.
Whatever the reason, I do not see cheaper rent turning into a higher savings rate in these profiles.
The part I expected to kill it, and did not
This could just be an income result, not a geography result. High earners cluster in expensive cities. Comparing the cities may really mean comparing the salaries.
So I split everyone into three equal income bands and repeated the comparison inside each one, where pay is roughly matched:
Negative would mean the high-cost group was slower. Every band is positive, so at similar income the high-cost group still reached the first million sooner. The bands are small once split three ways, so read the direction rather than the size.
At similar incomes, the high-cost group is still faster in all three bands. The middle band has the largest gap at five years. I do not have an explanation for that, and I am not going to make one up.
I expected the advantage to disappear once income was held roughly constant. It did not.
What this does not say
This does not show that moving somewhere expensive makes a person richer. Selection could easily produce this result on its own.
Everyone in this data reached a million. Someone who moved to an expensive city, found that the maths did not work and left would be missing from the sample. So would someone who stayed and never reached a million. The data contains the survivors. Expensive places may impose a harsher filter, making those survivors look better even if the place itself did nothing.
The data has other limits. Cost of living is self-described rather than indexed, so one person's high is another person's medium. Income is a snapshot near the time of writing, not an average across the years of accumulation. And the income bands become small once divided three ways, so I trust the direction more than the size of the gaps.
What I take from it
I came away less convinced that cheaper places help automatically. The cheap-area advantage is not visible here. Neither is the higher savings rate that is supposed to create it.
The more useful question is whether income will hold while costs fall. If that is true, moving somewhere cheaper has a strong case. But this data includes many people whose income was attached to the expensive place and who did better by staying.
The savings-rate column suggests something else too. Two groups with a fifty-thousand-dollar difference in median income saved almost exactly the same fraction of it. Whatever sets a person's savings rate, housing costs do not appear to explain it here.
I would put more weight on a study that tracked the same people before and after a move. A relocation study would answer a different question. That would be a better test of whether relocation changes the outcome.