Ask anyone in the prosperous suburbs of a booming capital how the economy is doing and you will hear that wages are up, jobs are plentiful and life is, broadly, fine. Travel two hundred kilometres to a mid-sized industrial town and the same question earns a flat laugh. Both answers are true, and that is the point. The gap between them is the subject of a months-long data investigation by Blog Dergisi — and it exposes a divergence that national averages are built, almost perfectly, to conceal. It is not a story about one country. It is the shape of the rich world's recovery everywhere.
The headline figures reassure. Average nominal wages rose roughly 4.3 per cent over the past year, comfortably ahead of a national inflation rate that has cooled toward target. On paper, real incomes are recovering. But "average" and "national" are the two words doing the most damage here, and it is worth seeing exactly how. When we rebuilt the picture region by region — pairing local pay with a basket of local costs, from rent to childcare to the weekly shop — the comforting average dissolved into a patchwork of winners and quiet losers. The mechanism is banal and merciless: a single mean can rise while most of the distribution beneath it falls.
Where the numbers turn negative
The pattern is not the one the headlines imply. The places slipping behind are rarely the poorest. They are the middle — regions built around a single industry, commuter belts an hour too far from a thriving metropolis, towns where pay rose three per cent while rents rose eight. In a cluster of formerly industrial districts we examined, real purchasing power has fallen for the third year running, even as the same workers collected nominal raises they were told to be grateful for. The raise was real. So was the erosion beneath it. Both showed up on the same payslip.
Housing is the lever that does most of the work. In the dynamic city-regions, soaring rents claw back much of the wage gain, but high salaries and dense labour markets cushion the fall. In the struggling middle the trap is meaner: pay growth is sluggish because the local economy is, yet housing and energy costs track national trends the local wage cannot. The squeeze that follows is invisible to official statistics anchored to a national consumer-price index. A single figure for "the cost of living" becomes a fiction the moment you cross a regional line — and the further from the capital you travel, the bigger the lie.
"The national average is a flattering mirror. It shows the median voter a face two-fifths of the country simply does not recognise as their own — and then wonders why they stop believing the statistics."
A regional-economics researcher — who reviewed our methodologyHow we counted
A finding this uncomfortable demands that we show our working, so here it is. We combined regional pay records with a locally weighted cost basket, adjusting each region's spending mix rather than imposing the national one — a town where half of every wage goes on rent does not inhabit the same economy as a city where a quarter does. Where data was thin we took the more conservative estimate, which means the divergence we found is, if anything, understated. The full dataset and every weighting choice are published alongside this piece, for anyone who wants to argue with them. Evidence over outrage: we expect the argument, and we welcome it.
The politics of all this are only beginning to surface, and they are combustible — the same charge across every advanced democracy, whatever the local flag on the ballot. A worker told the country is recovering while their own budget tightens for a third straight year does not conclude they are an outlier on a spreadsheet. They conclude the people reading the spreadsheet are lying to them. That sentiment — the feeling of being averaged out of the national story — is already redrawing electoral maps, and it explains why a dataset that looks benign in a finance ministry reads as incendiary in a town hall. It runs, too, into the renegotiation of where and how people work, as the regions losing ground watch the remote-friendly jobs cluster somewhere else entirely.
What our map shows, in the end, is that "purchasing power" is not one number but hundreds, and that the distance between the best and the worst is widening faster than any national statistic will admit. The recovery is real. It is also so unevenly shared that, for a substantial minority, it is no recovery at all. The question for the coming years is not whether incomes rise. It is whether the places quietly left off the map decide they have had enough of being told, from the capital, that they are doing fine.