============================================================================ WOMEN RECEIVE A SMALLER PENSION. MEN RECEIVE IT FOR FEWER YEARS. ONE OF THOSE HAS A NAME, A DATASET CODE AND A PLACE IN EU SOCIAL REPORTING. ============================================================================ Two things are true about sex and retirement in Europe and both come from the same statistical office. Only one of them is an indicator. The third thing in this piece is the one that cuts against us, and we checked it before writing the rest. By Mikal Monslaup with Claude (Anthropic) Kairunen on X Rickmyrolls on Reddit The reporting, the reading and the conclusions are mine. Claude did the searching, the counting, the checking and much of the drafting, and argued back when a claim was thinner than it sounded. Every figure here has a primary source you can open, because neither of us is asking to be believed. First published: 19 August 2026 (2026-08-19T11:45:00+02:00) Canonical: https://unformentoo.org/two-gaps-in-retirement.html IN SHORT ---------------------------------------------------------------------------- * Eurostat publishes the gender pension gap as a named indicator with its own dataset code. Across the EU in 2023, women aged 65 to 74 received pensions 24.0 per cent lower than men. * That figure is real, large and rightly measured. Nothing here argues for reducing it or for taking anything from anyone. * Eurostat also publishes life expectancy at 65 by sex. At the same retirement age a man draws his pension for about 3.4 fewer years than a woman. * There is no indicator for that. It is one subtraction from data already collected and published, and nothing is indexed against it. * Before writing this we checked the figure that could undo it, and it undid half of it. * Healthy life years at 65 are 9.2 for men and 9.6 for women. Women's extra 3.5 years are almost entirely years in poor health, not extra healthy retirement. * So the claim that men get fewer good years of retirement is close to false, and this piece does not make it. The claim that men receive fewer years of payment survives intact. * Denmark closes the argument. Its gender pension gap is 0.5 per cent, effectively solved, and Danish men still draw their pension for 2.8 fewer years. * The two gaps are independent. Closing one does nothing to the other, and only the one with a name is being managed. ---------------------------------------------------------------------------- 01. THE GAP THAT HAS A NAME It is measured, it is published annually, and it is large. ---------------------------------------------------------------------------- Eurostat maintains a dataset called Average gender pension gap by age group. It has a code, ilc_pnp13, an age breakdown, a country breakdown and a place in European social reporting. Anyone can open it. For the EU27 in 2023, among people aged 65 to 74, women's average pension was 24.0 per cent lower than men's. That is a serious number and this piece does not dispute a decimal of it. Nothing here asks for a smaller pension for anyone. The question is only what else sits beside it, measured with the same care, by the same office, from data it already holds. ---------------------------------------------------------------------------- 02. THE GAP THAT DOES NOT It takes one subtraction, from two numbers Eurostat already publishes. ---------------------------------------------------------------------------- A pension is an amount per year, and it is also a number of years. The first is measured. The second is not. Eurostat publishes life expectancy at 65 by sex. For the EU27 in 2023 it was 21.8 years for women and 18.3 for men. At the same retirement age, that is about three and a half fewer years of pension for a man. There is no indicator for this. No dataset code, no target, no age breakdown built around it, nothing indexed against it in European social reporting. It is a subtraction anybody could do and nobody has been asked to do. Statistics Norway is the same. Its table catalogue returns pensioners by sex, and nothing when you search it for expected pension years or pension duration. Say precisely what is and is not being claimed. Eurostat does not refuse to publish this. It publishes both halves of it. What does not exist is anyone reading the pension question against the mortality one. ---------------------------------------------------------------------------- 03. THE NUMBER THAT TOOK HALF OF THIS PIECE AWAY Before writing any of the above we checked the figure most likely to undo it. It undid a good deal of it. ---------------------------------------------------------------------------- The tempting next sentence, and the one we were going to write, is that men get fewer good years of retirement. Eurostat publishes healthy life years at 65 as well, so it can be checked rather than assumed. Healthy life years at 65 are 9.2 for men and 9.6 for women. A gap of four tenths of a year. Women's extra three and a half years are almost entirely years in poor health, not extra healthy retirement. So the welfare version of this argument is close to false and this piece does not make it. Men do not get meaningfully fewer healthy years of retirement. They get almost exactly the same number. What survives is narrower and still true. A man draws his pension for about three and a half fewer years, whatever the health status of those years, because a payment does not care how you feel while receiving it. And the check turned up something better than what it removed. The familiar line that women outlive men is, after 65, largely a statement about years in poor health. That is a fact about both sexes, it is nobody's talking point, and neither the pension gap nor the life expectancy gap will tell you on its own. It also sharpens the point rather than blunting it. Eurostat has published healthy life years for two decades. The data was never the problem. ---------------------------------------------------------------------------- 04. DENMARK, WHERE ONE GAP IS SOLVED AND THE OTHER IS NOT If the two gaps were the same thing, closing one would move the other. Denmark shows they are not. ---------------------------------------------------------------------------- Denmark's gender pension gap in 2023 was 0.5 per cent. For practical purposes it is closed, and it is the only country in this comparison where that is true. It took decades of deliberate policy and it worked. Danish men still draw their pension for 2.8 fewer years than Danish women. Solving one gap did nothing to the other. That is the clearest available demonstration that they are different quantities, and that only one of them is being managed. Not because anyone decided the other did not matter. Because one of them is an indicator with a name, and the other is an arithmetic operation nobody performs. ---------------------------------------------------------------------------- This piece asks for one table. Where a statistics office already publishes pension income by sex, and already publishes life expectancy by sex, publish the years of receipt alongside them. It would cost a subtraction. The reason it has not happened is not cost, and it is not data. It is that nobody has been asked, because there is no column for the question. ---------------------------------------------------------------------------- THE DOCUMENTS THIS RESTS ON ---------------------------------------------------------------------------- ilc_pnp13 Average gender pension gap by age group https://ec.europa.eu/eurostat/databrowser/view/ilc_pnp13/default/table hlth_hlye Healthy life years by sex, including life expectancy at 65 https://ec.europa.eu/eurostat/databrowser/view/hlth_hlye/default/table ---------------------------------------------------------------------------- SOURCE NOTES ---------------------------------------------------------------------------- * Every figure was pulled from Eurostat's dissemination API on 19 August 2026 rather than read off a summary page, and the working file is in the repository at research/pensions-old-age.json. One inconsistency is worth stating: Eurostat's healthy life years dataset gives life expectancy at 65 as 18.3 and 21.8 where its demography dataset gives 18.2 and 21.6. Different vintages of the same measure. The health split in section three uses the healthy life years dataset throughout, and the 3.4 year figure in section four uses the demography one. ---------------------------------------------------------------------------- SOFTENED OR OMITTED FOR ACCURACY ---------------------------------------------------------------------------- * {"wanted":"Men get three and a half fewer healthy years of retirement than women.","published":"Healthy life years at 65 are 9.2 for men and 9.6 for women, a gap of 0.4 years.","why":"We checked before publishing and the life expectancy gap is almost entirely years in poor health. Presenting it as a healthy-years gap would have been the easiest possible way to have this piece dismantled."} * {"wanted":"Pension systems cheat men.","published":"Men draw a pension for fewer years because they die earlier.","why":"The harm is the earlier death, not the pension arithmetic. Framing it as a benefits grievance would miss the finding and invite a rebuttal that would be correct."} * {"wanted":"The gender pension gap is overstated or unimportant.","published":"It is 24 per cent across the EU and 39.9 per cent in the Netherlands.","why":"It is real and large. This campaign is additive and that rule does not lapse because a convenient comparison turned up."} * {"wanted":"Eurostat refuses to publish how long each sex draws a pension.","published":"Eurostat publishes both components. No indicator combines them.","why":"There is no evidence of refusal and we did not look for a motive. The data is there and the question is not asked."} ============================================================================ This document is generated from the same source data as https://unformentoo.org/two-gaps-in-retirement.html. If the two ever disagree, the page is the one that was rebuilt last; both are regenerated together. ============================================================================