Overtime fell 30% — less strain, or less pay?

A number that sounds like unambiguous progress: Korean regular-worker overtime fell from 11.5 hours a month in 2011 to 8.1 in 2025, a drop of about 30%. Less overtime reads as a win — more evenings, less overwork. And it may be. But an overtime-hours figure carries only half of the thing people actually feel, because overtime is both a demand on your time and a line on your paycheck.

Hours are not won

Overtime is usually paid at a premium — that is the whole point of it, for the worker. So when overtime hours fall, two things happen at once: the time burden eases, and the overtime pay shrinks. A worker who wanted the hours has lost income; a worker who resented them has gained an evening. The hours series measures only the first half of that trade — the time — and says nothing about how the second half, the pay, landed.

This is the same gap we flagged when a youth jobless rate fell for two possible reasons: the single number moves cleanly in one direction while the meaning underneath forks. A 30% fall in overtime hours is consistent with a healthier labour market and with a squeezed one, and the hours figure alone does not adjudicate between them.

Why this fits everything else we count

It is also why we keep insisting that a pay ranking hides the distribution behind it. Monthly pay is built from base wage, bonuses, and overtime, and each of those moves differently. Strip out overtime and the headline may barely move while the composition underneath shifts from variable pay toward fixed — a real change in how a paycheck is built, invisible in a single number. Hours, pay, and their mix are three different measurements, and a story that quotes one as if it settled the others is quietly overreaching.

What we report, and what we won’t

We report the overtime-hours series, because it is real and reproducible. What we will not do is translate a 30% fall in hours into a verdict about whether workers are better or worse off — because the pay side of that trade is not in the hours number, and we do not manufacture figures we have not measured. We would rather show you the hours, and name plainly that pay is a separate series we are not claiming to have moved. This is a statistic, not a settlement.

The debate

So: is falling overtime a story of lighter loads, or lighter pay? For some workers it is one, for some the other, and the hours figure will not tell you the mix. What do you think? This is a talking point, not a verdict — argue it out below.