What Koreans are really arguing about when they argue about generations
Generational conflict is Korea's favourite explanatory frame, but the disputes it covers are about assets, pension arithmetic and pay systems, and the age variable is mostly standing in for those.
South Korea talks about generational conflict more than almost any comparable society, and it has built vocabulary to match. The label MZ, coined domestically around 2019 by a marketing research group, fuses millennials and Generation Z into one cohort of people born roughly between 1981 and 2010 — a thirty-year span no other country treats as a single group, and a category designed for advertisers rather than analysts. These are descriptions in search of a mechanism, and the mechanisms turn out to be mostly about property and institutional timing.
Housing is the material core. Seoul apartment prices rose steeply through the late 2010s into 2021 by every major index, and jeonse deposits — the lump sum a tenant lodges with a landlord in lieu of rent — rose with them. Whether a household owns, and when it bought, now explains more about its financial position than what it earns. Because purchase timing correlates strongly with age, a distributional question about asset ownership presents itself as a generational one. The framing obscures the variance that matters: among Koreans in their thirties, the gap between those whose families could assist with a deposit and those whose could not is wider than that between the average thirty-year-old and fifty-year-old.
Pensions supply the arithmetic that makes the conflict legible. The national scheme has charged a contribution rate of nine percent of income since 1998, targets an income replacement rate of forty percent, and is projected in the fifth official actuarial review, published in 2023, to exhaust its reserve fund around 2055. Those three numbers cannot coexist indefinitely. Current retirees receive substantially more than their contributions funded, a deliberate choice when the scheme was launched to cover a population with no other retirement income. Any reform therefore allocates a defined loss between cohorts and says so explicitly, which is what makes pension debate uniquely bad-tempered: the transfers are visible and the losers identifiable in advance.
The workplace version is a pay system, not a personality clash. Seniority-based pay scales remain common in larger Korean firms, meaning labour costs rise with tenure independently of role. The mandatory retirement age has been sixty since the middle of the last decade, and the case for raising it to sixty-five is strong on pension grounds — the gap between retirement and pension eligibility is where old-age poverty is produced. In a seniority-pay firm, extending retirement raises senior payroll and reduces hiring capacity, which is why younger employees at those firms oppose it. In a job-based pay firm, the trade largely disappears. Retirement-age reform and pay-system reform are therefore the same argument, and treating them separately guarantees deadlock.
Reported office friction dissolves under the same treatment. Younger employees leaving on time, declining company dinners and asking what a task is for are read as cohort traits, but they track institutional change: the statutory week was cut to 52 hours in 2018, harassment provisions entered the Labour Standards Act in 2019, and remote work normalised during the pandemic. Older employees under the same rules behave similarly, which is the test the generational reading fails.
Survey evidence also undercuts the frame from inside. On most attitudinal measures — economic security, gender, work — the variance within a Korean age cohort exceeds the variance between cohorts, and the divergence between young men and young women — the split behind the gendered cohort labels that entered media shorthand around 2021 — is on some questions wider than that between young and old. A category whose members disagree with each other more than with outsiders is not doing explanatory work.
None of this means age is irrelevant. Cohorts met different labour markets, housing prices and pension rules, and those differences compound over a lifetime. But “generation” is a cheap variable: universally available, intuitively satisfying, and usually a proxy for asset position and the timing of institutional change. The questions worth arguing about are about assets and rules — harder, and considerably less entertaining.