Korea's 43% Pension Rate Isn't Everyone's Rate

The Number 43% Assumes One Particular Person

Starting in January 2026, Korea’s National Pension income replacement rate rose from 41.5% to 43%. When you see this number in a notice or a news article, it’s natural to read it this way: does that mean I’ll get 43% of my salary back as a pension when I retire?

There’s a reason it reads that way. Even the Ministry of Health and Welfare’s own press release describes the replacement rate as the percentage of an individual’s average lifetime income that gets paid out as a pension. Read that sentence on its own, and it sounds like the same rate applies to everyone.

But open up the National Pension Act itself, and you won’t find the number 43% anywhere. Article 51, which sets the basic pension amount, contains only a constant—1.29—and a formula. 43% is simply the output you get when you plug one specific person into that formula. Figure out who that person is, and where you differ from them, and you can finally connect the number in the official explanation to your own pension.

The Law Says 1.29, Not 43%

Article 51 of Korea’s National Pension Act defines the basic pension amount like this: add the two amounts below, multiply by 1.29, and if your enrollment period exceeds 20 years, add 5% of that amount for every additional year.

• A: the average of the average monthly income of all subscribers over the three years immediately before you start receiving your pension. For 2026, the A value is ₩3,193,511 (~$2,300).
• B: the average of your own standard monthly income¹ over your enrollment period, after converting each year’s figure to the value of the year just before you start receiving benefits, using the revaluation rate².

The Ministry of Health and Welfare’s policy guidance writes this formula as 1.29(A+B)(1+0.05n/12), describing 1.29 as a constant that reflects the 2026 income replacement rate. Here, n is the number of months enrolled beyond 20 years.

Let’s plug a person into this formula. Suppose my average income (B) happens to equal the overall average (A), and I enrolled for exactly 40 years, all after 2026. The 20-year baseline amount is 1.29 × 2A, and since I’ve put in 20 more years beyond that, the 5%-per-year additions apply 20 times — doubling the amount. Dividing the annual figure, 1.29 × 2A × 2, by 12 months gives 0.43A per month. That’s 43% of my income.

43% is the ratio for an average earner who enrolled continuously for 40 years starting in 2026, without a single gap. This is exactly why this figure is called the nominal income replacement rate³. It isn’t the rate anyone actually receives — it’s a benchmark figure meant to represent the system’s benefit level through one idealized case.

Even under the same 43%, the ratio differs person to person

Change just one of the two conditions in our representative example—average income or 40 years—and the ratio shifts.

Start with the enrollment period. Article 63 specifies that if you’ve been enrolled for 20 years or more, you get the full basic pension amount; for 10 to under 20 years, you get 50% at the 10-year mark, plus an additional 5% for each extra year. If you haven’t reached 10 years by age 60, you receive a lump-sum refund⁴ instead of an old-age pension. Working through the formula, for an average earner, each year of enrollment adds roughly 1.075 percentage points.

Next is income level. Since both A and B appear in the formula together, if your income is below average, the overall average (A) pulls your share up; if it’s above average, it pulls your share down. So even with the same 40 years of enrollment, your ratio relative to your own income varies by income bracket.

The table below applies the 1.29 multiplier only to enrollment from 2026 onward and fixes the A value at the 2026 level. It excludes the dependent family pension amount, and keep in mind that actual pensions are recalculated using the A value and revaluation rate at the time you start receiving them.

Scenario	Enrollment period	Ratio to own income	Monthly pension (approx., based on 2026 A value)
Average earner (B = A)	10 years	10.75%	~₩340,000 (~$245)
Average earner (B = A)	20 years	21.5%	~₩690,000 (~$497)
Average earner (B = A)	30 years	32.25%	~₩1,030,000 (~$742)
Average earner (B = A)	40 years	43%	~₩1,370,000 (~$987)
Half the average income (B = 0.5A, ~₩1,600,000/~$1,153)	40 years	64.5%	~₩1,030,000 (~$742)
Double the average income (B = 2A, ~₩6,390,000/~$4,604)	40 years	32.25%	~₩2,060,000 (~$1,484)

Someone earning twice the average income doesn’t get double the pension—they get 1.5 times. Converting to a ratio against one’s own income, a person earning half the average comes out above 43%, while someone earning double comes out below it. 43% is only exactly right for someone whose B equals A.

Periods accumulated before 2026 are calculated using the numbers from that time

So far, I’ve assumed someone who only enrolled after 2026. But most current enrollees already have accumulated periods. The 43% doesn’t apply to those periods.

The supplementary provisions of the National Pension Act, revised in April 2025, state that the basic pension amount for enrollment periods before the law took effect follows the previous regulations. The National Pension Service also explains that the adjusted income replacement rate only applies to enrollment periods from January 1, 2026 onward, and does not apply to those already receiving pensions. In the example the Service gives, someone who is 50 in 2026 only gets the 43% applied to the remaining 10 years until age 59.

The previous constants differ by period. Enrollment from 1999–2007 used 1.8 (60% replacement rate), 2008 used 1.5 (50%), and after that it dropped by 0.015 each year, reaching 1.245 (41.5%) in 2025. The Service’s expected-pension formula also works by multiplying each year’s enrollment months by that year’s constant and summing them up.

So even with the same total enrollment period, the result differs depending on when it was accumulated. Calculating for an average-income earner looks like this:

• Someone enrolled for 30 years, all after 2026: 32.25%
• Someone enrolled for 20 years between 2006–2025, plus another 10 years from 2026–2035: about 34.3%

The reason the second person’s rate is higher is that the constants for the 2006–2007 period (1.8) and several years after 2008 (around 1.5) were larger than 1.29. Looked at another way, for this person, this reform only changed the remaining 10 years’ worth. The longer the history already accumulated, the smaller the portion affected by this increase — and the longer the period still to be accumulated, the larger it is.

When reading the figures the government has put out, you also need to look at what baseline they’re being compared against. A government policy briefing explained that this increase would raise the monthly pension amount by about ₩90,000 (~$65) for someone with 40 years of enrollment. If you plug in an average-income earner using the 2026 A-value, the difference between 43% and 41.5% comes to about ₩48,000 (~$35) a month, while the roughly ₩96,000 (~$69) figure is the difference between 43% and 40%. 40% is the rate that, under the pre-reform plan, would have applied from 2028 onward. The article doesn’t show the assumptions behind the calculation, so I couldn’t confirm which baseline the government used. Still, the same increase can look nearly twice as large or small depending on what you compare it to.

The numbers on the estimated pension screen also rest on assumptions

To see your own number directly, you can use the National Pension Service’s estimated pension inquiry. Once you verify your identity, it calculates based on your actual enrollment history. But even this number comes with its own assumptions attached.

The Service’s simple calculator, which you can use without logging in, states that it assumes enrollment started in January 2026 and computes using the 2026 A-value. It also notes that the actual pension is calculated using the A-value and re-evaluation rate in effect when you actually start receiving it, so the real figure may differ. Even on the identity-verified screen, it’s worth checking the notice explaining what assumptions were made about your future income and enrollment period.

Real-world enrollment periods are far from 40 years. According to reporting that cites National Pension Service data submitted to the National Assembly, the average enrollment period for new old-age pension recipients from January to March 2024 was 237 months — just under 20 years. The same reporting notes that, per the Service’s own projections, the average enrollment period for people starting to receive pensions from 2030 onward is also 20.3 years. If an average earner enrolled for 20 years entirely after 2026, that comes out to 21.5% in the table above. Mix in enrollment from before 2026, and the figure climbs a bit higher.

Here’s how you can lay out the sequence connecting the replacement rate to your own pension. Start with 43% as your baseline, then plug in, in order: how far your enrollment period falls short of 40 years, how your income differs from the average, and how that period splits across the 2026 divide. The final amount is something the Service’s inquiry screen calculates for you — but these three factors are how you can read the assumptions underneath that number.

Oswarld’s Lens

I read the income replacement rate as a number with “average” baked in twice. One average is my own lifetime income; the other is the average income of all plan participants. The 43% figure only holds when these two averages belong to the same person who has paid in for a full 40 years.

That doesn’t mean the number is wrong. To compare a system’s benefit levels across different periods or different countries, you need some benchmark case, and a representative example serves that purpose. The trouble starts when the output of that representative example gets communicated as if it were a promise made to an individual. Whether someone sees 43% and expects to receive more, or sees it and feels disappointed they’re getting less, both have skipped past the assumptions and taken only the number — in that sense, I think they’re making the same mistake.

The habit of asking about the comparison baseline and assumptions before looking at the output figure is especially useful when it comes to pensions. The same impression of impact can translate to ₩48,000 (~$35) or ₩96,000 (~$70) depending on the benchmark you use, and the same 30 years of contributions can shift the replacement rate by more than 2 percentage points depending on when those years were accumulated.

Closing

The 43% income replacement rate is the figure for an average earner who has paid in for 40 years starting from 2026. To connect that number to your own pension, you need to plug in three things that differ between that hypothetical person and you: your enrollment period, your income level, and how your contribution years split before and after 2026. The number that comes out of that calculation is the one sitting on the National Pension Service’s lookup screen — and that number, too, rests on assumptions about the future.

---

💬 When you first saw the number 43%, did you read it as “43% of my income”? Or did you take it to mean something else? Let me know in the comments.

📨 If you have a colleague or family member who once understood 43% differently than you did, send them this piece.

---

References & Further Reading

Primary sources

• National Law Information Center, National Pension Act Article 51 (Basic Pension Amount) · Article 61 (Old-Age Pension Beneficiaries) · Article 63 (Old-Age Pension Amount) · Article 77 (Lump-Sum Refund), Addendum (Act No. 20903, 2025.4.2) Article 5, Addendum (Act No. 8541, 2007.7.23) Articles 20 and 34. Current as of 2026.11.1: this is the original text confirming the 1.29 constant, the year-by-year prior constants, and the transitional measures for coverage periods before 2026.
• Ministry of Health and Welfare, National Pension Benefits policy guide: explains the basic pension amount formula, the definitions of A, B, and n, and notes that 1.29 is the income replacement rate constant for 2026.
• National Pension Service, Pension Reform Guide: source for the explanation that the adjusted income replacement rate applies only to coverage periods after 2026.1.1, and for the example of a 50-year-old subscriber.
• National Pension Service, Simple Pension Estimate Calculator: lists the 2026 A-value (₩3,193,511 (~$2,303)), the upper and lower bounds of the standard monthly income, and the assumptions behind the simple calculation.
• Ministry of Health and Welfare, “National Pension: Stronger Finances for the New Year, Greater Security for Retirement” press release: contains the definition of the income replacement rate and the explanation that it does not apply to existing beneficiaries.
• Korea Policy Briefing, “State Guarantee of Pension Payments” Written Into Law — What Changes for the National Pension Starting Next Year, 2025.12.29: source for the explanation of a roughly ₩90,000 (~$65) monthly increase based on 40 years of coverage.

Background

• The Hankook Ilbo, How Much National Pension Will I Actually Get? Why a 43% Income Replacement Rate Is Unrealistic, 2025.4.1: citing data submitted by the National Pension Service to the National Assembly, reports the average coverage period of new beneficiaries (237 months) and the projected coverage period for those starting to receive benefits in 2030 (20.3 years).

Past issues worth reading alongside this one

• The Country Where Children’s Paychecks Cover Their Parents’ Pensions: looked at the National Pension system through the lens of public finances and generational structure. This issue looks at the same system from the angle of how an individual’s benefit is actually calculated.

---

The author is Oswarld (Kwangseob Ahn). Current roles: Adjunct Professor at Sejong University, Strategy Consultant at INLEVEL9. Career, research, books, and recent work are kept current on the About page. Latest · July 2026: HEMA-2: A Consolidation-Aware Tri-Memory Architecture with Multi-Channel Scheduling for Lifelong Conversational AI.

📝 Glossary

Footnotes

1. Standard monthly income: the monthly income figure used as the basis for calculating both pension contributions and pension benefits. From July 2026 to June 2027, it is set within a range of a ₩410,000 (~$296) floor and a ₩6,590,000 (~$4,756) ceiling. ↩

2. Revaluation rate: the rate used to convert a past year’s standard monthly income into its value as of the year just before the pension is received. The Minister of Health and Welfare announces this rate annually. ↩

3. Nominal income replacement rate: the ratio produced by plugging an average earner with 40 years of coverage into the statutory formula. The ratio calculated using an actual beneficiary’s real coverage period and income is commonly called the effective income replacement rate. ↩

4. Lump-sum refund: a one-time payment—covering the pension contributions paid in (including, for workplace-based subscribers, the employer’s contributions) plus interest—given to someone with less than 10 years of coverage who, for example, turns 60. ↩