The supplementary insurance that became Korea's second payer

Korea’s National Health Insurance covers everyone and does not cover everything. The gap it leaves — copayments, and a large category of treatments and materials designated as non-covered — is where private indemnity insurance took root. Sold as a supplement, the product reimburses most of what a patient pays out of pocket, and it has been bought at a scale that makes the word supplement misleading. Something close to forty million policies were in force by the mid-2020s in a country of fifty-one million people, a penetration rate that makes indemnity cover a near-universal second layer rather than a niche financial product.

A second payer changes what providers do, because it changes which services are price-sensitive. Anything the national scheme covers is reimbursed at an administered fee that hospitals widely regard as thin. Anything it does not cover is priced by the provider, and if the patient’s own insurance reimburses most of the bill, the patient’s resistance to that price is weak. The predictable consequence is that the non-covered category grew, and grew fastest in treatments that are easy to repeat, hard to adjudicate and delivered outside hospital admission. Manual therapy is the standard example: a course of physical manipulation, prescribed in multiples, priced freely, and for years reimbursed with little friction. Vitamin infusions, some proton and laser procedures, and premium intraocular lenses fitted during cataract surgery followed the same logic at different price points.

The aggregate is substantial. Estimates of annual non-covered medical spending in Korea have run into the high teens of trillions of won, and indemnity insurance funds a large share of it. That expenditure is not distributed the way clinical need is. It concentrates in ambulatory specialties where the procedures are elective, the volume is controllable and the reimbursement is generous relative to the covered-service fee schedule — which is also to say it draws practice capacity away from the services the national scheme sets prices for.

For insurers the arithmetic broke. Loss ratios on indemnity lines ran well above one hundred percent through the early 2020s, meaning claims and expenses exceeded premiums on the product as a whole. The response was premium increases, which arrive unevenly: renewal terms differ sharply across the four generations of the product, and holders of the oldest contracts, written before 2009 with the most generous terms and the least ability to reprice individual behaviour, have absorbed the largest increases. Those are disproportionately older policyholders who bought early and cannot switch without losing coverage they would not be sold again.

The fourth-generation product, introduced in July 2021, was the structural attempt at a fix. It separates non-covered treatment into a rider and adjusts each policyholder’s premium according to their own claims in the preceding period, discounting for those who claim nothing and surcharging heavily at the top of the distribution. The design targets the right thing, since claims are extremely concentrated — a small minority of policyholders account for a large majority of non-covered claims. Its weakness is that it only governs new and switched contracts, so it does nothing about the tens of millions of older policies where the incentives that created the problem remain fully in force. Further reform of the product line was under active discussion through 2025 for that reason.

The wider lesson is about sequencing rather than about insurance. A public scheme that leaves a defined gap will have that gap filled, and whatever fills it will set incentives the public scheme did not choose. Korea now has a national payer that decides what care is worth in one category of treatment and a private market that decides it in another, with hospitals allocating staff and rooms across the boundary accordingly. Managing non-covered pricing directly, rather than through the insurance product that pays for it, is the harder and more direct route, and it remains the one least attempted.