Medigap loss ratios explained

The definition, in one line

A loss ratio is claims paid divided by premiums collected, over some period, for some group of policies. A block collecting $100 and paying $88 in claims has an 88% loss ratio.

“Loss” here is insurance jargon for a claim, not for losing money. A high loss ratio means a large share of premium is going back to policyholders as benefits. From your side of the table, high is not automatically bad — it means the policy is paying out.

The floor the law sets

Federal law does not leave this to the market. A Medicare supplement policy must be expected to return, as benefits, at least:

• 65% of premiums for individual policies, and
• 75% of premiums for group policies.

That is §1882(r)(1) of the Social Security Act.^{[Social Security Administration]} The implementing regulations sit at 42 CFR Part 403, Subpart B.^{[Electronic Code of Federal Regulations]} Most Medigap policies bought individually are on the 65% side.

So a block running at 70% is not violating anything. It is above the floor. The floor exists to stop a policy being sold that returns very little of what it takes in — it is a consumer protection, not a target.

The number worth watching is 100%

Here is where it gets useful to you. The legal floor is 65%, but the economically interesting line is 100%.

At 100%, a block is paying out in claims exactly what it collects in premium — before paying a single employee, commission, or overhead cost. Above 100%, it is paying out more than it takes in.

A block cannot stay there. It has essentially three ways out, and only one of them is available at scale: raise premiums. It cannot cancel policyholders — Medigap coverage is guaranteed renewable — and it cannot cut the benefits, which are standardised by plan letter.

That is why we publish loss ratios beside rate history rather than on their own. In our research corpus, the pattern recurs: blocks carrying state loss ratios at or above 100% turn up repeatedly alongside the largest filed increases. A loss ratio above 100% is not a prediction, and we will not dress it up as one. It is a pressure reading.

Read the state figure, not the national one

Rates are filed state by state, and a company’s experience varies enormously between them. The same carrier can be comfortable nationally and badly underwater in one state — different age mix, different medical costs, different history of who bought when.

When a national number and a state number diverge sharply, the state number is the one that bears on your premium, because your state’s regulator is the one reviewing the filing that affects you. Where we can publish both, we show both, and we label which is which.

What a loss ratio does not tell you

Four honest limits, because a number used carelessly is worse than no number:

• It is backward-looking. It describes an experience period that has already closed, often a year or more ago. It is a photograph, not a forecast.
• It is noisy in small blocks. A block of a few thousand lives can swing wildly on a handful of expensive claims. Treat a dramatic figure attached to a small block with caution.
• It says nothing about service. A block with a great loss ratio can still have poor claims handling. Different question, different source — your state insurance department publishes complaint data.
• It is not a company grade. Loss ratios belong to blocks in states, like rate increases do. Averaging them into a carrier-level score would be wrong in some state for nearly every company. Why blocks, not brands.

How to use it when you are actually shopping

Loss ratio is a second-pass filter, not a first-pass one. A sensible order:

1. Work out which plan letter you want, on benefits.
2. Find the blocks offering it to someone like you in your state.
3. Look at each block’s filed increase history — the shape over several years, not one figure.
4. Then use the loss ratio to sanity-check what you saw. A calm history with a loss ratio near or above 100% suggests the calm may not last. A history with one large increase and a loss ratio that has since come back down may mean the correction has already happened.

That last case is worth dwelling on, because it cuts against instinct: a block that has just taken a painful increase and returned to health can be a better place to be than one that has not taken its medicine yet.

Where our figures come from

Loss ratios on this site are published only when we can attribute them to a public regulatory source and cite it on the page. Until a figure clears that check, the page says so rather than showing an unsourced number. Our verification standard.