What plan design is worth

Plan design transfers cost. Purchasing reduces it. Three tools to run against your own plan, with nothing stored.

Drafting compass, protractor and ruler

This page collects the interactive tools from Just a tree in one place, and adds one that the article does not carry. Everything here runs in your browser against numbers you type in. Nothing is uploaded, nothing is saved, and nothing asks for your email.

Two different jobs

Plan design is a risk transfer tool. Moving a deductible or a copay does not change what a knee replacement costs. It changes how much of the cost, and how much of the risk, shifts from the plan to the employee.

Reducing what the plan spends is procurement. What you buy, where you buy it, and what you agree to pay, the same way the business buys everything else.

The first two tools show where the transfer runs out of room. The third prices it.

A suggestion before you start: these work best with your actual plan in front of you. A summary of benefits, last renewal's rate page, and ten minutes will get you further than an afternoon of reading commentary, mine included.

1. Where your cost sharing stops

Enter your deductible, your coinsurance, and your out of pocket maximum, or load a national benchmark with one of the preset buttons. The tool returns the claim size at which member cost sharing is fully exhausted, the point where every additional dollar belongs to the plan. On the average large-firm plan that point arrives near $23,000. The number to sit with is what share of your spend lands above yours.

A chart showing the share of a claim borne by the member as claim size grows. Cost sharing applies fully at small claims, then collapses toward zero once the out-of-pocket maximum is reached, which happens well below the claim sizes that drive most plan spending.

Where plan design stops working

Cost sharing is a lever with a hard stop. Past the out-of-pocket maximum, the plan pays every additional dollar and deductible design has no further effect. Enter your plan to find the stop.

Load a benchmark

Your plan design stops mattering above

$23,320

member share of the claim past the stop, plan pays 100% of each new dollar
A claim ofMember paysPlan paysMember share

What a typical employer plan looks like

Average single deductible, all firms$1,886
Average single deductible, 200 or more workers$1,670
Average single deductible, under 200 workers$2,631
Covered workers facing a deductible of $2,000 or more34%
Average coinsurance rate, hospital admission20%
Out-of-pocket maximum above $3,000, single coverage72%
Out-of-pocket maximum above $6,000, single coverage21%
Out-of-pocket maximum of $2,000 or less, single coverage12%
2026 statutory ceiling, ACA, self-only$10,600
2026 statutory ceiling, HDHP, self-only$8,500
Average single premium, for scale$9,325

2. The claims your plan likely sees

The same claim sizes, drawn as a population. Start by picking your funding ecosystem, since what a plan can influence depends on who holds the contracts. Then hover or tap any marker: each one tells you where the number came from, how much a plan can typically influence it, and roughly how many dollars that influence is worth on one case. Color is influence, size is dollars, and a big claim with a small dot is money you can see and cannot reach.

Share of people spending at least a given amount per year, with each reference point rated for how much a plan can influence it under two funding models.

Start here: pick your funding ecosystem

high impact mid low pharmacy, a separate contract Color is how much a plan can influence the claim. Size is how many dollars that comes to on one case. A big claim with a small dot is money you can see and cannot reach. Those dollar figures are modeled from published price differentials and assume the plan actually captures the gap. They are not measured plan results. A marker can rate mid or high with nothing movable on the case itself, where the lever works across the population rather than on the individual claim.
Figure 1. Share of people spending at least a given amount in a year, with reference points priced at commercial employer-plan allowed amounts. US employer-sponsored group coverage, with one labeled exception disclosed on the type 1 diabetes marker; Medicare, Medicaid and individual-market data are otherwise excluded. Imaging, general surgery and cardiology figures are professional plus facility fee, four largest commercial insurers, 2023 contract year, from Philips and Whaley in Health Affairs Scholar and JAMA Network Open. ACL reconstruction and its site-of-care differential are MarketScan commercial claims via two American Journal of Sports Medicine analyses. Maternity and newborn figures are employer-plan claims from KFF analysis of MarketScan. Knee replacement is UnitedHealth Group analysis of UnitedHealthcare commercial claims, 2023. Insulin is Health Care Cost Institute employer-sponsored claims, gross of manufacturer rebates. Type 1 diabetes total is Journal of Managed Care & Specialty Pharmacy, 2020. The GLP-1 figure is modeled from published net-price data in the Fringe Theory analysis. Site-of-care differential for infusion from Journal of Managed Care & Specialty Pharmacy, 2025. Dialysis from JAMA Network Open analysis of Health Care Cost Institute commercial claims. Markers that would otherwise sit on top of each other are nudged a few pixels vertically so both remain readable; horizontal position, which carries the dollar amount, is exact. Distribution fitted between published employer-plan spending bands from KFF analysis of MarketScan, and the million-dollar claim rate from Sun Life, 2026. Impact ratings are editorial judgment, not a measured quantity, and describe what a plan can typically influence rather than what any particular plan will achieve. Dot size is the dollars a plan can move on one case, which is an estimate derived from the cited spreads. General information about plan economics. Not legal, tax or clinical advice.

3. What a design change is actually worth

This is the addition, and for most readers it will be the useful one. Enter your current design in one column and the alternative you are weighing in the other. The tool returns the actuarial value of each, the difference between them, and a premium estimate for the move, decomposed so you can see what the fixed costs of the arrangement keep from you.

If you are running a midyear review or weighing renewal alternatives, this is built to sit beside that conversation. Model the deductible bump before the meeting and you will generally find it is worth less than the room assumes.

One thing this tool does on purpose: it does not model copay tiers in fine detail. The precision matches the size of the effect. Copay-level changes move the plan's economics so little that modeling them finely would only decorate a rounding error. If you want design changes members actually feel, the leverage generally runs the other way, toward targeted improvements. Richer physical therapy benefits that support surgical avoidance, $0 generics that support adherence, account funding that buys value without touching the medical rate. Those are the design conversations worth an hour, and this tool will at least show you what the traditional ones are worth first.

What a plan design change is actually worth

What a design change is actually worth

Enter the plan you have on the left and the one you are weighing on the right. This is a directional model, not a quote and not an actuarial opinion. It is built to size a change before the renewal meeting, so you walk in knowing roughly what the move is worth. Your own claims, your own underwriting, and your own carrier's rating will differ. Everything runs in your browser and nothing is stored. Money inputs are monthly except account funding, which is annual.

Current plan

Alternative

Assumptions and premium inputs

Actuarial value is computed by integrating member cost sharing over a claims distribution fitted to four public anchors: 12% of covered members with no claim, 44% under $1,000, 15% at $10,000 or more, and a mean of about $6,700, from KFF analysis of Merative MarketScan, with the million-dollar claim rate from Sun Life's 2026 stop-loss report. Induced utilization factors are 1.00, 1.03, 1.08 and 1.15 by metal tier, from the CMS Summary Report on Permanent Risk Adjustment Transfers for the 2024 Benefit Year. The fixed-cost share defaults to 15%, which is the retention assumption for a fully insured large group. This models a single member in a single plan year, in network, and ignores copay-only services and separate pharmacy accumulators.

Current AV

74.8%

Alternative AV

69.7%

Change

-5.1 pts

What moves the premiumPer monthPer year
If the whole premium moved with actuarial value
What survives the fixed-cost floor
Induced utilization effect
Estimated premium change
Where the member landsCurrentAlternative
Cost sharing stops at a claim of
Average member cost sharing per year
Plan liability AV, before account funding

Two actuarial values are reported and they answer different questions. Total AV is net of employer account funding and is the figure comparable to a metal tier. Plan liability AV is gross of it and is what drives the premium, because account dollars buy value without moving the medical rate. Modeled results describe what the fitted distribution implies, not what any particular plan will achieve. Not actuarial advice.

Where the argument lives

The tools stand on their own, but they were built to carry an argument, and the argument lives in Just a tree: where plan design stops reaching, where the money actually sits, and why the same small move keeps getting made anyway. If a number here surprises you, that piece is where the sourcing and the reasoning are.

Figures and benchmarks are drawn from published sources cited within each tool. Modeled results describe what published differentials imply, not what any particular plan will achieve. Fringe Theory is independent and unaffiliated. Views expressed are my own and do not represent those of my employer. Nothing here is legal, tax, medical, or investment advice.