> ## Content Index
> Fetch the complete content index at: https://www.thefringetheory.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# It's good to have CHOICE(s)
- URL: https://www.thefringetheory.com/ichra-choice-side-effects-employers/
- Published: 2026-10-08T22:51:15.000Z
- Updated: 2026-10-08T22:51:15.000Z
- Description: Exploring the pros and cons of ICHRA plans, as well as their utility in a volatile health insurance market.
- Author: Derek Winn
- Tags: Theory

CHOICE, the arrangement long known as ICHRA, moves the decision from the employer's plan to the employee's county. Risk, choice, carrier changes and out-of-pocket cost all get decentralized, landing person by person and market by market instead of on one renewal spreadsheet. Call it the decentralized plan. That shift is what makes it useful, and it is also what makes it hard. The problems a health plan solves are still there under CHOICE. What changes is who has to solve them.

This article picks up where [In case of emergency, break glass](https://www.thefringetheory.com/ichra-choice-arrangement-employers/) left off, with a closer look at what sits behind ICHRA, which CMS has started calling CHOICE.

- When was the last time your plan moved risk off its own books, and who decided where that line sat?
- How many insurers did your last renewal actually get in front of?
- If the plan stopped existing, what would change for your oldest employees? (read on for the answer)

## Let's talk bias

Health plan purists will pound the table on owning the risk and controlling outcomes, and they will often call attention to ICHRA as a perpetuation of the status quo. What they may be slower to admit is that many high performing self-insured plans have leaned on outside pools for years. Since the exchanges opened in 2014, many have considered pathways to transfer catastrophic risk from a self-insured group health plan through a combination of charitable funds, foundations, and more, ultimately using individual plans as a backstop. The offer is built to be compelling, often wiping out premium and out-of-pocket costs entirely, and that appeal lands hardest with members facing a serious diagnosis. That is also where the plan's savings come from.

Whether a given arrangement fits your plan is a question for your own counsel, and I would put it to them before rolling anything out. That said, I have spoken with ERISA counsel who have concerns about these programs, but cannot find an issue with how *some* are structured. Given this, simply writing off a move to a CHOICE model as more of the status quo is a bit thin.

Similarly, the CHOICE champions have their share of blind spots. The common pitch (not necessarily true) is a stable, large risk pool that every CFO should flock to. What tends to get glossed over is three fold.

- **The pool is not one pool.** It is priced rating area by rating area, and in 2026, 165 counties have a single insurer on the exchange by KFF's count, up from 93 a year earlier.
- **Plan quality and networks vary a great deal.** Many individual plans are built on narrower networks than the group plan your employees are used to. In the EBRI and Morgan Health survey this summer, roughly nine in ten large firms said some assurance on network quality would make them more likely to adopt.
- **The market moves.** Insurers enter and leave every year, and the big names have left before.

Insurers selling individual coverage, by county, 2026

1 insurer (144)2 (654)3 to 4 (1,421)5 or more (923)

144 of 3,142 counties have a single insurer selling individual coverage in 2026\. Hover or tap a county for its count. Source: CMS Marketplace public use files, plan year 2026, as compiled for the [CHOICE employer tool](https://www.thefringetheory.com/ichra-choice-employer-tool/), pulled September 24, 2026\. KFF's count of single-insurer counties, 165, uses its own method. Confirm any county on HealthCare.gov or the state exchange.

None of that is a reason to dismiss CHOICE, but it is a reason to price it person by person and review it county by county. In today's market your toolbox can't have just one or two tools in it. You won't use a sledgehammer to hang a picture, and you won't use a handsaw to fell a full grown tree.

- **Where self-funding tends to fit.** A plan that has managed its risk well over time and has not yet met a risk it cannot address. It rarely hurts when the workforce is concentrated in a few markets and there is an appetite to manage the plan year round.
- **Where CHOICE tends to fit.** Generally the opposite, a plan facing escalating uncontrolled risks, some of them without a clear solution, compounding renewal increases, a workforce spread across many rating areas, or a group that has been priced out of coverage entirely.

Plenty of employers sit somewhere in the middle, and that is where the arithmetic matters more than the philosophy. A practitioner rule of thumb is that a CHOICE arrangement usually needs a current employee-only group rate north of roughly $850 a month to pencil out. The [CHOICE employer tool](https://thefringetheory.com/ichra-choice-employer-tool/?ref=thefringetheory.com) runs that comparison against the plans where your people actually live.

## What changes after an employer moves to CHOICE?

Every effect below comes back to the same thing. The decentralized plan takes problems a group plan solved once, for everyone, and hands them out one employee at a time.

### The whole population gets de-rated

A CHOICE arrangement stops rating the group as a group. The plan's own claims history stops setting its price, because there is no longer a group plan to price. A self-funded plan still owes its run-out claims, but going forward there is no residual group plan left behind to collect a better or worse experience.

What changes is how the population is priced. A group plan over 50 employees generally blends everyone into a composite rate by tier. The individual market prices each person on their own, by age within a 3 to 1 band, by tobacco use, by where they live, and by the ages of everyone on the policy.

Given that, the winners and losers inside the workforce are sorted by age and family composition, and health does not enter into it. A 25-year-old's individual premium may sit well under what the composite charged them. A 63-year-old's may sit well over. The employer can still decide how much of that difference to absorb, but it can no longer make the price difference disappear.

### The oldest band can leave the risk

Employees on Medicare generally can't be sold an individual market plan, but a CHOICE arrangement can reimburse Medicare premiums. On a traditional group plan, an employer with 20 or more employees generally carries its working Medicare-eligible employees as primary, and the Medicare Secondary Payer rules generally keep it from steering them off the plan. Under CHOICE, the arrangement itself is the employer's plan, so the question shifts from steering people off it to what it reimburses. The employer funds an allowance, and the employee picks up Part B, Part D and any supplement or Medicare Advantage plan on their own.

The catch is the class rules, which were written with exactly this move in mind. The first question for counsel is whether your Medicare-eligible employees can be treated any differently from everyone else in their class, and I would expect the answer to be no. So the saving is real, but it tends to arrive as a byproduct of converting a whole class.

### The employee picks the plan, and more of the bill

The decision an HR team used to make once a year, which carrier and which plans, becomes thousands of small decisions made by employees in their own counties. They can pick the insurer and plan that work for them, buy up and pay the difference, or buy down. That is a real benefit for an employee whose doctors are in one local system, and a real burden for one who has never shopped for insurance. It is also why 85% of brokers in TASC's survey this year cited employee confusion.

Premium is only a portion of what they pay. Individual plans generally carry [higher deductibles than group plans](https://www.thefringetheory.com/what-a-deductible-increase-saves/), about twice as high on average. KFF puts the average marketplace deductible at $3,786 for 2026, up 37% in a year. The comparable employer figure is $1,886, for a worker with single coverage and a deductible in 2025\. Those are different populations and different plan mixes, so treat the gap as a rough guide. An allowance that covers the premium may still leave an employee holding a bigger bill the first time they actually use the plan, and buying up closes some of that gap at the employee's expense.

Some CHOICE arrangements let the unspent part of an allowance, say for someone who buys down, go toward other qualified medical expenses rather than back to the employer. That can offset some of the out-of-pocket exposure. It is an employer design choice, though, and before it goes in, ask your advisor and compliance how it interacts with HSA eligibility for employees on HSA-qualified plans.

### Carrier exits become the employee's problem

Under a group plan, a carrier leaving the market is a renewal problem for HR and the broker. Under CHOICE it lands on every employee who picked that carrier. To be fair, a group plan that changes carriers moves everyone at once, and under CHOICE only the employees with the departing insurer have to move. The difference is who does the work. The decentralized plan turns one renewal problem into many small enrollment problems, each one landing on an employee.

The big names have left before. Aetna left the exchanges for 2018, came back in 2022, then left again for 2026, taking about a million members across 17 states with it. For 2027, Cigna is leaving all 11 of its states, about 369,000 members. CareSource, PacificSource, Medica and several hospital-owned plans are leaving too, roughly 775,000 enrollees in all by one count.

For a group on CHOICE, each of those exits is a plan change its employees did not choose, landing right before open enrollment. That is the volatility a group should expect, and plan for, before it moves. The exits and the new entrants are the same feature, seen from either side of open enrollment.

### Local competition gets new buyers

Ask anyone what is wrong with healthcare pricing and you will often hear some version of the same thing, there is not enough competition. That complaint has more evidence behind it than most. Researchers looking at the first year of the exchanges estimated benchmark premiums would have run about 11% lower if every insurer selling individual coverage in each state had joined the marketplace.

Here's why that matters for CHOICE. A group plan has to cover every county where you have people, which tends to narrow the field to whoever already covers the whole state. A CHOICE arrangement puts each employee in front of every insurer selling in their county, including ones that do not "do" employer group at all. In the 30 states that run on HealthCare.gov, 106 of the 176 insurer filings for 2026 sit outside the BUCA family, and nine in ten counties have at least one non-BUCA option on the shelf. These insurers generally fall into two camps.

- **Individual market specialists.** Oscar puts it plainly in its own annual report, its plans are offered in the individual market. Centene's Ambetter is the largest marketplace carrier in the country, at 4.4 million members across 29 states at the end of 2024\. Neither company is typically on the list when a mid-size employer goes out to bid.
- **Local plans, often hospital-owned.** There are 28 health-system-owned plans on the exchange in those 30 states, reaching 27% of counties. A local plan only has to win one market, and the individual market lets it sell county by county. New ones are still showing up. 22 Health, a joint plan from two Broward County health systems, launched for 2026 in a single county.

Median share of its state's counties each insurer serves, 2026

Hospital-owned (32)**24%**

Other insurers (74)**57%**

BUCA family (70)**87%**

Medians across 176 state-level insurer filings (count in brackets) in the 30 HealthCare.gov states. Counties served are on-exchange individual medical service areas. Source: CMS Plan Attributes and Service Area public use files, PY2026\. Author's calculation.

CHOICE is a small slice of that market today, but it is growing, and the money has noticed. The HRA Council counts more than half a million employees and dependents across ICHRA and QSEHRA, roughly 3% of the 19.2 million people enrolled on the exchanges this February. It also found the number of employers offering an ICHRA roughly doubled from 2025 to 2026\. Since the fall of 2024, the companies building the plumbing for CHOICE have announced well over $300 million in new funding. Thatch announced a $108 million round at a $1 billion valuation this September, with ADP, Paychex and Eli Lilly among its backers, and Remodel Health announced more than $100 million at the end of 2024\. Venteur, Zorro and SureCo each raised between $20 and $23 million in 2025, with JPMorgan Chase's Morgan Health and Kaiser Permanente's venture arm among the investors, and Take Command and newer names like Kyra have raised money too.

So what does that money say? I think the people writing the checks expect this market to grow well past half a million lives, because that is the only way a billion-dollar valuation works, and payroll companies that already sell to well over a million small businesses are not tourists. That said, venture money is a forecast. Investors get paid to bet early, and they get plenty of bets wrong, and most of this money is going to the administrators rather than to the insurers that have to show up in your county. Still, I think there is room for more traction from here, especially if we see a resurgence of local health plans.

Let's face it, though, more insurers is not the same as cheaper care. Competition between insurers works on the premium. It does little about what the hospital or the drug costs, and when researchers compared 2017 premiums, provider-owned plans and traditional insurers came out about the same. The average number of insurers per state fell from 9.6 to 9.0 this year, and three of the hospital-owned plans in this count, Baylor Scott & White, Providence and Cox HealthPlans, have already announced 2027 exits. Local competition can leave as fast as it arrives.

### What the employer keeps, and what it gives up

Plan sponsors know the time that goes into understanding, managing and controlling a plan's costs. Under CHOICE most of that work stops. There is no plan design to negotiate, no network to steer, no stop-loss to place and no claims report to read. What is left is the allowance, how it is set by class, and what it costs to administer. Some will miss the control. Others will be glad to manage one number instead of a whole plan. Keep in mind, CHOICE doesn't stop prices from rising at or near the pace you would see on a group plan, and all of this varies by local market.

What the employer may give up is harder to measure. Whether CHOICE helps or hurts retention is, as far as I can find, not yet measured anywhere public. The case for it is portability and choice. The case against is that a plan the employer no longer runs is a plan the employer can no longer use as a reason to stay. My guess is that it cuts by age, younger employees may barely notice and older ones may notice a great deal, but I can't prove that, and anyone who says they can is ahead of the data.

## What to do

- **Map the workforce by rating area before anything else.** Count employees by county and pull how many insurers sell in each one. A workforce sitting in single-insurer counties is a different decision than one sitting in the Dallas suburbs.
- **Price your current employee-only rate against the local benchmark.** If it sits well under roughly $850 a month, CHOICE may be a harder case to make. If it sits well over, the conversation is worth having.
- **Price the plan design, not just the premium.** Compare what an employee would owe on the plans they can actually afford with the allowance, against what they owe on the group plan today.

The real decision is which problems you still want the plan to solve for everyone, and which ones you're comfortable handing to each employee. The decentralized plan is not the right answer for every employer, but it is a real one. You won't use a sledgehammer to hang a picture, and you won't use a handsaw to fell a full grown tree. Most employers have both jobs somewhere in the building.

## The Tail

The competition argument rests on local insurers staying put, and even expanding. Several of the hospital-owned plans in this count are already leaving for 2027\. If CHOICE adds buyers to the individual market faster than the subsidized population shrinks, local plans may have a reason to stay and grow. If it does not, employers may find they moved their people into a market that is consolidating behind them.

Lastly, CHOICE may yet get a shot in the arm from Washington. The House has twice voted to write it into law, and CMS and the SBA spent this fall promoting it by name. The Senate has left it out both times, and the White House's own plan talks about sending money "directly to" people without naming CHOICE. The ground work is laid. Whether the next round of reform builds on it is the open question.

## Sources

1. Dafny, Gruber and Ody, ["More Insurers Lower Premiums: Evidence from Initial Pricing in the Health Insurance Marketplaces,"](https://nber.org/system/files/working%5Fpapers/w20140/w20140.pdf?ref=thefringetheory.com) NBER Working Paper 20140, 2014; [NBER Digest](https://www.nber.org/digest/nov14/insurer-competition-and-premiums-health-exchanges?ref=thefringetheory.com), November 2014.
2. Oscar Health, Inc., [Form 10-K, fiscal year 2025](https://www.sec.gov/Archives/edgar/data/1568651/000156865126000011/oscr-20251231.htm?ref=thefringetheory.com), SEC.
3. Centene Corporation, [Form 10-K, fiscal year 2024](https://www.sec.gov/Archives/edgar/data/1071739/000107173925000027/cnc-20241231.htm?ref=thefringetheory.com), SEC.
4. CMS, Plan Attributes and Service Area public use files, plan year 2026, [data.healthcare.gov](https://data.healthcare.gov/?ref=thefringetheory.com), read October 5, 2026.
5. Howard, Herring, Graves and Trish, ["Provider-Owned Insurers in the Individual Market,"](https://www.ajmc.com/journals/issue/2018/2018-vol24-n12/providerowned-insurers-in-the-individual-market?ref=thefringetheory.com) American Journal of Managed Care 24(12), 2018.
6. KFF, ["How Has Insurer Participation in the ACA Marketplaces Changed in 2026?"](https://www.kff.org/affordable-care-act/how-has-insurer-participation-in-the-aca-marketplaces-changed-in-2026/?ref=thefringetheory.com) June 11, 2026.
7. Becker's Payer Issues, ["2 Florida systems partner on ACA plan,"](https://www.beckerspayer.com/payer/2-florida-systems-partner-on-aca-plan/?ref=thefringetheory.com) November 2025.
8. healthinsurance.org, [insurers exiting the Marketplace after 2026](https://www.healthinsurance.org/?p=1012944&ref=thefringetheory.com), updated August 4, 2026.
9. Fierce Healthcare, ["Aetna to exit the ACA exchanges in 2026,"](https://www.fiercehealthcare.com/payers/aetna-exit-aca-exchanges-2026?ref=thefringetheory.com) May 2025.
10. EBRI and Morgan Health, ICHRA employer chartbook, July 28, 2026.
11. HRA Council, annual data report, 2026: more than 20,000 businesses and more than half a million employees and dependents across ICHRA and QSEHRA, from administrators covering 75 to 80 percent of the market.
12. HHS ASPE, effectuated Exchange enrollment estimate as of February 2026 (19.2 million), June 2026.
13. TASC, 2026 Broker ICHRA Survey, September 10, 2026.
14. 45 CFR 147.102 (individual market rating factors); 26 CFR 54.9802-4 (ICHRA classes and conditions); 42 CFR 411 subpart E (Medicare Secondary Payer, working aged).
15. KFF, ["The Average Marketplace Deductible Grew by About $1,000 Per Person in 2026,"](https://www.kff.org/affordable-care-act/the-average-marketplace-deductible-grew-by-about-1000-per-person-in-2026-with-more-enrollees-shifting-to-higher-deductible-plans-as-enhanced-tax-credits-expired/?ref=thefringetheory.com) May 2026.
16. KFF, [2025 Employer Health Benefits Survey](https://www.kff.org/health-costs/report/2025-employer-health-benefits-survey/?ref=thefringetheory.com), October 2025.
17. Fierce Healthcare, ["Thatch secures $108M series C... to reach unicorn status,"](https://www.fiercehealthcare.com/finance/thatch-secures-108m-series-c-backed-general-catalyst-and-index-ventures-reach-unicorn?ref=thefringetheory.com) September 16, 2026 (also lists the $38M Series A, September 2024, and $40M Series B, April 2025).
18. MedCity News, [Remodel Health raises more than $100 million (Oak HC/FT, Hercules Capital),](https://medcitynews.com/?p=132639&ref=thefringetheory.com) December 2024.
19. Fierce Healthcare, ["Morgan Health joins $20M series A for ICHRA company Venteur,"](https://www.fiercehealthcare.com/finance/morgan-health-joins-20m-series-ichra-company-venteur?ref=thefringetheory.com) April 1, 2025.
20. HLTH, ["Zorro Raises $20M,"](https://hlth.com/insights/news/zorro-raises-20m-to-transform-benefits-administration-with-ai-powered-ichra-platform-2025-06-12?ref=thefringetheory.com) June 12, 2025.
21. FinTech Global, ["SureCo secures $23m to expand ICHRA platform" (Health Velocity Capital, Kaiser Permanente Ventures),](https://fintech.global/2025/10/02/sureco-secures-23m-to-expand-ichra-platform/?ref=thefringetheory.com) October 2, 2025.
22. Business Wire, ["Edison Partners Announces $25M Growth Investment in Take Command,"](https://www.businesswire.com/news/home/20230919140638/en/Edison-Partners-Announces-25M-Growth-Investment-in-Take-Command?ref=thefringetheory.com) September 19, 2023.
23. Caplight, [Kyra Health company profile](https://www.caplight.com/company/kyra-health?ref=thefringetheory.com) (seed round, April 2026, amount undisclosed).
24. Insurance Business, ["CHOICE arrangements get a federal push,"](https://www.insurancebusinessmag.com/us/news/benefits/choice-arrangements-get-a-federal-push--and-a-compensation-question-589181.aspx?ref=thefringetheory.com) September 9, 2026, citing the HRA Council 2025-2026 report (employers offering ICHRAs up 99%).
25. U.S. Congress, [H.R. 6703, Lower Health Care Premiums for All Americans Act,](https://www.congress.gov/bill/119th-congress/house-bill/6703?ref=thefringetheory.com) passed the House December 17, 2025; CHOICE provisions also in the House-passed H.R. 1, May 2025.
26. SBA, [release 26-89,](https://legacy.sba.gov/article/2026/09/03/sba-recognizes-innovative-choice-arrangements-major-win-small-businesses?ref=thefringetheory.com) September 3, 2026; CMS, [press release](https://www.cms.gov/newsroom/press-releases/cms-sba-georgia-recognize-innovative-choice-arrangements-major-win-small-businesses?ref=thefringetheory.com), September 14, 2026.
27. White House, "The Great Healthcare Plan," January 15, 2026, as summarized by [Butler Snow](https://www.butlersnow.com/news-and-events/president-trump-reveals-the-great-healthcare-plan?ref=thefringetheory.com).

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. Figures described as modeled or derived are my own arithmetic from the public sources listed, not disclosed data.