How to Lower Employee Benefits Costs Without Losing People
How to lower employee benefits costs: plan design, reimbursement arrangements, pre-tax structures, renewal tactics, and the cuts that quietly backfire.
My first renewal quote came back at nineteen percent. I called the broker expecting a negotiation and got a shrug instead. That is the market, he said, and he was mostly right, which is the part nobody warns you about. There was no argument to win. There were only different plans to buy.
What I learned over the next three renewals is that cutting benefits cost is almost never a negotiation. It is a design exercise, and it happens in a six-week window once a year. Outside that window there is very little you can do without hurting somebody.
Inside that window there is more room than most owners realize, and almost all of it sits in three places: how the plan is built, how the benefit reaches the employee, and how the dollar is taxed on the way.
This guide covers the tactics rather than the numbers: what actually moves the total, what each move costs you in goodwill, and which popular tactics reliably make things worse. I build the people and records side of this at FirstHR. This is general information rather than legal, tax, or insurance advice.
Find Out Where the Money Goes Before You Cut Anything
Before any tactic, list everything you spend on benefits, with enrollment and each line's share of the total. Most employers who try to cut costs start with the perks, save a rounding error, and never touch the health premium, where most of the controllable money sits.
Two things usually fall out of that list immediately. The first is the gap between headcount and enrollment. Your health cost is the premium multiplied by enrolled employees, not by everyone on the payroll, and a team with several people on a spouse's plan can be paying for a plan design sized for a group that does not exist.
The second is the mandatory floor. Payroll taxes and workers' compensation are not a benefits decision, and separating them out stops you from congratulating yourself on cutting something you never controlled. The legally required group is a cost of employing anybody, and it survives every strategy on this page.
The one exception inside that floor is worth a phone call this week. Workers' compensation rates follow job classification, and small businesses are routinely assigned a rate reflecting the riskiest job in the company. Running a classification audit costs nothing and regularly returns more than any perk you were considering cutting.
Then ask your carrier or broker for the utilization picture. In a small group you will not get claims-level detail, but you can usually get enrollment by tier, plan selection, and which offered benefits nobody has touched. That report is the difference between cutting on evidence and cutting on instinct.
Put all of it on one sheet before you form a view. A row for every benefit, what it cost you last plan year, how many people are actually enrolled rather than eligible, and what share of the total that line represents.
| A | B | C | D | E | F | G | H | I | J | |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Benefit or line item | Required or discretionary | Employer cost last plan year | Employees eligible | Employees enrolled or using | Uptake percent | Share of total spend | Weight in a stay-or-go decision | Decision (keep, redesign, cut) | Notes and next step |
| 2 | Health insurance premium | Discretionary | ||||||||
| 3 | Dental and vision | Discretionary | ||||||||
| 4 | Employer payroll taxes | Required | Not a benefits decision. List it so you do not count it as a saving | |||||||
| 5 | Workers' compensation premium | Required | Audit the classification codes before anything else | |||||||
| 6 | Retirement contributions and match | Discretionary | ||||||||
| 7 | Paid time off and holidays | Discretionary | ||||||||
| 8 | Life and disability coverage | Discretionary | ||||||||
| 9 | Wellness platform or subscription | Discretionary | ||||||||
| 10 | Stipends (home office, phone, wellness) | Discretionary | ||||||||
| 11 | Food, social budget, and events | Discretionary | ||||||||
| 12 | Training and development budget | Discretionary | ||||||||
| 13 | Other: | |||||||||
| 14 | Other: | |||||||||
| 15 | TOTAL | |||||||||
| 16 | ||||||||||
| 17 | Enrollment, not headcount, is what you are paying for. Count the people actually on the plan. | |||||||||
| 18 | Fill the uptake column before the decision column. Cutting on instinct is how the wrong line gets cut. | |||||||||
| 19 | Date completed and who to revisit this with: |
Once the cost and enrollment columns are filled, look at the two that argue with each other: cost and weight in a stay-or-go decision. The lines where they disagree are where the decisions actually are.
Employee Benefits Analytics at Small Scale
Employee benefits analytics, at this size, is four numbers tracked across years rather than a dashboard anybody logs into. The carrier will not release member-level claims for a small group, but the summary reporting you can get is enough to show which lines are working and which are being funded out of habit.
| What to track | What it tells you | Where the number comes from |
|---|---|---|
| Employer cost per enrolled employee per month | The only cost figure that survives a change in headcount | The carrier invoice divided by enrolled lives, never by payroll headcount |
| Enrollment rate and the split across tiers | Whether you are funding a plan design sized for a group that does not exist | The carrier enrollment report, or your own payroll deduction register |
| Uptake on each discretionary line | Which benefits are being paid for out of habit rather than used | Vendor logins, claim counts, and the inventory sheet above |
| Renewal increase three years running | Whether your increases are the market or specific to you | Your own renewal letters, which nobody keeps and everybody needs |
Two cautions come with numbers this small. One expensive claimant can move the trend line for a twenty-life group further than any design change you make, so a single bad year is not a trend. And your own history is a weak benchmark, which is why measuring your cost per employee against a national employer survey says more than measuring it against yourself.
Pull the same figures at the same point every year, roughly a week before you ask for the renewal. Numbers that arrive after the plan is chosen are a report. Numbers that arrive before the renewal conversation are the only version that changes what you pay.
Plan Design: Deductible, Network, and the Contribution Split
Plan design is where most of the controllable money sits, and it comes down to three dials: how much risk sits in the deductible rather than the premium, how wide the network is, and how the premium is split between you and the employee. Each one is quotable, so ask for numbers rather than guessing.
| Design change | What it does to your premium | What it does to employees | How to soften it |
|---|---|---|---|
| Raise the single and family deductible | Reduces the fixed monthly cost you pay all year | Shifts risk onto whoever actually gets sick, which is a minority in any given year | Fund a health savings account or an integrated reimbursement arrangement with part of the saving |
| Move to a narrower network | Can produce a substantial reduction where the network exists | Ranges from invisible to unusable depending on whose doctor is excluded | Check the directory against the providers your team names before you switch |
| Increase the employee share of the premium | Reduces your cost immediately and proportionally | Reads as a pay cut, because on the payslip it is one | Announce a full cycle ahead and show the per-paycheck difference in dollars |
| Contribute a fixed dollar amount rather than a percentage | Caps your exposure to future increases at a number you choose | Employees absorb the increase above your contribution | Set the amount against the employee-only premium and revisit it annually |
| Fund employee-only richly, dependents lightly | Concentrates spend where enrollment is highest | Penalizes employees with families, who are frequently the ones least able to move | Consider a tiered contribution by salary band instead |
| Add a lower-cost plan alongside the current one | Reduces cost only to the extent people choose it | Preserves choice, which protects the employees who need the richer plan | Requires clear comparison materials or nobody switches |
The deductible dial is the biggest and the most abused. Moving risk out of the premium works because most employees never reach the deductible in a given year, so you are lowering a fixed cost in exchange for a contingent one. The failure mode is banking the entire saving and letting employees absorb the whole shift.
The right pairing is a higher deductible plus employer money flowing back. A qualifying high deductible plan lets employees open a health savings account, and your contribution to it is excludable, which means it reaches them whole. An integrated health reimbursement arrangement does something similar and costs you nothing unless somebody claims.
Narrow networks are the underrated one. The saving is genuine, and so is the risk of buying a plan your team cannot use. Before signing, ask three or four employees to name their doctors and check those names against the directory yourself. Ten minutes of that is worth more than any brochure.
On the contribution split, the structural choice matters more than the percentage. Contributing a fixed dollar amount rather than a fixed percentage caps your exposure to next year's increase at a number you chose, which converts an open-ended commitment into a budget line. It also transfers the increase to employees, so it needs to be paired with an annual review rather than set and forgotten.
When a Reimbursement Arrangement Beats a Group Plan
A reimbursement arrangement replaces a renewal you receive with an allowance you set. That is the whole argument, and for a small employer it is usually worth more than the raw dollar difference. You stop being a price taker on your single largest benefits line.
Two versions matter. A qualified small employer arrangement (QSEHRA) is capped and open only to employers below 50 full-time equivalents. Those employers must offer no group health plan, health reimbursement arrangement, or health flexible spending account to anyone (HealthCare.gov). An individual coverage arrangement has no dollar cap, no size limit, and more design complexity.
| Delivery route | Who absorbs the annual increase | Budget predictability | Administrative load |
|---|---|---|---|
| Fully insured group plan | You, at whatever the renewal says | Low. The number arrives rather than being chosen | Low. The carrier runs the plan |
| Level-funded plan | You, based on your own claims experience | Moderate within the year, low across years | Moderate. More reporting and more decisions |
| Qualified small employer arrangement | The employee, above the allowance you set | High. You choose the number and it holds | Low to moderate. A per-employee administrator fee |
| Individual coverage arrangement | The employee, above the allowance you set | High, with class design flexibility | Moderate. Class rules, notices, and coverage verification |
| No health benefit at all | Not applicable | Total | None, and it shows in hiring |
The arrangement is cheaper in three recognizable situations. Your group renewal is climbing faster than an allowance would; your team is spread across states where no single network works; or your enrollment is low enough that you are maintaining a group plan for a handful of people. Any of those makes the switch worth pricing.
It is not cheaper in two others, and both deserve saying plainly. Individual market premiums for older employees in high-cost areas can exceed what a group rate would have been, because group rating spreads age across the whole census.
Marketplace subsidies are the second problem. Under 26 U.S.C. 36B(c)(4), an arrangement that counts as affordable coverage ends an employee's premium tax credit outright for those months, and an unaffordable one cuts the credit by one twelfth of the permitted benefit each month. A lower-paid worker can finish net worse off even as your cost falls.
For taxable years beginning in 2026 the qualified small employer ceiling is $6,450 for self-only coverage and $13,100 for family coverage, per IRS Revenue Procedure 2025-32. Those are ceilings, not targets. Funding half of one is still a real benefit, and unlike a premium it costs you nothing in any month where nobody claims.
Self-Funding and Level-Funding at Small Scale, Honestly
Level funding is the small employer version of self-insurance. You pay a fixed monthly amount split into a claims fund, administration, and stop-loss premium, and you get part of the unused claims fund back if the year goes well. The saving is real. So is the reason it is available.
The mechanism that makes the first-year quote attractive is medical underwriting. A fully insured small group plan is priced under community rating rules that ignore your claims. A level-funded plan is not, so a healthy group gets a rate the fully insured market is not allowed to offer them. That is a genuine arbitrage, and it only lasts while the group stays healthy.
Per the Kaiser Family Foundation 2025 Employer Health Benefits Survey, 37 percent of covered workers at firms with 10 to 199 workers are in a level-funded plan, while 27 percent are in a plan their employer reports as self-funded.
At small scale the volatility is the whole story: with a couple of dozen enrolled lives, a single catastrophic claimant is not a statistical event, it is your entire year.
Three details decide whether the arrangement is survivable. The stop-loss attachment point, both per person and in aggregate, is what caps your exposure, so read it before the premium. Terminal liability coverage decides whether you owe for claims incurred before the plan ended but reported afterward. And the refund terms decide whether a good year actually returns money to you or mostly to the administrator.
The last piece is legal rather than financial. The Department of Labor states that most employers sponsoring a fully or partially self-funded group health plan exercise discretionary authority over it and are therefore fiduciaries of it, meaning they must run it in participants' interest (Department of Labor).
Fiduciary duty is a standard of conduct rather than a filing, so errors surface at audit or when a participant complains, both of which are the expensive way to find out.
Pre-Tax Structures That Cut the Bill on Both Sides
A Section 125 cafeteria plan lets employees pay their share of premiums and certain other costs before tax. Those amounts leave the wage base for federal income tax and for both halves of Social Security and Medicare, which means the employer saves 7.65 percent on every dollar routed through it, up to the Social Security wage base.
This is the closest thing to free money on the list. It does take paperwork: the IRS describes a cafeteria plan as a separate written plan, and it comes with nondiscrimination testing, so it is not a payroll checkbox. It is still inexpensive to establish.
Work the arithmetic on your own numbers. If ten employees each contribute $4,000 a year toward their premium and all of it runs pre-tax, that is $40,000 removed from the wage base and roughly $3,060 of employer payroll tax you do not owe. The employees each keep several hundred dollars they would otherwise have paid in tax. Nobody takes home less.
A health flexible spending account extends the same treatment to out-of-pocket costs, and a dependent care account does it for childcare. Both produce the same 7.65 percent reduction on the amounts employees elect, and the dependent care account costs the employer nothing beyond the administrator fee.
The health account puts one risk on the employer. The IRS requires the full annual election to be available at any time during the coverage period, regardless of how much the employee has contributed so far (Publication 969). If an employee spends the whole election early and then leaves, you absorb the difference.
Two honest caveats. Pre-tax elections reduce reported Social Security wages slightly, which has a marginal long-run effect on the benefit an employee eventually collects.
The second caveat is nondiscrimination. Where a plan favors highly compensated employees (broadly, officers, major owners, and top earners) on eligibility, contributions, or benefits, the IRS requires you to include in their wages the value of the taxable benefits they could have selected (Publication 15-B).
The Small Employer Tax Credit Most Businesses Never Check
A small employer can claim a credit worth up to 50 percent of the premiums it paid, for two consecutive tax years. To qualify it needs fewer than 25 full-time equivalent employees and average annual wages below an inflation-adjusted threshold.
The employer must also pay at least half the cost of employee-only coverage and buy that coverage through the Small Business Health Options Program (SHOP). The rules are worth ten minutes of checking (Internal Revenue Service).
The credit is 35 percent rather than 50 for tax-exempt employers, and it operates on a sliding scale, so the full amount goes only to the smallest and lowest-wage employers.
The wage ceiling is where most teams fall out. For tax year 2025 the instructions to Form 8941, the form used to claim the credit, require average annual wages of less than $67,000 per full-time equivalent, and the credit already starts shrinking above $33,000 (IRS).
There is a catch. The credit requires a qualified health plan bought through the SHOP marketplace, and the IRS has had to publish relief for employers whose county offers no such plan at all. Check whether it applies to you, take it if it does, and do not build a benefits strategy around it.
Cut What Nobody Uses, Protect What People Weigh
Pull the uptake data before you pull a benefit. A perk with four percent participation is money back with no cost to morale. A benefit that 60 percent of your team uses is a pay cut wearing a different name, and it will be received as one.
The test that works is two columns: what a benefit costs you per year, and how much weight it carries in somebody's decision to stay. Those two are not correlated. A catered lunch can cost a lot and still carry little weight in whether someone stays. A retirement match with immediate vesting carries far more.
| Benefit | Typical uptake at a small employer | Weight in a stay-or-go decision | Cut it? |
|---|---|---|---|
| Health coverage | High among eligible employees | Highest of any benefit | Redesign it, do not remove it |
| Retirement match | Moderate to high where auto-enrollment exists | High, and rising with employee age | Change the formula before removing the match |
| Paid time off | Universal | High | Adjust accrual and carryover rules, not the headline number |
| Dental and vision | Moderate | Moderate, and high value per dollar spent | Keep. Cheap relative to how visible it is |
| Life and disability coverage | Low awareness, low claims | Low until somebody needs it | Move to employee-paid voluntary coverage |
| Wellness platform subscriptions | Often low | Low | Cut first and measure the reaction |
| Catered food and social budget | High attendance, low weight | Low | Cut, but explain why rather than letting it vanish |
| One-size stipends | Varies wildly by employee situation | Low to moderate | Replace with an excludable benefit or a menu |
Where a benefit is valued but expensive, voluntary benefits are the middle path. The employee pays the premium, you arrange the access and the payroll deduction, and the group rate is better than anything they could buy alone. Your cost is administration, and the offer still counts as something you provide.
The cheapest research available is asking. A short survey tells you which two or three benefits your team actually weights, and businesses routinely discover they are funding something nobody cares about while underfunding the thing that keeps people.
The Renewal Negotiation, and When to Market the Plan
Start the renewal conversation 120 days before the plan year ends, and understand what is negotiable before you start. In the small group market, 42 U.S.C. 300gg lets the rate on a non-grandfathered plan vary only by family size, rating area, age, and tobacco use. Which plan you buy changes the price; your own claims history never enters the rate. There is no persuasion available.
That single fact reframes the whole exercise. You are not negotiating a price down. You are choosing among priced options, and the work is making sure you see all of them in time to compare properly.
Broker compensation disclosure is not a favor you are asking for. Under section 408(b)(2)(B) of the Employee Retirement Income Security Act (ERISA), added by the Consolidated Appropriations Act, 2021, a broker or consultant expecting $1,000 or more from a group health plan covered by ERISA must disclose its direct and indirect compensation to the plan fiduciary in writing (Department of Labor).
Line the options up in one place as the quotes come back, so the renewal as offered, the same plan at a higher deductible, a level-funded quote, and a reimbursement allowance are all answering the same questions in the same order.
| A | B | C | D | E | F | |
|---|---|---|---|---|---|---|
| 1 | What to compare | Current plan | Renewal as offered | Same plan, higher deductible | Level-funded quote | Reimbursement allowance |
| 2 | Carrier or route, and plan name | |||||
| 3 | Date quoted, and how long the quote holds | |||||
| 4 | Employer cost per enrolled employee per month | |||||
| 5 | Employer cost per year at our current enrollment | |||||
| 6 | Change against what we pay today | |||||
| 7 | Employee-only contribution per paycheck | |||||
| 8 | Family contribution per paycheck | |||||
| 9 | Single deductible | |||||
| 10 | Family deductible | |||||
| 11 | Out-of-pocket maximum | |||||
| 12 | Network name | |||||
| 13 | Doctors our team named: how many are in this network | |||||
| 14 | Employer money returned to employees (health savings account or reimbursement arrangement) | |||||
| 15 | What a median-wage employee pays in a typical year | |||||
| 16 | What a median-wage employee pays in a bad year | |||||
| 17 | Who absorbs next year's increase | |||||
| 18 | What happens if we have one expensive claimant | |||||
| 19 | What it takes to reverse this decision next year | |||||
| 20 | Administration on us: setup, notices, ongoing | |||||
| 21 | ||||||
| 22 | Our decision and the reason for it | |||||
| 23 | Announced to the team on | |||||
| 24 | Open enrollment opens on | |||||
| 25 | Date to start this exercise again (120 days before the next plan year ends) |
The rows that decide this are near the bottom rather than the top. Two options with the same employer cost are not the same option if one of them puts next year's increase on your side of the table and the other puts it on an employee earning your median wage.
Marketing the plan means getting quotes from several carriers on a clean census. Two things make that go wrong: a census with stale dependent data, which produces quotes you cannot compare, and multiple brokers approaching the same carrier, which usually gets both quotes blocked. Pick one representative and give them the whole market.
Whatever you decide, the deadline that matters is open enrollment, not the start of the plan year. Employees need materials, a comparison, and a window to act.
Dwight SchruteSales3 people
Darryl PhilbinWarehouse1 person
Kelly KapoorCustomer Service1 personWellness Programs and the Evidence Problem
The best evidence available says a typical workplace wellness program does not reduce medical spending. That evidence is a randomized controlled trial of more than 4,800 University of Illinois employees. Its design was sound and its result is inconvenient, which is exactly why it is worth knowing before you sign anything.
The reason vendor return-on-investment claims look so much better is selection. Most of those studies compare people who joined a program with people who did not, and the people who join are healthier to begin with.
The Illinois researchers measured that gap. In an earlier paper from the same study, employees who chose to participate had lower medical spending and healthier behaviors than nonparticipants in the year before the program began (Quarterly Journal of Economics, 2019). Randomization removes that gap, and once it is gone the savings largely disappear.
None of that makes a wellness program worthless. It makes it a benefit people may enjoy rather than a cost control measure, and it should be budgeted accordingly. If you want to spend here, a simple wellness stipend gives employees the choice and costs you nothing in platform fees. Remember, though, that cash stipends are taxable wages.
If you do run a program with financial incentives attached, the legal caps matter. A health-contingent wellness reward, one that depends on meeting a health target or completing an activity tied to a health factor, is limited to 30 percent of the total cost of employee-only coverage, rising to 50 percent where the additional amount relates to tobacco prevention or reduction (29 CFR 2590.702).
Programs involving medical examinations or health questions must also be voluntary under disability and genetic information law, and that is where most small employer programs get into trouble.
The Cuts That Reliably Backfire
Three tactics show up constantly, look like savings on a spreadsheet, and cost more than they return. Each one has a compliant version that works, and the difference is almost always timing and communication rather than the change itself.
Cutting the employer contribution share without warning is the first and worst. Employees read the payslip, not the memo, and a ten-point shift on a family premium is thousands of dollars of take-home pay. The change itself is legitimate. Learning about it from a paycheck is the problem, and it is the fastest way to convert a cost saving into a resignation.
There is a compliance dimension too. If you are large enough to be subject to the employer coverage requirement, raising the employee-only contribution can push the plan past the affordability threshold and create a penalty exposure that dwarfs whatever you saved.
Switching carriers mid-year is the second. When the carrier changes, deductible and out-of-pocket accumulators can reset to zero, so an employee who met a $3,000 deductible in May can start again in June. At worst, you saved four months of premium difference and handed somebody a several thousand dollar bill they had already paid once.
Some state insurance laws soften this, but only in part. Florida (Fla. Stat. 627.666) and New Mexico (13.10.5.12 NMAC), for example, make the new insurer credit deductible expenses from the 90 days before its policy starts. Deductible spending from earlier in the year counts only if the new carrier agrees to credit it.
Chasing the cheapest premium into a network nobody can use is the third, and it is the one that produces the worst possible outcome: you pay the premium and nobody can use the benefit, because everybody is out of network. The plan looks identical on a comparison sheet. It is not the same product.
Reducing a safe harbor retirement match mid-year is allowed only if you are operating at an economic loss or your safe harbor notice warned that the match could be reduced. It takes a plan amendment and a supplemental notice, it takes effect no earlier than the later of the amendment date or 30 days after that notice reaches employees, and the plan must then pass nondiscrimination testing for the whole year (26 CFR 1.401(k)-3).
Dropping dental and vision saves very little and is highly visible, which is the worst ratio available.
Degrading the plan quietly runs into the rule that a material modification to the summary of benefits and coverage, made outside a renewal, needs 60 days of advance notice to enrollees (29 CFR 2590.715-2715).
The pattern underneath all of these is the same. The saving is real and the damage comes from the surprise. A change announced a cycle early with the arithmetic shown will be grumbled about and absorbed. The same change discovered in a paycheck, at a pharmacy counter, or in a doctor's office produces a different company.
Nick Anisimov, FirstHR founderThe Order to Do This In
Run the free changes first, the structural ones second, and the ones employees feel last. That order matters because each stage produces information the next stage needs, and because the early moves buy credibility for the harder conversation later.
| Tactic | How much it saves | Effort | Effect on employees |
|---|---|---|---|
| Audit workers' compensation classification codes | Moderate to large where misclassified | Low, one phone call | None |
| Route premium contributions through a Section 125 plan | Moderate and permanent | Low, one-time setup | Positive. They keep more of their own money |
| Remove benefits with single-digit uptake | Small | Low | Slightly negative unless explained |
| Confirm eligibility and remove ineligible enrollees | Small to moderate | Low | Negative for the affected individual only |
| Raise the deductible and fund an account with part of the saving | Large | Moderate, annual | Neutral if the offset is real, negative if you bank it all |
| Move to a narrower network | Large where available | Moderate, needs verification work | Ranges from none to severe. Verify before switching |
| Price a level-funded plan | Large in a healthy year | Moderate, plus ongoing decisions | None visible in year one |
| Switch to a reimbursement arrangement | Large and predictable | Moderate, one-time setup | Mixed. More control for them, more work for them |
| Claim the small employer tax credit if eligible | Large where it applies, rarely applicable | Low, worth checking once | None |
| Increase the employee premium share | Large and immediate | Low | Strongly negative without a cycle of notice |
| Cut a retirement match | Large | Low mechanically, restricted legally | Severe, and it damages the people most likely to stay |
Read the last two rows as a warning rather than a menu. They save the most per unit of effort and they carry the most damage, which is exactly why they are the ones a business under pressure reaches for first. Doing everything above them first is usually enough.
The administration underneath all of this is the part that quietly decides whether any of it holds. Eligibility dates, waiting periods, enrollment records, and the notices that go with each change are what turn a plan on paper into a plan that works, and keeping them in one place rather than in somebody's inbox is what FirstHR is built for.
Finally, set the review as a recurring calendar item rather than a crisis response. Benefits costs do not spike; they climb steadily and get noticed all at once. An hour every quarter, and a serious exercise 120 days before renewal, keeps you from ever having to make the panicked version of these decisions.
Frequently Asked Questions
How can a small business lower health insurance costs?
Four routes account for nearly all of the real saving. Change the plan design, which usually means a higher deductible paired with an employer-funded account so employees are not simply worse off. Change the contribution split, deliberately and with notice, rather than quietly. Change the delivery route by pricing a level-funded plan or a reimbursement arrangement against the fully insured renewal. And take employees’ share of premiums pre-tax under a Section 125 plan, so the money escapes payroll tax on both sides. Shopping the market matters too, but in the small group market carriers price on age, location, family size, and tobacco rather than on your claims, so the saving comes from choosing a different plan rather than from negotiating a better one.
Is a QSEHRA cheaper than a group health plan?
Frequently, and the reason is control rather than price. A qualified small employer health reimbursement arrangement (QSEHRA) swaps a carrier’s renewal for an allowance you choose, so your budget no longer moves on somebody else’s schedule. Under IRS Revenue Procedure 2025-32, the cap for taxable years beginning in 2026 is $6,450 when only the employee is covered and $13,100 when family members are covered too, and you can fund any amount below it. Cheaper for you does not always mean cheaper for employees, though. Older workers can pay more in the individual market than they would at group rates. And when the arrangement counts as affordable coverage, an employee loses the marketplace premium tax credit for the months it covers, which can leave a low-paid worker worse off than before.
Does raising the deductible actually save money?
Yes, and it is usually the largest single design change available to a small employer. A higher deductible moves risk out of the premium: you cut the fixed monthly cost you pay all year, and employees take on exposure that only some of them will ever hit. The mistake is pocketing the entire saving. Employees at smaller firms already shoulder more of it. The Kaiser Family Foundation 2025 Employer Health Benefits Survey puts the average single deductible for covered workers with a deductible at $2,631 in firms with 10 to 199 workers against $1,670 in firms with 200 or more. Handing part of the saving back through a health savings account contribution or an integrated reimbursement arrangement keeps the change from looking like a pure cut.
Can I lower costs by changing how much I contribute toward premiums?
You can, and it is the fastest change available, which is exactly why it does the most damage when done carelessly. On the payslip a larger employee share is a pay cut, and employees will read it that way. Two rules make it survivable. Announce it a full cycle before it takes effect, and show the change per paycheck in dollars rather than in percentage points. And if your company is big enough to fall under the employer coverage requirement, run the affordability arithmetic first: a bigger employee share can push the employee-only cost past the affordability threshold, and the resulting penalty exposure can dwarf the saving.
Is level funding a good idea for a small employer?
It can be, if your group is healthy, your cash position can absorb a bad quarter, and you accept that you are trading price for volatility. With level funding you pay one fixed amount each month that covers a claims fund, administration, and stop-loss insurance, and you may get some of the unused claims fund back after a good year. The Kaiser Family Foundation 2025 survey found level-funded coverage among 37 percent of covered workers at firms with 10 to 199 workers. The real risk arrives at renewal. Your own experience sets the new price, so one expensive claimant can produce an increase that the fully insured market will not undercut when you try to go back.
Do wellness programs reduce health care costs?
The best available evidence says no, at least not on the timescales employers are sold. In a randomized controlled trial among more than 4,800 employees at the University of Illinois, published in JAMA Internal Medicine in 2020, a comprehensive wellness program had no significant effect after 24 months on clinical measures such as weight, blood pressure, cholesterol, or blood glucose, or on medical use. It also made employees more likely to report having a primary care physician and improved their self-reported health beliefs. Vendor return on investment claims usually come from comparing volunteers with non-volunteers, and healthier employees are the ones who sign up. Budget any program as something people like rather than as a cost reduction.
Can I switch health insurance carriers in the middle of the plan year?
Usually you can, and usually you should not. A new carrier can restart every deductible and out-of-pocket tally at zero. Some states, including Florida and New Mexico, require partial credit: the new insurer must recognize deductible costs incurred in the 90 days before its policy began, while anything earlier is credited only if the new carrier agrees. An employee who had met a $3,000 deductible by spring can end up owing much of it again, which turns a cost saving for you into a several thousand dollar surprise for them. If a mid-year move is unavoidable, make accumulator transfer a written condition of the deal, confirm that in-progress treatment and prior authorizations carry over, and tell affected employees individually rather than in a group email.
What is the cheapest way to offer health benefits to a small team?
For a business that cannot get a usable small group quote, a reimbursement arrangement is normally the lowest-cost route that still counts as a real benefit. You set a monthly allowance, employees buy their own coverage, and you reimburse tax-free against substantiated claims. Nothing is spent unless somebody claims, there is no renewal increase to absorb, and the administration is a per-employee fee rather than a plan to run. The trade is that employees do the shopping, which some find liberating and others find stressful, and that the arrangement interacts with marketplace subsidies in ways worth checking before you announce anything.
Make HR the easy part of running your business
Onboarding, e-signatures, training and core HR in one place. Up and running in 5 minutes.
- 7-day free trial
- No credit card required
- Cancel anytime