Which Workers' Comp Carriers Actually Write Hard Accounts
- Evan Swan
- 3 days ago
- 7 min read
Quick answer: which carriers write hard-to-place workers' comp
Four markets write hard workers' comp accounts, in order of cost: standard admitted carriers (clean accounts, roughly sub-1.20 mod), specialty and non-standard admitted carriers (high mods, hazardous classes), PEO co-employment (which substitutes the PEO's blended mod, usually near 1.00), and the residual market. No single carrier is "best" — the account's mod, class code, loss pattern, and state determine which of the four will price it competitively.
Sources: NCCI 2026 State of the Line; NAIC 2024 market share data. Verified August 4, 2026 by CPR Business Solutions.
"Which carrier writes this?" is the wrong first question. The right one is "which of the four markets does this account belong in?" — because the carrier list inside each market is short, well-known, and largely interchangeable, while sending an account to the wrong market wastes six weeks and burns the relationships you'll need later.
This guide covers the four markets, the terminology that trips people up, what underwriters at each tier are actually deciding, and how to route an account on the first try.
The 2026 market: better than it's been in a decade
Before routing anything, know what market you're routing into. Per NCCI's 2026 State of the Line, private-carrier net written premium held at $41.6 billion for 2025 ($45.6 billion including state funds), and the calendar-year combined ratio came in at 91% — the twelfth consecutive year of underwriting gains for the line.
Three numbers from that report matter for a hard account:
The residual market is down to roughly 5% share. The pool is not where hard accounts are landing anymore. Carriers are competing for business that would have been declined outright in 2015.
Lost-time claim frequency fell 2% in 2025 — a milder decline than the long-term average, but still a decline.
Medical and indemnity severity each rose 4%. Frequency is falling, cost per claim is climbing — which is why underwriters scrutinize claim patterns more than claim counts.
The practical read: the accident-year combined ratio was 102%, meaning current-year business is being written at a small underwriting loss and profitability is coming from reserve releases on old years. That's a late-soft-market signature. It won't last forever, and while it holds, it is the best window in years to move a distressed account back into voluntary coverage.
First, a terminology correction: there is no "E&S workers' comp"
You will hear "we'll take it to the E&S market" about a hard comp account constantly. It is, in almost every state, not a real thing — and the confusion causes real problems.
Statutory workers' compensation is an admitted-market product. Because benefits are set by state statute and backstopped by state guaranty funds, workers' comp is normally written only by admitted carriers, and surplus lines placement of statutory comp is unavailable in most states. There is no non-admitted market for the underlying coverage the way there is for property or professional liability.
What people mean when they say "E&S comp" is usually one of three real things:
Specialty or non-standard admitted carriers — admitted paper, but with an appetite built for high mods and hazardous classes and priced accordingly. This is the actual answer 90% of the time.
Program business — an MGA or program administrator with delegated underwriting authority on admitted paper, focused on one industry or class family.
Excess workers' comp — genuinely non-admitted in many states, but it sits above a self-insured retention. It's for self-insured employers, not for a 1.6-mod contractor who needs a first-dollar policy.
Why the precision matters: a client told their comp is going "to surplus lines" reasonably asks whether they'll still have guaranty-fund protection. On admitted specialty paper, they will. Getting this wrong in front of a sophisticated buyer costs credibility you don't get back.
The four markets, and where each account belongs
1. Standard admitted carriers
The large national writers. Per NAIC 2024 data on a $57.5 billion market, the leaders are Travelers (6.67% share, $3.83B), Hartford (6.45%, $3.71B), AmTrust (5.86%, $3.37B), Zurich (5.01%), Chubb (4.09%), Berkshire Hathaway (3.60%), Liberty Mutual (3.23%) and Old Republic (2.60%). The top ten groups together hold about 42% of the market.
That fragmentation is the useful fact. No carrier holds even 7% of workers' comp. Appetite varies enormously by state and class, and a carrier that declines an account in one state may write it happily in another. "Travelers declined it" is not market feedback; it's one underwriter in one state.
Send here: mods under roughly 1.20, clean-to-moderate class codes, no lapse history, three years of stable loss experience.
2. Specialty and non-standard admitted carriers
Carriers whose whole business model is the accounts the standard market declines. Expect rates meaningfully above standard — the trade is availability and sustainability, not price.
Note that AmTrust and Berkshire Hathaway appear in both tiers. Several of the largest writers run separate specialty and non-standard divisions with entirely different appetites from their standard operations, which is why a decline from a carrier's standard unit says nothing about its specialty unit.
Send here: mods from roughly 1.20 to 2.00+, hazardous classes (roofing, demolition, framing, tree service, oil and gas), accounts with a single explainable severe claim, and accounts returning from the residual market with a clean recent year.
3. PEO co-employment
Structurally different from the other three. In a co-employment arrangement the client's employees are layered onto the PEO's master policy, so the client's payroll is rated on the PEO's blended mod — often near 1.00 — rather than its own. For a distressed account, this is frequently the cheapest available answer by a wide margin, because it sidesteps the mod entirely rather than paying for it.
The trade-offs are real and worth stating up front. The client gives up direct control of the comp program, takes on the PEO's administrative fee structure, and inherits the PEO's audit process — which is more complex than a standalone audit and generates most of the disputes we see. A PEO placement also doesn't repair the client's own mod; it parks it. Plan the exit at the same time you plan the entry.
Send here: mods roughly 1.30 and up where the client also wants payroll and HR infrastructure, fast-growing companies, and accounts that need coverage bound quickly.
4. The residual market
The assigned risk plan, administered by NCCI in most states, with state-fund equivalents elsewhere. It will write essentially any compliant employer that asks. At roughly 5% of the market it is smaller than it has been in years, and it should be the last option considered, not the first fallback.
Send here: only after the other three have genuinely been worked. And if an account does land here, treat it as a 12-to-24-month bridge with an active exit plan, not a destination.
Two structures that don't fit the four-market model
Competitive state funds. About twenty states operate a state fund that competes with private carriers. California's State Fund is the clearest case — it writes voluntary business, applies schedule modifications, and pays dividends. A California account placed with State Fund is having a different experience than an NCCI-state account in assigned risk, and conflating the two leads to bad advice.
Monopolistic states. North Dakota, Ohio, Washington and Wyoming require coverage through the state fund exclusively. There is no carrier selection to make. The work is rate-tier improvement inside the state system, and any national account with operations in these states needs them carved out of the main program.
What the underwriter is actually deciding
Each tier is answering a different question, and a submission that answers the wrong one gets declined regardless of quality.
Standard carriers ask: does this fit the box? Automated or near-automated. A single disqualifying feature — mod over threshold, class code off-appetite, lapse in coverage — ends it. Narrative doesn't help.
Specialty carriers ask: is this bad or is it explainable? This is where narrative earns money. "Three of five claims came from one crew at a location closed in 2024" reprices the account. A specialty underwriter is pricing the forward risk, not the past.
PEOs ask: will this client dilute our master mod, and will they pay? Credit and payroll stability weigh nearly as heavily as loss history.
The residual market asks: are you legally required to carry coverage? That's the whole underwriting question.
Routing an account on the first try
Work these in order before you send anything anywhere:
Pull the mod worksheet, not just the mod. The worksheet shows which claims are driving the number and whether any are correctable. A mod being disputed is a different account than a mod that's settled.
Read the loss runs for pattern, not total. Frequency and severity route to different markets. Ten small claims and one large claim totaling the same dollars are not the same submission.
Verify the class codes against the NCCI Scopes Manual. A misclassification inflates both premium and the expected-loss baseline, and fixing it before marketing is worth more than any negotiation afterward.
Map the state footprint. Monopolistic states carve out. Independent-bureau states (California, New York, New Jersey, Pennsylvania and others) have their own rating rules. Assessment stacks vary enormously — price the all-in cost, not the manual premium.
Then pick the market — and send to two or three carriers inside it, not a dozen across all four. Shotgunning a hard account produces a file full of declinations that follows the client to every future submission.
The four mistakes that cost placements
Shopping before diagnosing. Sending a submission to fifteen markets to "see who bites" burns the market and produces a declination history. Diagnose, then send to three.
Treating a decline as market feedback. With no carrier above 7% share and appetites that vary by state and class, one decline means one underwriter said no.
Comparing quotes across markets on rate alone. A PEO quote, a specialty admitted quote and a residual quote have different fee structures, different assessment exposure and different renewal trajectories. Compare three-year all-in cost.
Binding and disengaging. The mod that made the account hard is repairable, but only if someone works the claims and documents the prevention during the policy year.
When to bring in a wholesale specialist
Most retail agents can place clean accounts through their direct appointments. The economics shift toward a wholesale specialist when an account has been declined by two or more standard carriers, when the mod is above 1.40 (or above 1.50 outside construction), when there's a lapse or assigned-risk history, or when PEO and standalone options need to be modeled against each other rather than guessed at.
CPR Business Solutions has placed hard-to-write workers' comp since 2021. We hold direct relationships across the specialty admitted markets, the national PEO market, and the residual-market process — and we stay engaged after binding on the claims advocacy and mod-cleanup work that turns a one-time placement into a falling rate over three years.
Submit an account at proposals@cprbrokers.com or call (704) 256-5945 and we'll tell you which of the four markets it belongs in before you market it.
Related placement guides:
Comments