Initiated 2026-09-28 · Price $55.69 (as of 2026-09-25 close, stockanalysis.com) · Mkt cap $5.11bn · Financials / specialty P&C insurance · Model: verified

Rating: Hold — Conviction: Low

RLI is one of the most consistent underwriters in US specialty insurance: 2025 was its 30th straight year of underwriting profit. It returns nearly all its earnings through special dividends. The cycle is turning against it, though. E&S property premium is falling market-wide, casualty (60% of premium) already runs at a ~98% combined ratio, and reserve releases flatter the headline 85.8%. At ~2.9x book and ~17x annualized operating earnings, a premium to W. R. Berkley, the price assumes discipline holds through the softening. The model's base case sits below the price, which argues against a Buy. It stops short of a Sell because the reported numbers have not yet deteriorated. Conviction is Low.

Model value range vs price
Bear $22.05Base $47.09Bull $77.67Price $55.69

Business overview

RLI writes specialty commercial and personal lines through wholesale and retail brokers in three segments. Casualty (H1 2026 GPW $646mm) covers commercial excess, personal umbrella, transportation and professional lines. Property ($354mm) covers E&S commercial property, marine and Hawaii homeowners. Surety is $84mm. H1 2026 net premiums earned rose 3.5% to $828.5mm. The combined ratio was 85.8%: casualty 98.2%, property 59.4%, surety 90.4%. That included $61.6mm (7.4 points) of favorable prior-year development. Net investment income rose 16% to $88.4mm on a $4.9bn portfolio, of which ~$1bn is equities, ~55% of book value. Earnings turn on property pricing, casualty loss trends and reserve releases, and the reinvestment yield.

Competition

RLI competes with W. R. Berkley, Kinsale, Skyward, Markel and the Lloyd's market in E&S, and increasingly with standard carriers returning to classes they had left. Industry stamping-office data show E&S property premium fell 13.7% in H1 2026 while transaction count rose 15.2%, a sign of rate compression. Non-professional liability rose 11.2% (WSIA, reported 2026-08-05). RLI's E&S property premium fell a smaller 6%, and it says submissions now go to "over 45 markets," with quoted rates "about the 2022 timeframe." Pressure shows up first in property price and then in casualty once capacity migrates. On stockanalysis (2026-09-25), W. R. Berkley trades at 2.64x book and 14.1x forward earnings, Kinsale at 3.70x and 15.6x, and Skyward at 1.92x and 10.5x. RLI trades at 2.92x book. Its 20.9x forward P/E is not comparable, because its 11.7x trailing figure includes investment gains.

Bull case

  1. Casualty keeps hardening. Casualty premium rose 11% on ~10% rate. Personal umbrella rose 26% (17% rate) and transportation 19% (8% rate), while competitors pull back. Plays out if liability rates stay near double digits into 2027 and property premium stabilizes. Model: npe_growth
  2. Discipline holds the combined ratio in the mid 80s. Property still runs at 57–59% even at 2022-era rates, and transportation claim counts are falling. Plays out if releases continue and the 2026 cat load stays moderate. Model: combined_ratio
  3. Investment income compounds. New money yields 4.9%, ~60bp above book yield, and the $300mm of 2036 notes adds float. Plays out if the 10Y stays near 5%. Model: investment_yield

Bear case

  1. Property softening spreads. E&S property is already −13.7% market-wide, and standard carriers are re-entering. Plays out if liability rate increases fade in H2 2026, as management expects, and NPE shrinks in 2027. Model: npe_growth
  2. Casualty reserves turn. Casualty is already at ~98% after releases, and social inflation on umbrella and transportation erodes redundancy. The combined ratio moves toward 95%. Plays out if prior-year development turns adverse in casualty. Model: combined_ratio, cost_of_equity
  3. The premium multiple compresses. At 2.9x book against Berkley's 2.6x, a normalising ROE re-rates the stock. Model: not in the value range. Insurer value comes from residual income on book, and multiple shocks move only the cross-checks, so this bullet is expressed through bear:2's combined ratio and cost of equity.

Valuation & balance sheet

Metric (definition) Current Own history (range or 5y avg) Peers Source, as-of
P/B (book $19.09/share, June 2026) 2.92x n/a (unverified) WRB 2.64x · KNSL 3.70x · SKWD 1.92x stockanalysis.com, 2026-09-25
Forward P/E (stockanalysis consensus) 20.9x (distorted; ~16.7x on annualized H1 operating EPS $3.34) Trailing 11.7x incl. gains WRB 14.1x · KNSL 15.6x · SKWD 10.5x stockanalysis.com, 2026-09-25; Q2 release
Combined ratio (GAAP) 85.8% H1 2026 83.6% FY2025; 83.4% H1 2025 n/a (unverified) Q2 2026 and FY2025 releases
ROE (TTM, net income incl. gains) 25.2% n/a (unverified) WRB 20.2% · KNSL 30.3% · SKWD 17.3% stockanalysis.com, 2026-09-25

Model-implied value range (from model-summary.json; insurer module, residual income and justified P/BV, midpoints): Bear $22.05 · Base $47.09 · Bull $77.67 per share, i.e. implied returns of −60.4% / −15.4% / +39.5% vs $55.69. These ranges show how the bull and bear drivers translate into value; they are not price targets. The price sits about 30% of the way from base to bull, so the market pays for a softer cycle than the base case assumes.

Re-rating. The two methods agree within 4% (residual income $46.23, justified P/BV $47.96). The base embeds a de-rating: justified P/BV of ~2.5x FY2025 book ($19.29/share, est. value_b ÷ FY0 book per share), against 2.89x today on the same book, a 14% de-rating.

Tail. Casualty is ~60% of premium, so a harsher sensitivity was run on a scratch copy: NPE +2% / −4% / −3% / 0% / +2%, combined ratio 90% rising to 100% then 97%, investment growth near zero, a 3.3% yield and a 10.5% cost of equity. It gives $12.54 (−77.5%).

Why Hold and not Sell. The base case is a forecast of the cycle, not an observation. Reported property margins (57–59%) leave room to absorb rate cuts, and investment income is rising 16–17%. The Q2 buyback at $51.25, about $12mm or 0.2% of market cap, is not a valuation endorsement. The two $2.00 specials return capital; they are not growth.

Balance sheet: net leverage not meaningful. The only debt is $300mm of 5.375% notes due 2036 ($297.0mm carried), 0.57x FY0 pre-tax earnings plus interest and ~14% of debt plus book. Coverage is 32.7x. Cash is $32.1mm, and there is no maturity before 2036. Ratings: n/a (unverified).

Model note: tier full, status built, verification verified (LibreOffice recalculation matched all 2,846 formula cells). No unverified inputs and no assumptions without basis. Scenario consistency is OK on all 14 rows.

Scenario stress test

Reasoned from the bull/bear drivers above. The model column comes from the scenario overlays (Base case + shock).

Scenario Effect Mechanism Magnitude Model Δ value vs Base ($/sh)
S1 Fast equity crash − Risk-premium spike and a mark on ~$1bn of equities (proxied in cost of equity); no operating channel Med −$4.00
S2 Slow bear / recession − Exposures shrink, yields fall, surety losses rise. Much milder than Base − Bear ($25.04), because the bear case is a reserve and pricing cycle, and casualty losses are not recession-driven Med −$3.36
S3 Rapid rate shock ± Higher reinvestment yields vs a higher cost of equity; AOCI mark not modeled Med −$3.82
S4 Slow rate grind + Portfolio reinvests higher over years; modest cost-of-equity drag Low +$2.42
S5 Soft-landing cuts ± Lower yields vs a lower cost of equity Low +$2.93
S6 Recession-driven cuts − Yields fall and the insured base shrinks Low −$2.75
S7a Credit liquidity shock − Risk-off; bond and equity marks (proxied) Med −$5.75
S7b Slow default cycle − Surety losses on contractor defaults; portfolio credit losses Low −$2.34
S8 Stagflation − Social and loss-cost inflation on umbrella and transportation outruns pricing; higher yields only partly offset Med −$4.73
S9a Dollar spike 0 No material effect, not modeled; domestic insurer — $0.00
S9b Dollar slide 0 No material effect, not modeled — $0.00
S10 Melt-up + Lower cost of equity; equity-portfolio gains (proxied) Med +$4.83
S11 Energy supply shock − Transportation and marine claim costs rise Low −$0.25
S12 Mega-cap/AI derating − Equities are ~55% of book; the mark is proxied by +25bp on the cost of equity Low −$2.09

No scenario moves base value by more than ~12%, so nothing is labelled High. Currently active/on watch per the playbook: state.md lists S3 as partially active; the macro log records all three legs crossed from 2026-09-24 (10Y 5.17% on 2026-09-25). S8, S10 and S11 are on watch. S3, S8 and S10 are Med here, and S11 Low.

Model value change vs Base, by scenario
S7a Credit liquidity shock−$5.75S8 Stagflation−$4.73S1 Fast equity crash−$4.00S3 Rapid rate shock−$3.82S2 Slow bear / recession−$3.36S6 Recession-driven cuts−$2.75S7b Slow default cycle−$2.34S12 Mega-cap/AI derating−$2.09S11 Energy supply shock−$0.25S9a Dollar spike$0.00S9b Dollar slide$0.00S4 Slow rate grind+$2.42S5 Soft-landing cuts+$2.93S10 Melt-up+$4.83

What would change the call

Upgrades if:

Downgrades if:

Watch items

Sources