Japan non-life big-three reinsurance and cat overlay matrix

TL;DR

The “Japan non-life big three” — Tokio Marine Holdings, MS&AD Insurance Group Holdings, and Sompo Holdings — share a Japan-domiciled, multi-line P&C franchise but differ on five overlay axes that drive the analytical readings: catastrophe exposure by peril, reinsurance-program shape, overseas specialty / London market presence, equity-cross-shareholding portfolio size and wind-down pace, and capital return policy. Japan-side cat exposure splits into earthquake (large-loss tail with a public-private retrocession layer through Japan Earthquake Reinsurance), typhoon / wind (annual-frequency loss driver), flood (rising secondary peril), and snow / hail (smaller but volatile). Reinsurance programs combine domestic intra-group retrocession, proportional treaties for non-cat lines, and excess-of-loss (XL) cover for cat lines purchased in the global reinsurance market. Tokio Marine has the largest overseas earnings share among the three (including a London-market platform and US specialty footprint); MS&AD’s structural feature is a dual-domestic operating-company model (MSI plus ADI) with a Toyota / mobility ecosystem linkage; Sompo’s distinctive features include the Sompo Japan domestic core and a remediation / governance arc that has shaped recent strategic disclosure. All three are running multi-year cross-shareholding wind-down programs that mechanically lower ESR equity-risk capital and free up capital for buybacks. Combined-ratio readings split into voluntary auto, fire, casualty, marine, and other commercial lines, with overseas combined ratios reported separately at the group level.

Wiki route

This overlay sits under insurance INDEX and is the cat / reinsurance companion to japan-nonlife-big-three. Read it together with the global solvency framework comparison matrix for the regulatory-capital lens, the nat-cat reinsurance Japan entry for the cat-reinsurance mechanics, the earthquake insurance public-private scheme for the household-EQ retrocession layer, the foreign-reinsurer Japan landscape for the global reinsurer counterparty set, the Japan non-life underwriting cycle for the rate-cycle backdrop, the Lloyd’s Japan syndicate operating model for the alternative-syndicate route, the marine / P&I cover market entry for the specialty-lines context, and the captive insurance Japan market entry for the corporate-self-insurance contrast.

The clean entity anchors for this overlay are Tokio Marine Holdings, MS&AD Insurance Group Holdings, and Sompo Holdings; the operating-company anchors are Tokio Marine & Nichido Fire, MSI, ADI, and Sompo Japan. The asset-management affiliate anchor is Tokio Marine Asset Management. The retrocession / reinsurance counterparty anchors are Japan Earthquake Reinsurance, Toa Reinsurance, Munich Re Japan, Swiss Re Japan, and Lloyd’s Japan.

Why this matrix matters

Reading the Japan non-life big three from financial-statement summaries alone (premium, combined ratio, ESR) does not reveal what actually drives the volatility on the balance sheet. Three structural questions sit underneath:

1. What is the cat-loss profile by peril, and how much of the volatility is ceded to reinsurance versus retained? Earthquake, typhoon, and flood losses have different return-period shapes. A high cat-retention layer means good underwriting years drop more profit to the bottom line but bad years burn more capital.
2. How does overseas business interact with domestic underwriting volatility? Each group has built a different overseas mix: Tokio Marine has the largest US specialty and London-market presence; MS&AD has US, Asia, and a Lloyd’s-syndicate route; Sompo has a US / Europe specialty footprint and Asia exposure.
3. How is the legacy cross-shareholding portfolio being wound down, and what does that do to ESR and capital return? All three groups carry a long-tail equity book inherited from postwar keiretsu and group relationships. Reducing it lowers ESR market-risk capital and frees buyback capacity; running it down too quickly damages relationships and the dividend income line.

The overlay matrix below records these axes as routing rather than as as-of-date numbers. Specific cat-loss numbers, ESR ratios, reinsurance retentions, and equity portfolio sizes change every fiscal year and should be sourced from each group’s integrated / annual report and from FSA materials.

Tokio Marine Holdings — overseas-led specialty franchise

Tokio Marine Holdings is structurally the most overseas-weighted of the big three. Its domestic non-life core is Tokio Marine & Nichido Fire. Overseas, the group operates a US specialty insurer (Philadelphia / Pure / HCC platforms), a Lloyd’s syndicate / London-market presence, and Asia / EMEA platforms. The overseas earnings share is the highest in the big three, which means the group’s combined ratio and ESR sensitivity have meaningful exposure to non-Japan cycles (US specialty soft markets, London casualty cycles, overseas cat events).

On cat, Tokio Marine’s Japan-side exposure covers the standard set (earthquake, typhoon, flood, snow / hail). Domestic earthquake (commercial property and shintaku-style risk) is separate from the household earthquake insurance pool — the household scheme is run through Japan Earthquake Reinsurance and a government retrocession layer routed through the earthquake insurance public-private scheme. Overseas cat exposure includes US hurricane (Atlantic and Gulf), US wildfire, European windstorm, and Asian typhoon.

Reinsurance programs combine intra-group retrocession (US specialty platforms cede some retro back to Tokio Marine Reinsurance), proportional reinsurance for non-cat lines, and XL cover for cat lines purchased in the global market — counterparties include the major European reinsurers (Munich Re, Swiss Re, Hannover Re, SCOR), Bermuda reinsurers, Lloyd’s syndicates, and Japanese domestic reinsurance counterparties such as Toa Reinsurance.

Capital return policy at Tokio Marine Holdings is the most active among the big three: a progressive dividend plus opportunistic buybacks calibrated against ESR. Cross-shareholding wind-down has been a multi-year program and is a source of one-off realized gains plus a structural reduction in ESR equity-risk capital.

MS&AD Insurance Group Holdings — dual-domestic and mobility-tilted

MS&AD Insurance Group Holdings is structurally distinguished by its two-major-operating-company model: Mitsui Sumitomo Insurance (MSI) and Aioi Nissay Dowa Insurance (ADI). MSI carries the larger commercial / industrial book; ADI has a Toyota-dealer-network mobility insurance heritage and is the group’s mobility / connected-car ecosystem anchor.

On cat, MS&AD’s Japan-side exposure includes the same peril mix as Tokio Marine. Overseas, the group has built a Lloyd’s syndicate route (Lloyd’s structures and MS Amlin platform inherited from prior acquisitions), Asian non-life platforms, and US specialty exposure. The London-market / Lloyd’s footprint is one of the structural differentiators against Sompo.

Reinsurance programs follow a similar architecture to Tokio Marine: intra-group retrocession plus proportional and XL cover purchased externally. The dual-operating-company structure means group-level reinsurance optimization sits across both MSI and ADI and is reported at the holdings level.

Capital return is dividend plus periodic buybacks calibrated against ESR. Cross-shareholding wind-down is in progress and is a recurring capital-policy topic.

Sompo Holdings — domestic-core with overseas specialty

Sompo Holdings operates Sompo Japan Insurance as the domestic non-life core. The overseas portfolio combines US specialty (Endurance / Sompo International heritage), Europe, and Asia. The group has, in recent fiscal years, been managing remediation / governance disclosures arising from agency-channel and supplier-related issues; the disclosure arc has consequences for mid-term plan messaging and regulatory dialogue. See the agency / brokerage Japan landscape for the agency-channel context.

On cat, the Japan-side peril mix is the same as the other two groups; overseas cat exposure tracks the US specialty and Europe specialty platforms. Reinsurance program architecture is similar — intra-group retrocession plus external proportional and XL cover.

Capital return is dividend plus opportunistic buybacks calibrated against ESR. Cross-shareholding wind-down is in progress. The remediation arc has been a public-disclosure topic affecting capital-policy messaging in recent fiscal years.

Cat-peril, reinsurance, and overseas overlay

All three groups run a common reinsurance architecture and a common cat-peril mix on the Japan side, but the size and shape of each layer differ:

• Earthquake (commercial / shintaku property) — Distinct from the household earthquake insurance pool (earthquake insurance public-private scheme). Cat XL is the primary reinsurance protection. The biggest single-event scenarios drive the reinsurance-program retention sizing.
• Typhoon / wind — High-frequency loss driver. Recent years (Faxai 2019, Hagibis 2019, and subsequent events) reset reinsurance pricing across the program. Annual cat budget assumptions feed combined-ratio guidance.
• Flood — Secondary peril with increasing climate-attributed loss frequency. Cat XL and reinsurance reinstatement provisions are the main mitigants.
• Snow / hail — Smaller but volatile peril, especially for motor and agricultural lines.
• Overseas (US hurricane, US wildfire, European windstorm, Asian typhoon, cyber, casualty) — Subsidiary-level cat XL plus group-level retrocession.

The reinsurance counterparty set is concentrated among the major European reinsurers (Munich Re, Swiss Re, Hannover Re, SCOR), Bermuda reinsurers (Renaissance Re, Everest Re, AXIS), the Lloyd’s market, and Japanese reinsurance counterparties including Toa Reinsurance. The household earthquake retrocession layer routes through Japan Earthquake Reinsurance and a government layer documented in the earthquake insurance public-private scheme entry.

Combined-ratio and ESR sensitivity decomposition

Combined ratio (loss + expense / earned premium) is the headline underwriting profitability metric. For the big three, it decomposes by line — voluntary auto, fire (residential and commercial), casualty, marine, accident, and other commercial — and by region (domestic vs overseas). Cat losses appear as a one-line addition to the loss ratio when reported, with a separate “cat budget” assumption disclosed in mid-term plans.

ESR sensitivity for the big three is driven by:

• Domestic interest-rate shock — Smaller than for life insurers but still present because of reserve discounting and bond-portfolio mark-to-market.
• Equity-market shock — Meaningful because of the cross-shareholding portfolio; this sensitivity is mechanically falling as the wind-down progresses.
• FX shock — Material because of overseas earnings, foreign-denominated reserves, and foreign-currency-denominated reinsurance recoverables.
• Credit-spread widening — Bond portfolio mark-to-market and reinsurance counterparty exposure.
• Catastrophe shock — Direct cat-module charge under FSA economic-value standard; mitigated by reinsurance recoverables subject to counterparty quality.

See the global solvency framework comparison matrix for the regime-level decomposition and the ESR explainer for the metric definition.

Big comparison matrix table

The matrix below lists axes that differentiate the big three. Specific numbers (cat budgets, combined ratios, ESR, equity portfolio sizes) are date-specific and should be confirmed against each group’s current integrated / annual report and FSA materials.

Identity, group, and operating structure

Domestic cat exposure by peril

Reinsurance program shape

Overseas portfolio and specialty footprint

Equity holdings and cross-shareholding portfolio

Capital return policy and ESR

Combined-ratio decomposition (conceptual)

ESR sensitivity at a conceptual level

Recent M&A and strategic deals (conceptual)

Specialty-lines mapping

Historical and structural context

The current big-three shape reflects two decades of consolidation:

• Original lineage. The pre-consolidation Japan non-life industry had a much larger number of stand-alone insurers (Tokio, Nichido, Yasuda, Sumitomo, Mitsui, Taisei, Dowa, Aioi, Nippon Fire, Nissan Fire, Chiyoda, Koa, and others). The current big three are the consolidation products of merging most of these legacy carriers.
• Tokio Marine Holdings formation. Tokio Marine and Nichido Fire merged to form Tokio Marine & Nichido Fire (the domestic operating core under Tokio Marine Holdings).
• MS&AD formation. Mitsui Sumitomo, Aioi, and Nissay Dowa merged operations to form the current MS&AD structure with two domestic operating companies (MSI and ADI) under MS&AD Insurance Group Holdings.
• Sompo formation. Sompo Japan and Nipponkoa merged to form Sompo Japan Nipponkoa (now Sompo Japan Insurance) under Sompo Holdings.
• Overseas-acquisition wave. All three pursued overseas P&C acquisitions through the 2010s. Tokio Marine acquired Philadelphia Insurance, HCC, and Pure; MS&AD acquired MS Amlin (Lloyd’s-syndicate route); Sompo acquired Endurance Specialty and rebranded as Sompo International for the overseas specialty book.
• 2019 typhoon reset. Typhoon Faxai and Typhoon Hagibis in 2019 generated material domestic cat losses that reset reinsurance pricing and cat-budget assumptions across the big three for subsequent fiscal years.
• 2025 ESR rollout. The FSA economic-value-based solvency regime was rolled out from April 2025 onward, replacing the traditional solvency margin ratio as the headline domestic regulatory capital metric. The transition reset disclosure language. See economic-value-based solvency for the regime and ESR for the ratio.
• Cross-shareholding wind-down. All three are running long-tail equity-holding reduction programs over multiple years, with consequences for ESR market-risk capital, dividend-income line, and buyback capacity.

Reading the disclosure

Each group publishes the relevant data in similar but not identical disclosure formats:

Match the line-item naming carefully: “voluntary auto” versus “compulsory auto liability (CALI),” “fire” subdivided into residential / commercial, “casualty” with general / liability / E&O subdivisions, and “overseas” sub-segmented by region differ in label and detail across the three.

Decision use

Use this overlay when reading the non-life big three against each other rather than against the broader market. The matrix surfaces several practical analytical questions:

• Headline-ESR comparison alone is misleading. Two groups with the same ESR but different cat-retention layers and different overseas earnings shares carry different volatility. Pull each group’s ESR sensitivity disclosure rather than relying on the headline.
• Cat-budget interpretation needs the reinsurance program. A low cat budget paired with a high reinsurance retention is structurally different from a low cat budget paired with a low retention.
• Overseas combined ratio is a separate cycle. US specialty soft / hard markets and London casualty cycles move independently of Japan-side conditions. Reading group combined ratio without splitting domestic and overseas is misleading.
• Cross-shareholding wind-down has dual signal. Reducing the equity book lowers ESR market-risk capital (positive) and lowers dividend-income contribution (mixed). The pace and the buyer side matter.
• Remediation context affects messaging. Public-disclosure remediation arcs (notably the recent Sompo arc) affect mid-term-plan messaging, regulatory dialogue, and capital-policy guidance. Read alongside the agency / brokerage Japan landscape.
• MS&AD dual-operating-company model. Group-level numbers aggregate MSI and ADI; single-operating-company comparison against Tokio Marine & Nichido Fire or Sompo Japan requires segment-level reading.
• IAIG / ICS reporting. Group-level ICS reporting applies in parallel with domestic ESR for designated IAIGs — see Japan IAIG / ICS mapping and the global solvency framework comparison matrix.

Boundary cases / caveats

• Numbers are conceptual. This page is an overlay route. Cat budgets, ESR ratios, retention layers, combined ratios, overseas earnings shares, and equity portfolio sizes are date-specific and should be sourced from each group’s current integrated / annual report and FSA materials.
• Reinsurance is not zero-risk. Reinsurance recoverables shift loss risk to counterparties but introduce credit risk to those counterparties. Diversification across multiple reinsurers and use of collateralized structures matter.
• Cat budget is a planning assumption. The cat budget disclosed in mid-term plans is an underwriting-year planning number, not a guarantee. Actual cat experience can exceed or undershoot the budget materially in any single year.
• Earthquake household scheme is separate. The household earthquake insurance pool runs through Japan Earthquake Reinsurance and a government layer documented in the earthquake insurance public-private scheme entry. Commercial property earthquake is a separate private layer.
• Overseas combined ratios reflect non-Japan cycles. US specialty soft / hard markets, London casualty cycles, and overseas cat events drive the overseas combined ratio independently of Japan-side conditions.
• Cross-shareholding wind-down is sequenced. The pace at which long-tail equity holdings are reduced is constrained by counterparty / customer relationship considerations and is not purely capital-driven.
• Remediation arcs are public disclosure. Governance / agency-related remediation disclosures (most notably the recent Sompo remediation arc) affect mid-term-plan messaging and regulatory dialogue. See the agency / brokerage Japan landscape for the channel context.
• MS&AD dual-operating-company model. MSI and ADI are separate licensed operating companies under the holdings. Group-level reinsurance and capital optimization span both. Comparing big-three “single operating company” combined ratios requires care.
• IAIG group supervision. Group-level ICS reporting applies to designated IAIGs in parallel with domestic ESR — see Japan IAIG / ICS mapping. The headline domestic capital metric remains ESR.
• Specialty alternatives. The Lloyd’s Japan syndicate operating model and the captive insurance Japan market are alternative routes for specialty risk that interact with the big-three commercial book.
• Foreign reinsurer landscape. The foreign-reinsurer Japan landscape entry expands the counterparty set on the cession side, including Munich Re Japan, Swiss Re Japan, and Lloyd’s Japan.
• Underwriting-cycle backdrop. The Japan non-life underwriting cycle is the rate-cycle backdrop that determines commercial line pricing and combined-ratio direction independent of cat experience.

Related

• INDEX
• japan-nonlife-big-three
• natcat-reinsurance-japan
• earthquake-insurance-public-private-scheme
• foreign-reinsurer-japan-landscape
• japan-non-life-underwriting-cycle
• global-solvency-framework-comparison-matrix
• lloyds-japan-syndicate-operating-model
• marine-insurance-and-pi-cover-market
• captive-insurance-japan-market
• esr-economic-value-solvency
• japan-iaig-ics-mapping
• insurance-agency-and-brokerage-japan
• tokio-marine
• tokio-marine-nichido-fire
• tokio-marine-asset-management
• msad
• mitsui-sumitomo-insurance
• aioi-nissay-dowa-insurance
• sompo
• sompo-japan-insurance
• japan-earthquake-reinsurance
• toa-reinsurance
• munich-re-japan
• swiss-re-japan
• lloyd-japan
• insurance-license-and-solvency
• FinWiki index

Sources

• FSA: Economic-value-based solvency regulation hub.
• IAIS: Insurance Capital Standard activity / topic page.
• General Insurance Association of Japan: industry overview.
• Tokio Marine Holdings: integrated / annual reports.
• MS&AD Insurance Group Holdings: integrated reports and disclosure library.
• Sompo Holdings: integrated / annual reports.