Munich Re and the Business of Pricing Catastrophe

Reinsurance is the only industry whose product is the ability to lose an enormous amount of money in a single year without failing. Munich Re is among the oldest and largest participants, and its discipline through pricing cycles is the clearest available lesson in underwriting through a market that periodically forgets what risk costs. This case study closes the banking and finance pillar of the Germany Company Stories hub.

How does the reinsurance cycle actually work?

Through capital rather than through risk. When a major catastrophe consumes industry capital, supply contracts and prices rise sharply. High prices attract new capital, including from investors buying catastrophe bonds, which expands supply and pushes prices down until the next major loss.

The critical point is that price movements are driven far more by capital availability than by changes in underlying hazard. A quiet loss year does not mean risk has fallen; it means capital has accumulated.

Discipline in this environment means declining business when prices fall below the level the risk requires, which reduces revenue and invites criticism for losing market share. Firms that grow through soft markets almost always pay for it in the next hard one.

The measurable expression of discipline is the combined ratio across a full cycle rather than in any single year. A reinsurer that beats its peers in soft years is usually taking risk it has not been paid for.

Why are secondary perils such a problem?

Because they are frequent, geographically diffuse and poorly modelled. Wildfire, flood, hail and severe convective storm produce moderate individual losses that aggregate into very large annual totals, and the models for them are far less mature than those for hurricane and earthquake.

Peak perils are modelled well because they are rare, well documented and physically well understood. A hurricane model can be validated against a century of landfall data. A wildfire model must account for vegetation, land use, building codes, ignition sources and suppression capacity, all of which change faster than the historical record can capture.

Exposure growth compounds the problem. Losses rise partly because more property of higher value is located in hazard-exposed areas, which means loss growth reflects development patterns as much as climate.

The practical consequence for reinsurers is that annual aggregate covers, which pay when accumulated smaller losses exceed a threshold, have performed poorly, and many have been repriced or withdrawn. The industry has responded by raising attachment points so that primary insurers retain more of the frequent losses.

Is climate risk still insurable?

Annually, yes. Property catastrophe cover is written for one year at a time, which means prices reset every renewal and no reinsurer is committed to a long-term view of climate trends.

That annual repricing is the industry's core defence, and it is also why the insurability question is genuinely about affordability rather than about availability. Cover exists at a price; the question is whether the price exceeds what property owners will pay.

In several exposed markets that threshold has been reached, with primary insurers withdrawing from specific regions and state-backed schemes expanding to fill the gap. That transfers risk from private balance sheets to public ones without reducing the underlying hazard.

The systemic issue is that risk-based pricing works as a signal only if people can act on it. Where insurance prices correctly reflect hazard and development continues anyway because coverage is subsidised, the signal is suppressed and exposure keeps growing.

What makes cyber risk different from natural catastrophe?

Correlation and non-stationarity. A hurricane affects a defined geography; a vulnerability in widely deployed software can affect every insured simultaneously regardless of location, which breaks the diversification that makes insurance work.

The second difference is that the hazard is adversarial. Natural perils do not adapt to defensive measures, while attackers do, which means historical loss data has limited predictive value for future frequency and severity.

Reinsurers have responded with explicit war and state-sponsored attack exclusions, event definitions limiting aggregation, and conservative capacity deployment. Those measures are necessary and they narrow the product substantially, which frustrates buyers who wanted broad cover.

The growth opportunity is nonetheless real, because demand vastly exceeds supply and pricing reflects genuine uncertainty. The firms likely to succeed are those that treat cyber as a modelling and data problem rather than as a premium growth opportunity, which is the same discipline that separates results in the asset management analysis.

How does a reinsurer make money in a low-loss year?

From underwriting margin and from investment income on the float, and the second is more important than most observers assume. A reinsurer holds premiums for years before paying claims on long-tail business, and that portfolio generates returns that can rival underwriting profit.

That makes interest rates a major driver of reinsurance economics. A period of higher rates raises investment returns on both new premium and reinvested maturities, which improves earnings independently of underwriting conditions.

It also creates a subtle discipline risk. When investment income is strong, the temptation is to accept thinner underwriting margins because total return still looks acceptable, which is precisely how soft markets are financed.

The reinsurers with the best long-run records generally separate the two explicitly, requiring underwriting to stand on its own terms and treating investment income as a separate result. That sounds obvious and it is rarer in practice than it should be.

What should a corporate risk manager take from this?

That insurance pricing tells you something real about your risk, and that it is worth listening to. When a reinsurer raises attachment points on a peril, it is expressing a data-based view that frequency has changed, and that view is usually earlier than internal risk assessments.

The second practical point is retention strategy. In a market shifting risk back to buyers, the question is not how much cover to buy but which layer to retain. Retaining frequent, moderate losses and insuring the tail is usually cheaper than the reverse, and it aligns with how the market is now structured.

Third, examine aggregation across your own portfolio the way a reinsurer would. Concentration of sites, suppliers or customers in one hazard zone is the corporate equivalent of a correlated book, and it is frequently invisible until an event reveals it.

Finally, treat insurability as a location factor in capital investment decisions. A site that cannot be insured economically has a higher cost of capital and a lower resale value, and those effects arrive well before any physical loss does.

What is a catastrophe bond and why does it matter?

A security whose principal is forgiven if a defined catastrophe occurs, allowing capital market investors to take insurance risk directly. It matters because it changed the supply dynamics of the reinsurance cycle permanently.

Before these instruments, reinsurance capacity was limited to reinsurers' own capital, so hard markets persisted until retained earnings rebuilt. Now capital can enter within months of a loss event through new issuance, which shortens hard markets and caps price increases.

For investors the appeal is diversification: catastrophe risk is largely uncorrelated with equity and credit markets, which is unusual and valuable in a portfolio.

The structural consequence for reinsurers is that they compete against capital with a lower required return, and their defence is in complexity. Standardised peak-peril risk migrates to the capital markets; bespoke, multi-peril and casualty risk requiring underwriting judgement stays with traditional reinsurers.

How do reinsurers actually manage accumulation risk?

By modelling the whole portfolio against defined event scenarios rather than by assessing contracts individually. The relevant question is never whether a single treaty is priced correctly, but what the entire book loses if a specific event occurs.

That requires knowing where every insured exposure sits geographically, which is harder than it sounds because a reinsurer sees primary insurers' portfolios in aggregate rather than location by location. Data quality on exposure is the binding constraint on the whole discipline.

Retrocession, reinsurance purchased by reinsurers, provides the final layer, transferring peak exposures to other carriers and to capital markets. The cost and availability of retrocession is itself cyclical and tends to be scarcest precisely when it is most wanted.

The transferable idea for any business is the distinction between contract-level and portfolio-level risk. A set of individually sound exposures can still be collectively fatal if they fail together, and only a scenario view reveals it.

What role does reinsurance play in the wider economy?

It determines what can be built and financed. A property development that cannot obtain insurance cannot obtain a mortgage, and an infrastructure project without construction and operational cover cannot reach financial close.

That makes reinsurance pricing a quiet but powerful allocator of capital across geographies. As cover becomes expensive or unavailable in hazard-exposed regions, investment redirects, which is a market response to physical risk operating faster than any policy mechanism.

For corporate planners the implication is that insurability should be tested early in site selection, alongside energy cost and logistics, and treated as a permanent locational factor rather than an administrative detail settled after the decision.

Why does reinsurance concentrate in a few cities?

Because the business depends on a dense cluster of underwriting, actuarial, modelling and brokerage expertise that is difficult to assemble anywhere else. Munich, Zurich, London and Bermuda each host that concentration.

The cluster effect resembles the industrial pattern described in the hidden champions analysis: specialist knowledge circulates informally between firms, talent moves within the cluster, and the aggregate capability exceeds what any single employer could build.

That is also why these positions are durable. A jurisdiction can offer favourable regulation and tax treatment and still fail to attract the business, because the constraint is people rather than policy.

Frequently Asked Questions