Why There Is No Single Global Insurance Company

There is no worldwide insurance monopoly, and the reason is not regulatory accident or insufficient ambition. Every path to one is blocked by a different mechanism — and each blockage is itself the explanation of how insurance works. The intuition is compelling: pool everyone, spread the risk maximally, cut out redundant overhead, reinvest the float, and premiums should collapse toward nothing. Following it produces a tour of exactly the constraints that shape the industry. ## Each escape hatch, and the wall it hits **"More members make it cheaper."** Pooling reduces the *uncertainty* of the average, not the cost. Every house brings its own expected claims in alongside its premium, and the benefit of scale decays as 1/√n — effectively exhausted at national scale. There is no mathematical force pushing toward global. → Risk Pooling Reduces Uncertainty, Not Cost **"Cut out the profit margin."** In the best underwriting year in over a decade, the industry ran a combined ratio around 96.5, and homeowners specifically at 99.7 — essentially zero underwriting profit. About 97% of the premium is the cost of risk and operations. A perfectly benevolent operator saves single-digit percentages. → Combined Ratio: Where an Insurance Premium Actually Goes **"A global pool is maximally diversified."** It is the exact opposite. Diversification requires independence, and one global balance sheet is maximally exposed to every global catastrophe simultaneously — the most fragile structure that could be built, and one no regulator would permit. → Correlated Risk: The Failure Mode That Actually Kills Insurers **"Invest the float and hand back the returns."** This is Berkshire Hathaway's model, run for sixty years by the best capital allocator of the era. It produced an enormous investment company, not free insurance: regulators require the capital to be held as reserves, claims inflate alongside returns, and home insurance generates little float. → Insurance Float: The Berkshire Hathaway Model and Its Limits **"Charge everyone the same flat rate."** This is the first thing competition destroys. A competitor cherry-picks the low risks, leaving a pool that costs more than it collects, and the spiral runs to insolvency. → Risk-Based Pricing and the Adverse Selection Spiral **"Run it as a member-owned non-profit."** Already exists, at scale, for over a century. Mutuals land 10–30% cheaper with an occasional dividend. → Mutual Insurance and the Assessable Mutual **"Pay only when someone actually has a loss."** The assessable mutual — historically how insurance began, and abandoned for good reasons: assessments cannot be collected after the fact, clustered losses produce crushing bills, and the real per-household arithmetic is an order of magnitude above the intuition. **"Prevent the losses instead."** The best idea in the set, and the only lever that lowers the cost floor — but bounded, since inspection costs approach the home premium and many perils give no warning. → Loss Prevention and Risk Engineering: The Only Lever That Lowers the Cost Floor **"Then nationalise it."** Governments already have, precisely where private markets fail — flood, terrorism, deposits, health, pensions. Where they nationalised working competitive lines, the recurring failure is political underpricing funded by debt. → Public Insurance: Where Governments Already Nationalised Risk ## The thing that is already global The worldwide pool exists. It is the **reinsurance layer** — wholesale, invisible to consumers, and deliberately several firms rather than one, because its entire function is to spread catastrophe across independent balance sheets. → Reinsurance: The Global Risk Layer Consumers Never See ## The single sentence The premium is chained to the physical expected loss of the insured thing — a real, recurring cost that does not shrink with pool size, cannot be invested away, and is already almost the whole price. You can pay it earlier, smoother, slightly smaller, or with someone else's money. You cannot make houses stop needing repairs. ## The generalisable shape This is a clean instance of a recurring pattern: a simple single-lever intuition fails because the system contains feedback loops that activate the moment the lever is pulled. Here they are adverse selection, competition compressing margin to cost, catastrophe correlation, and capital regulation — four independent mechanisms, each sufficient on its own. The intuition is not foolish; it is one that has been tested repeatedly by serious people, and the industry's structure is the accumulated record of what happened each time.