Insurance Float: The Secret Weapon of Smart Investors

Insurance float is the single most powerful concept in the insurance business, and most investors completely ignore it. They look at premiums. They look at claims. They look at revenue growth. And they miss the real engine – the massive pool of money sitting between when premiums are collected and when claims are paid. That pool is float, and understanding it is the difference between seeing an insurance company as a boring utility and recognizing it as one of the most attractive business models ever invented.

If you want to understand why certain insurance companies compound wealth at rates that make tech investors jealous, float is where you start. Let me break it down in a way that actually makes sense.

What Exactly Is Insurance Float and Why Should You Care?

Float is dead simple in concept. You pay your car insurance premium in January. Maybe you file a claim in September. Maybe you never file one at all. Between the moment the insurance company collects your money and the moment they pay it out (if ever), that cash sits in their accounts. Multiply that by millions of policyholders and you get billions of dollars just sitting there, available to invest.

Here is the critical distinction: float is not debt. Nobody is charging interest on it. The insurance company did not borrow this money from a bank. Policyholders handed it over voluntarily, as part of a contract. The insurer holds it, invests it, and keeps every dollar of investment return.

In 2025, the largest insurance operations are sitting on float measured in hundreds of billions. Berkshire Hathaway alone carries over $170 billion in float. That is $170 billion of other people’s money generating investment returns for the company’s shareholders. No interest payments. No bond covenants. No maturity dates demanding repayment. Just capital, available for deployment.

Compare this to how a hedge fund operates. A hedge fund manager raises $1 billion, charges 2% management fees plus 20% of profits, and has to deal with investors who might pull their money at the worst possible time. An insurance company with $1 billion in float pays nothing for that capital – or in the best case, gets paid to hold it – and the money stays as long as policyholders keep renewing. Which model would you rather own?

The duration of float matters too. Life insurance and long-tail liability policies generate float that sticks around for decades. A workers’ compensation claim from 2025 might not be fully settled until 2040. That is fifteen years of investable capital from a single policy. Auto insurance float turns over faster, but the sheer volume of policies creates a massive, self-renewing pool.

How Does the Math Behind Profitable Float Actually Work?

This is where it gets quantitative, and where most analysts lose the plot.

The cost of float is determined by underwriting results. If an insurance company collects $100 million in premiums and pays out $95 million in claims and expenses, it made a $5 million underwriting profit. The cost of float in this case is negative – the company was paid to hold other people’s money. This is the holy grail.

The metric to watch is the combined ratio: total claims plus expenses divided by premiums earned. A combined ratio below 100% means the company is making money on underwriting alone. Every point below 100% is essentially a subsidy that makes the float cheaper.

• Combined ratio of 95% – the company earns 5 cents of underwriting profit per dollar of premium. Float is free and then some.
• Combined ratio of 100% – breakeven underwriting. Float costs nothing. All investment income is pure profit.
• Combined ratio of 105% – the company loses 5 cents per premium dollar on underwriting. Float has a cost, but if investment returns exceed that 5% drag, it is still a winning proposition.
• Combined ratio of 115%+ – now you are in trouble. The cost of float exceeds what most conservative investment strategies can earn. The business model is broken.

Here is a concrete example. Say an insurer holds $10 billion in float and runs a combined ratio of 97%. That means it earns roughly $300 million from underwriting alone. Now it invests that $10 billion. Even a conservative portfolio yielding 4.5% in today’s rate environment generates $450 million. Total economic benefit: $750 million, from capital that cost the company nothing. Try getting that deal from your bank.

Now flip it. Same $10 billion in float, but the combined ratio is 108%. The company is losing $800 million on underwriting. Investment income of $450 million only partially offsets that loss. The float is costing 3.5% annually – worse than just issuing corporate bonds. This is what happens when insurers chase market share in soft markets and write policies at prices that cannot support the underlying risk.

The key insight: float is only valuable when it is cheap or free. A company with $50 billion in float and a combined ratio of 112% is in worse shape than a company with $5 billion in float and a combined ratio of 93%. Absolute size of float means nothing without underwriting discipline.

Why Is Negative-Cost Float the Ultimate Competitive Advantage?

Negative-cost float – where the company gets paid to hold other people’s money – is the rarest and most powerful form of capital in all of business. It is better than equity (no dilution), better than debt (no interest), and better than retained earnings (no opportunity cost from prior profits). It is, in effect, capital with a negative price tag.

Very few companies achieve this consistently. The ones that do tend to share a few characteristics.

Conservative reserving. They set aside more than they think they will need for claims. This means fewer nasty surprises and more favorable reserve releases in later years. When a company consistently releases excess reserves, those dollars drop straight to the bottom line and reduce the effective cost of float retroactively.

Willingness to shrink. When pricing in the market gets irrational – competitors writing policies at combined ratios of 110%+ just to grab market share – the disciplined insurer walks away. They let premium volume drop 20%, 30%, even 40% rather than write money-losing business. This takes serious management courage, because Wall Street analysts will hammer you for “declining revenue” even when the decision is mathematically obvious.

Structural cost advantage. GEICO built its entire model on direct-to-consumer distribution, cutting out the agent middleman. That structural cost advantage meant they could price competitively and still maintain profitable underwriting. In 2025, companies leveraging telematics, AI-driven claims processing, and parametric insurance products are building similar structural edges. The technology does not change the fundamental model – it just makes it cheaper to operate.

Long-tail business selection. Companies that write reinsurance, workers’ compensation, and specialty liability tend to hold float for much longer periods than personal auto or homeowners’ insurers. Longer duration means more compounding time on invested float. A dollar of workers’ comp float held for twelve years generates far more investment income than a dollar of six-month auto float.

Progressive, for example, has maintained a combined ratio below 96% for most of the past decade while growing float steadily. That consistency is why the stock has compounded at over 20% annually. The market eventually recognizes that a growing pool of negative-cost capital is an extraordinary asset.

How Should You Evaluate Insurance Companies as Investments?

Most investors screen insurance stocks the same way they screen any other company – P/E ratio, revenue growth, maybe dividend yield. That approach misses the point entirely. Here is what actually matters.

Float per share and float growth. Is float growing faster than share count? A company that doubles its float over ten years while keeping share count flat has doubled the investable capital working for each shareholder. This is the insurance equivalent of a tech company scaling its user base without dilution.

Cost of float over full cycles. Do not look at one year. Look at the average combined ratio over 10-15 years, including catastrophe years. Every insurer looks great in a benign loss year. The question is what happens during a Category 5 hurricane season or a pandemic. Companies that maintain sub-100% combined ratios through the worst years have genuinely low-cost float. Companies that swing wildly between 90% and 120% have unpredictable float economics that make valuation unreliable.

Investment portfolio quality. What are they doing with the float? A company parking everything in short-term Treasuries is safe but leaving returns on the table. A company with 80% of float in speculative equities is gambling with policyholders’ money. Look for a conservative, diversified approach – mostly investment-grade bonds with a sensible allocation to equities and alternatives.

Balance sheet strength relative to exposures. This is the survival question. An insurance company with $7 billion of float and $40 billion of liquid assets can absorb a billion-dollar catastrophe loss and barely flinch – it represents maybe 2.5% of liquid assets. A company with the same float but only $8 billion in total assets is one bad year away from insolvency. In insurance, surprises are never symmetrical. They are always bad. The margin of safety in the balance sheet is not optional.

Management incentives. Does management talk about float growth and cost of float in their communications? Do they have compensation tied to underwriting profitability rather than premium volume? If the CEO’s bonus is based on growing revenue, they will write bad business to hit targets. If it is based on combined ratio and float growth, incentives are aligned with shareholders.

Here is a practical screening approach for 2025:

1. Start with combined ratio – trailing five-year average below 98%
2. Check float growth – positive trajectory over the past decade
3. Verify balance sheet – net worth at least 25% of total float
4. Read management commentary – do they use the word “float” and explain its cost?
5. Compare valuation – price to book value relative to peers with similar underwriting quality

Companies that pass all five filters are rare. That is precisely why they tend to outperform.

Key Takeaways

• Float is free investable capital. Insurance companies collect premiums upfront and pay claims later, creating a pool of money they can invest for their own benefit with no interest cost.
• The combined ratio determines float cost. Below 100% means the company is being paid to hold float. Above 100% means float has a cost that investment returns must overcome.
• Negative-cost float is the ultimate competitive advantage. Getting paid to hold other people’s money and then earning investment returns on it creates a compounding machine that is nearly impossible to replicate outside of insurance.
• Float size without underwriting discipline is worthless. A massive pool of expensive float is worse than a small pool of cheap float. Always check the cost before celebrating the volume.
• Evaluate insurance stocks on float economics, not revenue growth. Float per share, cost of float through full cycles, investment portfolio quality, and balance sheet strength are the metrics that matter.
• The best insurance companies walk away from bad deals. Willingness to shrink in soft markets and maintain underwriting discipline is the clearest signal of management quality in the insurance sector.

Insurance float is not a complicated concept. Collect money now, invest it, pay claims later. But the gap between understanding it intellectually and finding companies that execute it well – that is where the real edge lives. In a world where everyone is chasing the next growth story, the patient investor who understands float economics is playing a completely different game. And historically, it is a game that wins.