Insurance Float Is Money You Hold Before You Have to Pay It Out

The Mechanism

An insurer collects premiums in advance and pays claims later sometimes much later. Between collection and payment it retains the money and invests it. That pool is called a float

The float is not profit and is not the insurer's money in any permanent sense. It represents obligations that will eventually be paid. But as long as it is held the investment income belongs to the insurer

The reason why it is worth understanding this as a financing structure and not as an insurance curiosity is that it reverses the usual sequence. Almost all other companies pay for their inputs before charging their clients and finance the gap. An insurer collects first and pays later so the gap runs in its favor and the financing comes from the clients themselves

Why Duration Matters

The length of the gap varies greatly by line of business and that variation drives the economy

Auto and property insurance settle quickly often in a matter of months so the float is short and turns over quickly. Certain lines of liability can take many years to resolve because a claim may not be reported for years and litigation further extends the timeline. Those long-tail lines generate float that can be reversed over long periods

Longer-duration flotation is more valuable and also more dangerous because the final cost is estimated for years before it is known

How Large the Pool Gets

The size of the float relative to the business is a function of duration and is compounded in a way that is easy to underestimate

A line that settles in three months has about a quarter of a year of premiums at any given time. A line that settles in four years has something like four years of premiums because each year's premiums are still on the balance sheet while those from previous years are settled. Therefore the same annual revenue produces a pool of an order of magnitude different depending only on how long it takes for claims to be resolved

This is why a long-tail insurer's balance sheet is so different from its bottom line. The premiums written in a year can be modest compared to the investment portfolio and the investment portfolio generates most of the visible profits

The Combined Ratio

Whether the float is a giveaway or a cheat is measured by the combined ratio which adds the loss ratio i.e. claims paid as a percentage of premiums to the expense ratio i.e. operating costs as a percentage of premiums

A combined ratio of less than 100 percent means that the insurer made money on the underwriting before any investment income. In that case the float was obtained at a negative cost that is the insurer charged a fee for holding funds that it would later release

A combined ratio greater than 100 percent means guaranteeing a loss of money and floating carries a real cost. It may still be worth it if the investment returns exceed that cost but the margin of safety is smaller and the model depends on the cooperation of the markets

The Cost of Float, Worked Through

Putting numbers on it makes the leverage visible. Take for example an insurer that writes 100% of the premiums on a long-tail line maintains a float of 200 and earns 4 percent on the investment portfolio

With a combined ratio of 97 underwriting brings in a profit of 3 and investments bring in 8 for a before-tax total of 11. The cost of the 200 float is a negative 1.5 percent. The insurer is paid 1.5 percent annually to hold the money it will eventually pay out while at the same time earning 4 percent on that money

With a combined ratio of 105 the subscription loses 5 against the same 8 of investment income so the total falls to 3. The cost of float is now positive 2.5 percent. The company continues to make money and it does so only because the return on investment exceeds the cost of financing which is precisely the position of a leveraged investor

The comparison worth keeping is that the second insurer runs an investment portfolio funded at 2.5 percent and the first runs a portfolio funded at less than nothing. Neither is unusual. The difference between them is entirely the underwriting and it's the entire business

Why Most Insurers Do Not Achieve It

Insurance is almost a commodity and competition is intense. When capital is abundant insurers compete on prices and premiums fall below what the expected claims justify. This is called a soft market

The discipline required to write less business when prices are inadequate is unusual because market share is visible immediately and underwriting losses appear years later when claims develop. Growth-compensated management faces an obvious temptation

Insurers that have built capital for decades are generally those that are willing to contract sharply during weak markets and expand aggressively after a catastrophe has removed capital from the industry and driven up prices

The Cycle Has a Mechanism

The soft and hard pattern of the market is not a state of mind. It follows from the fact that the industry's ability to do business is a function of its capital and its capital moves in large steps

A major loss event destroys capital throughout the industry at once. Capacity falls the supply of insurance contracts and prices rise sharply. High prices then generate strong returns which rebuild capital and attract new entrants and the added capacity causes prices to compete downward again until prices become inadequate again. The cycle ends when the next large loss wipes out the capital that the good years built up

What makes it persistent is the delay between writing a policy and knowing if it has been priced correctly. In most companies selling below cost quickly produces a bad number. Here it produces a good number quickly because the premium comes now and the claim comes in several years so the feedback that would correct the behavior is exactly what is missing during the period when the behavior occurs

Reinsurance Moves the Float As Well As the Risk

Insurers buy insurance themselves ceding a portion of their premiums to a reinsurer in exchange for the reinsurer accepting a portion of the claims. The usual description of this is risk transfer and it is also a float transfer

Ceded premiums are premiums that the primary insurer never retains so the fund available for investment is reduced along with the exposure. That's the trade. A company can manage a larger book of business than its capital would support by ceding part of it and what it gives up is both the technical benefit of the ceded share and the investment income from the float that would have accompanied it

The structure of the arrangement determines how much of each moves. A proportional treaty in which the reinsurer takes a fixed share of premiums and losses moves both in the same proportion. A deal that responds only above a threshold costs one premium and leaves most of the float in place as it is rarely paid out

The reason this belongs in an analysis and not a footnote is that the divestiture changes the meaning of the holder's figures. Gross premiums describe the business written and net premiums describe the business retained and a company that grows on a gross basis while giving up an increasing share is increasing its business without increasing the pool of money it has. Reading the growth from the wrong line produces the wrong conclusion about where the profits will come from

Where the Estimates Hide

The analytical difficulty is that claims not yet settled are estimates. Insurers set aside reserves for expected future claims and those reserves are management judgments

Insufficient reserves hurt current profits and create a problem that arises later as reserves strengthening. Watching whether an insurer's prior year reserves develop favorably or unfavorably over time is the best way to check whether reported profits have been real

It is worth understanding the mechanics of why this works. Reserves are established in the year a policy is written and are revised as information arrives. If the original estimate was too low later years must be added and that addition is charged to the subsequent year's earnings. Thus a company that has been under reserves reports good years followed by unexplained charges and the charges arrive labeled as an adjustment to prior periods rather than a loss on business written today

Insurers disclose reserve developments from previous years and reading several years in sequence is the closest thing to an audit of past honesty. Consistent favorable developments suggest conservative reserves. Consistent adverse developments suggest that earnings were being borrowed from the future and tends to be a habit rather than an accident

The Investment Side Has Its Own Constraint

The last part is that the free float cannot be invested however the insurer wants and the restriction comes from the liabilities it is subject to

The money has to be available when claims are due which advocates matching the duration of investments with the expected duration of claims. A portfolio of long-term assets versus short-tail liabilities may be forced to sell at a bad time. This is why insurance portfolios are dominated by bonds and not by whatever offers the highest expected return

Regulation reinforces this as insurers hold capital for both underwriting risk and investment risk and riskier assets require more capital reducing the return on that capital

There is also a correlation problem specific to this business. A major catastrophe generates claims and can coincide with falling asset prices so the insurer needs cash at a time when its portfolio is worth the least. Long-tail liability lines have a more subtle version of the same problem since the inflation that increases the ultimate cost of settling old claims is often the same inflation that depresses the bond portfolio held against them

What to Read First

All of the above boils down to a short list of things that are disclosed and are usually omitted in favor of the earnings figure

The combined ratio over a complete cycle rather than a year since any insurer can produce a good year if it does business cheaply and any insurer can produce a bad year if it is in the path of a storm. The trend and the average are the information

The previous year's reserves development table read over several years for the reasons above. This is the one that most often contradicts the narrative of the rest of the report

Premium growth compared to what's happening with prices. A company that grows faster than the market during a weak market is winning business based on price and the consequences of that are several years away

The expense ratio is separated from the loss ratio because the two components of the combined ratio behave completely differently. Expenses are under management's control and losses are not so a company that improves its combined ratio through expenses has done something lasting and another that improves it through a year of benign claims has not

And the composition of the investment portfolio versus the duration of the liabilities since that is where a conservative-looking insurer can assume the risk that it does not assume on the underwriting side

The Bottom Line

Float is investable money held against future claims and its value depends entirely on underwriting discipline. Check the combined ratio and reserve development of the previous year before believing in the profits of any insurer. Pooling is a form of borrowing whose interest rate is the result of underwriting which is why the same business can be extraordinarily good or quietly mediocre with an investment portfolio that looks identical from the outside