Berkshire Annual Letter 1997 (Part 4)
Unless you understand about “float” and how to measure its cost, you will never be able to make a good estimate about Berkshire’s intrinsic value.
Float is money that is held but not owned.
In the insurance industry, there is float because premiums are received before losses are paid. This time interval can sometimes extend to decades. During this time, the money can be invested by the insurer.
The downside of it is that the premiums are usually not sufficient to cover the expenses and insurance coverage that the insurer must eventually pay. This is the “underwriting loss”, which is the cost of the float.
An insurance business will have value if its cost of float over time is less than the cost of obtaining the funds from alternative means. Otherwise, it is a lemon.
(Interesting, for certain national plans in my country, sometimes they like to cite an underwriting loss as a good enough reason to increase insurance premiums. Now I know otherwise! An underwriting loss can be the norm.)
It is not so straightfoward for investors to value an insurance company. Insurance losses have to be estimated (these errors can be huge) so investors can’t really calculate a company’s true cost of float.
For Berkshire, their insurance business has been a huge winner. In the last 5 years, their cost of float has been negative. This means they have been paid for borrowing the money.
Furthermore, the amount of float that they have been able to generate has grown at a compounded rate of 21.7%. That means more money to invest!
The accounting irony of all these is that float is shown on the balance sheet as a liability. In actual fact, it has a value to Berkshire greater than an equal amount of networth would have had.