<THE BORING PORTFOLIO>
GenRe and the Cost of Float
ALEXANDRIA, VA (August 18, 1999) -- On Monday, I wrote that GenRe's cost of float for the quarter is still attractive, at a cost of capital of 3.35% for the quarter (annualized). A couple of people took exception to that, pointing to an 111.8 % consolidated combined ratio. The difference of opinion comes down to whether you view combined ratio or after-tax underwriting loss as a percentage of float as a better interim indicator of cost of float.
The way I look at it -- and believe me, I'm still learning from people who work in the insurance industry -- the latter is a better indicator of interim underwriting results because of the significant timing differences in the recognition of premium revenues and the recognition of losses and expenses. A premium could be earned a number of quarters ago while associated loss reserves are still being adjusted. For instance, there could be significant timing differences in adjustment of reserves related to the $1.6 billion hail storm that hit Sydney, Australia in April and the premiums being earned on the associated contracts.
Essentially, underwriting losses and expenses recognized in an accounting period are not all ascribable to the premiums earned during that period, especially for an insurer writing a highly diversified book of insurance that includes catastrophe coverage and long-tailed lines of insurance. If you took one book of policies and never added new policies to it, then the combined ratio and the underwriting gain or loss as a percentage of float will converge over time and both will give you the long-term cost of capital.
For a company writing large catastrophe coverages, for instance, you're going to see quarters where the combined ratio is well below 100%. For all of those quarters where that's the case, you could see one quarter with a huge loss and a combined ratio significantly above 100%. No particular quarter is a good indication of the attractiveness of that float, but rather the combined ratio or the cost of float over the course of the policies' lives is a better indication of the attractiveness of the capital.
That's why I looked at the cost of GenRe's float in the way that I did, rather than using the consolidated combined ratio. In that light, the after-tax underwriting loss for GenRe divided by 66% of Berkshire's float of $22.8 billion results in an after-tax cost of capital of 3.35% on an annualized basis. Compared with cost of equity, where the expected return on equity is at least 11% before and after tax (because the cost of equity is not tax-deductible), float is a much more attractive source of capital.
Compared with almost any form of debt, it's also attractive. On just the base LIBOR rate, one-year debt costs 3.7% while longer-term debt costs more. So, no, I'm not doing handstands over the quarter's combined ratio at GenRe, but I look at it more along a continuum than on an immediate, static basis of an 111.8% combined ratio.
Overall, combined cost of float for the quarter was 2.33% on an annualized basis. Here's why this is a good thing. If you could have a genie grant you three wishes, you would want 1) interest rates to stay the same forever; 2) inflation to stay the same forever; and 3) unlimited opportunities to write insurance at a cost of 2.33% forever. If an insurance company is run well and can raise capital at that cost over the long term (with the expectation that inflation and other unwelcome conditions will come along on occasion), and that source of capital is combined with a rational investment approach, you have a nearly perfect business.
You don't have to discover the cure for cancer or invent the transistor to create substantial shareholder returns over the long run in such a business. If you have two sources of capital, float and equity, and the float you generate is equal to 50% of equity, then your cost of capital looks like this:
(0.3333 * 2.33%) + (0.6667 * 11%) = 8.1%
Unless a company is highly leveraged, almost no company operates at this cost of capital. If you can invest at a rate higher than this over the long term, you're going to create shareholder value. Of course, the investment track record at Berkshire is well known -- its returns on capital have dwarfed its cost of capital for decades. Under the above cost of capital, though, you don't have to be a superhuman investor to show economic income -- a topic I'll discuss on Friday.
Make a Living Foolin' Around.