One of my readers, Steve Milos, asked me the following question:

Free cash flow is a metric that I like to use when judging investments in most types of companies. However, I?m not sure how to apply it to insurance companies, or even how to calculate it, given the uncertainty of claims. Do you use it? How do you calculate it? Currently, I?ve used P/E, P/Book, dividend yield, combined ratio as metrics for insurance companies instead.

Before I start, for those that have access to RealMoney, I would refer you to the following two articles:

Parsing the Financials of the Financials, and

Time to Get Personal with Insurers (free at TSCM).

Quoting from a recent article at RealMoney:

The free cash flow of a business is not the same as its earnings. Free cash flow is the amount of money that can be removed from a company at the end of an accounting period and still leave it as capable of generating profits as it was at the beginning of the accounting period. Sometimes this is approximated by cash flow from operations less maintenance capital expenditures, but maintenance capex is not a disclosed item, and changes in working capital can reflect a need to invest in inventories in order to grow the business, not merely maintain it.

And quoting from the first article that I cited above: The cash-flow statement is of great use in gauging the health of industrials and utilities, but it tells us next to nothing about financials. One of the best values of cash-flow statements is that they enable one to attempt to derive estimates of free cash flow (the amount of cash that a business generates in a year that is left over after it has paid all of its expenses, including capital expenditures to maintain its existing business). Deducting maintenance capital expenditure from EBITDA often approximates free cash flow.

However, cash-flow statements for financials can’t in general be used to derive estimates of free cash flow, because when new business is written, it requires capital to be set aside against risks. Capital is released as business matures. In order to derive a free cash-flow number for a financial company, operating earnings would have to be adjusted by the change in required capital.Sadly, the change in required capital isn’t disclosed anywhere in a typical 10K. Even the concept of required capital can change depending on what market the financial institution is in and what entity most closely controls the amount of operating and financial leverage it is allowed to take on.

Federal or state regulators sometimes impose the biggest constraints on leverage — this is particularly true for institutions that interact closely with the public, i.e., depository institutions and life and personal-lines insurers. For companies that raise capital in the debt markets or do business that requires a strong claims-paying-ability rating, the ratings agencies may lay on the tightest constraints.

Finally, in rare instances of loose regulatory structures, the tightest constraint can be the company’s calculation of how far it can push its leverage before it blows up. Again, this is rare; many companies estimate the capital required for business, but regulatory or rating agency standards are usually tighter.

Actuaries have their own name for free cash flow. They call it distributable earnings. It is equal to earnings less the change in capital necessary to support the business. When sales are growing, typically distributable earnings are less than earnings. When sales are shrinking, typically distributable earnings are more than earnings.

As pointed out in the second quotation above, the hard part is knowing what entity requires the most capital to be held against the business. Is it the regulators, the rating agencies, or the company itself? The tightest constraint determines the capital that is required.

The second part of what makes it hard is that the capital standards for the rating agencies are dimly known to outsiders. Internal company capital standards are not known to outsiders either. Finally, the regulatory standards for capital are known but complex. The formula is known, but not all of the data that goes into the calculation is public. A further difficulty is that different companies run at different percentages above the minimum capital standards, and typically, that is not disclosed.

Aside from that, there is the problem of whether the reserves are fairly stated, but the nuances of that are beyond this discussion. What can an insurance analyst do to get something near free cash flow?

Ask questions on leverage policy. Ask the company how they decide what the maximium amount of business they could write next year is. Premiums-to-surplus? Statutory net income/loss limits? How much more could you borrow at the holding company at your current rating? Questions like this cut through the clutter of what you don’t know, and allow you to estimate how much capital they will have available to increase dividends, buy back stock, or buy other strategic assets. You can also read reports from the ratings agencies, since they focus on this.

In practice, I have a “back of the envelope” feel for how loose or tight capital is at firms that I analyze. I spend more time on pricing power, since it correlates better with stock price performance in normal situations. I look for the sustainability of underwriting margins. I also graph Price-to-Book versus Return on Equity, looking for companies that earn a lot on their net worth, and have a reasonable chance of sustaining those earnings.

I hope that helps explain how to analyze insurance companies, approximating some aspects of free cash flow. If you have a question, pose it below so others can benefit from your question and my answer.