With all the predictions of a real estate bubble, it's sometimes forgotten that there's no nationwide real estate market. What happens in Pomona tells us nothing about real estate trends in Poughkeepsie. At any point some places will do better than others, some worse and the great trick is to find to which is which.

One of the best measures of local real estate demand is nothing more complex than population and construction trends. A growing population combined with an expanding housing stock generally suggests firm prices. A growing population with an insufficient housing stock means home prices are likely to soar. Less population suggests reduced demand and a softer local market.

Given this background, it's not surprising that so much attention is given to the Census Bureau's listing of the 100 fastest-growing counties in the U.S. If you believe that demand counts, then all things being equal population is an important predictor of future real estate values and growing communities implicitly suggest more household wealth.

But not all things are equal and communities do not always grow. The Census report reflects data that ended July 1st, 2005 and even though last summer was not too long ago, the world has become a very different place.

Hurricanes Katrina, Rita and Wilma devastated large areas along the Gulf Coast. There's no question that massive population declines will be shown in these areas with the next Census report if only because much of the housing stock was destroyed and the job base demolished. The result is that one will simultaneously expect to see reports of rising prices for those homes in good shape (because they're scarce, in demand and countable) and massive numbers of unreported cases where homes are unsold, unsalable and uninhabitable.

Unfortunately, it's possible that the situation in the gulf region may actually get worse. We now have large numbers of people living in trailers, the very form of housing least able to withstand heavy storms.

With a new hurricane season set to begin in several months, households that might easily have survived a typical storm season may well be less secure than a year ago.

Watching all of this with great care are the companies that provide homeowner insurance. If you're an insurance company you like to have an investment portfolio that grows, few claims and premiums that more than pay the freight. Unfortunately, changing weather patterns are causing insurance companies to raise premiums.

Amazingly, that's the good news.

The bad news is that in some cases insurance companies will not provide coverage in given areas at any price. The Wall Street Journal reports , for example, that coverage on Cape Cod is being refused by several insurance companies even though the area has not been hit by a hurricane in 50 years. The paper also says that 28,000 policyholders in eight counties in New York state are losing coverage.

The loss of insurance coverage is a problem for three reasons.

First, you want homeowner's coverage because it provides a number of important protections.

Second, homeowner's coverage is typically a contractual requirement: If you have a mortgage you must carry fire, theft and liability insurance to protect the lender's security for the loan -- your house. Fail to maintain such coverage and you will get an unpleasant letter from your lender discussing reasons why it may be necessary to call your loan in the next 30 days or so. The solution in most areas is to keep private-sector coverage or join a state-provided insurance pool where premiums tend to be high and coverage tends to be low.

Third, if homeowner's insurance is tough or impossible to get, financing is impacted and home values can weaken. Even counties in the top 100 can be impacted.

It's easy to understand why insurance companies would not want to cover a beach-front palace in the face of dire hurricane predictions. But rates are being raised far from ocean-front lots, which means that homeowners who represent little risk are effectively subsidizing those who live along the coasts.

Insurance companies are right to raise rates in the face of increasing pay-outs. The theory is simple: Those who represent the most objective risk should pay the highest rates. But insurance companies are not right to raise rates when increases are not directly related to local losses or because of poor portfolio performance.

The time has come to look carefully at insurance company economics. What is the relationship between premiums and payments? As New York Attorney General Elliot Spitzer has shown, overcharges do occur: In one New York case announced last December, 10,000 overcharged state residents will get back $400 each.

We need more Spitzers and we need more states to see what's really going on with insurance coverage -- while insurance coverage is still available.

For more articles by Peter G. Miller, please press here .