Population shifts, not storms, have caused rise in property losses

CHAMPAIGN, Ill. -- Don't blame staggering property losses of the 1990s solely on changing weather or global warming. A review of 50 years of insurance and weather records puts the blame on the movement of an ever-growing U.S. population to the coasts and major urban areas.

Large property losses in 1990-1996 raised concerns that intense storms were occurring at a greater frequency than in the past. However, after looking at insurance records and storms in the United States, "we can see there is no evidence of an upward trend in catastrophic weather events," said Stanley Changnon, chief emeritus of the Illinois State Water Survey.

"Once you take loss data and consider such factors as demographics, population, the value of money, insurance coverage and overall wealth, you can readily see that the climate is not changing, we are," he said. "The increasing losses really are the result of an increased target-at-risk."

Such factors previously were not used in the loss equation. Without accounting for these factors, loss reports have helped create "the impression of an ever-increasing weather-loss situation with time," said Changnon, a professor of geography and atmospheric sciences at the University of Illinois.

In a recently published paper in the Bulletin of the American Meteorological Association, Changnon and his colleagues showed there had been no increases in thunderstorms, hailstorms, tornadoes or hurricanes. In fact, the frequency of these storms has decreased in the past 30 years.

For the study -- parts of which also were published recently in the journals Natural Hazards and Meteorological Applications -- past losses were adjusted to current dollars to re-examine storms and insurance losses between 1950-1989 and 1990-1996.

During the two periods, the distribution of extreme storms by region stayed the same, striking most in the South, Central, Southeast and Northeast, respectively. However, the growth of losses has been greater in the Southeast, South and West Coast. These major losses have followed the migration of people to the coasts and major urban areas. An increase in the number of storms qualifying as catastrophes in the mountains was linked to thunderstorm and hail damage in the rapidly growing Denver metropolitan area.

If global warming caused the rising losses, Changnon said, more extreme storms would be occurring, but that has not happened. In 1950-1996, there were 22 weather events exceeding $1 billion in insured losses. There were five in 1990-1996, but five also occurred in 1950-54 and five in 1960-65.

Changnon also looked at the intensity of catastrophes, dividing the number of $100 million storms by the number of annual catastrophes. "When you do this, you get an intensity value in which the 1990s don't stick out anymore," he said. "In fact, the 1990s fall below two other time periods, which center on 1950 and 1965."