State's household income takes hit in survey

Decline, worst in nation, doubted by state economic analyst

SCOTT WYLAND, THE OLYMPIAN

Originally published September 26

OLYMPIA -- A new Census survey suggests that Washington state's median household income fell by 8.2 percent during the last two years of the booming 1990s.

But the survey has found at least one detractor: a state economic analyst who challenges the research methods.

The survey says Washington, Louisiana and Alabama had a decline in median household income between 1998 and 2000; of the three, Washington's showed the steepest drop.

"That's some tough company to keep," said William Dillingham, senior economist for the state Employment Security department.

Washington households had a median income of $42,024 in 2000, on par with the national median of $42,148.

Experts offered several theories to explain Washington's poor showing: It's the tech economy, it's the manufacturing economy, it's a mystery.

One expert dismissed the survey altogether.

"You're not seeing an actual fact; you're seeing an estimate based on a random sampling," said Bret Bertolin, senior analyst for the state Office of the Forecast Council. "It is possible this is not right."

The state's economy was robust during this period and the average wage went up substantially, putting the Census Bureau's methodology into doubt, Bertolin said.

The average wage in Washington went up by 7.9 percent in 1999 and 7.4 percent in 2000, Bertolin said. When adjusted for inflation, that adds up to more than a 10 percent rise in that period, he said.

If the median income drops, the average income isn't likely to increase, Bertolin said. "It's hard to reconcile those," he said.

The state determines the average wage using data that all Washington employers are required to submit each year, Bertolin said.

This information is far more complete and reliable than the partial Census sampling, which according to the survey's authors has a 1 in 10 chance of being wrong, he said.

Median vs. average

Median income should also be clarified, he said.

It's a dividing line. It means half the population earns income above that margin, and half earn below it, Bertolin said.

That's different than the average, in which everyone's income is added together then divided by the number of earners.

Dillingham said the bursting of the technology stock bubble let the air out of Washington's household income.

"That's the downside of such a remarkable half-decade," Dillingham said.

Exercised stock options boosted some workers' income, he said -- until the stock market took a dive.

Dillingham compared it to a tidal wave of wages surging onto our shores. "That tsunami of wages was stock options. Now we're back to normal tidal action," he said.

Boeing's impact

Other experts said the Washington data tell the story of the quiet downsizing that's been happening behind the technology boom.

"Reduced manufacturing employment appears to be behind the drop in median income," said Christopher Haugen, policy analyst with the University of Washington's Daniel J. Evans School of Public Affairs.

The Boeing Co. announced last week that it planned to cut as many as 30,000 jobs in its commercial airplane division, after the aviation industry was staggered by the Sept. 11 attacks.

But the work force has been shrinking all along, Haugen notes: Boeing has cut 20,000 jobs since 1997.

The poverty rate for the state's residents remained nearly steady from 1998 to 2000, rising 0.4 percent to 9.7 percent, according to the survey. That suggests it's not the poor who are losing income.

"It appears that middle-income families were not doing as well as we thought," Haugen said.

Differing view

But Bertolin, of the Forecaster Council, argues that a decline in median income is much harder to explain when the economy is vibrant.

Nothing short of massive layoffs would cause the median household income to dip as much as this survey describes, Bertolin said.

Boeing's job cuts in the past several years wouldn't come close, he said.

What's more, deflated stock options wouldn't necessarily affect the median, he said.

If a slew of investors who were pulling in $10 million a year from stocks suffered a $9 million loss in dividends, they would still be above the median -- so the median would not change, he said.

The same could be said for people earning below the median, Bertolin said. If workers with a yearly income of $30,000 were laid off, the median would be unaffected, he said.

For the median to change, people earning at or above the dividing line would have to lose enough income to drop below it, Bertolin said.

People should be careful not to mistake this study with the official 2000 Census, he said. The true Census numbers on household income will be out in eight or nine months.

"When the Census data comes out, we'll see who's right," Bertolin said.

Scott Wyland is a business reporter for The Olympian. He can be reached at 360-357-0748 or scottolympian@yahoo.com. The Associated Press contributed to this report.

The Olympian Copyright 2001

back to main Census 2000 index