Only 53 percent of Baltimore's households returned their Census 2000 forms by the April 18 deadline, one of the worst response rates of any big city in the nation.
Census Bureau data released yesterday found that Baltimoreans' initial response to this year's census - by mail, telephone or the Internet - fell 10 percentage points from their 63 percent return rate in the 1990 Census. That was the sharpest drop among any of the country's 100 most populous cities.
Census officials had no definitive explanation. They suggested that growing immigrant populations and concerns about privacy may have contributed to the poor response in Baltimore.
But they expressed confidence that follow-up efforts in the spring and early summer to find and count those who did not initially return their census forms would produce a more complete and accurate final count for the city.
"We feel that the outcome will be a very good census," said Fernando E. Armstrong, the Census Bureau's regional director.
Rep. Elijah E. Cummings, a 7th District Democrat, could not account for the falloff in participation either.
"But I do know that the state and the city made a tremendous effort to try to get people to recognize why the census was important," he said. "I'm not as pessimistic as some may be. I think the numbers will improve greatly by the time the final numbers are presented to the president."
City leaders had hoped to do better in this census than in 1990. Instead, they did far worse.
Baltimore's 53 percent return rate tied the city with St Louis and Jersey City, N.J. Only New Orleans (52 percent) and Newark, N.J., (46 percent) fared worse.
Nationally, the return rate was 67 percent, up from 65 percent in 1990. That ended three decades of declining return rates and drew praise from Census Director Kenneth Prewitt. "Civic obligation, contrary to skeptical voices, is alive and well across America's communities," he said.
The statewide response in Maryland was 69 percent, down from 70 percent in 1990. It ranked 15th among the states.
Baltimoreans' failure in large numbers to mail, phone or email their census data to the Census Bureau does not guarantee the city will suffer a severe undercount in the 2000 census.
Federal enumerators were recruited locally and sent out to the addresses that were missing from the initial returns. Their job was to find and count the people living there.
But the low initial response made the follow-up bigger, more difficult and more expensive, Prewitt said.
That work ended in July. The results are now being combined with the initial returns. State population numbers will be sent to the president in December. The final local population data won't be released until April 1, 2001.
Baltimore Planning Director Charles Graves said the city's poor return rate can be explained in part on demolition. Several high-rise housing projects and 5,000 other dwellings have been razed since the city gave its list of addresses to the Census Bureau, he said. "Those units don't exist anymore. That probably contributed" to the low return rate, he said.
The final Census 2000 count is considered crucial to the city's political and financial future. Undercounting in 1990 has been blamed for the loss since then of $40 million in federal aid.
Low counts may also cost the city representation in Washington when congressional districts are reapportioned using the 2000 Census numbers.
Baltimore's fortunes in Annapolis could also be affected. With a population estimated last year at 632,680, the city has lost more than 12,000 residents a year since 1990, and it has been losing political clout to the state's growing suburban counties.
The Washington suburbs did far better than the city in getting their residents to answer the census. Montgomery County's response rate was up one point to 77 percent this year. Prince George's County was up two points to 68 percent.
The return rate in Baltimore County fell, from 76 percent in 1990 to 74 percent this year.
Baltimore's showing may also reflect poorly on Cheryl Benton, the former Democratic political strategist paid $32,500 by the city to ensure a more accurate count by enlisting the help of churches and service groups.