Baltimore stands to lose more government aid than most large cities because of the census undercount of blacks, according to a Morgan State University study to be released today.
Baltimore could lose up to $39 million a year, said Robert B. Hill, the author of the study. But other experts said that figure might grossly overstate the undercount's potential impact on the city.
"People have done this kind of simplistic analysis, sometimes for shock value," said Peter A. Bounpane, assistant director of the Census Bureau.
He added, "It doesn't sound right to me."
William P. O'Hare, a University of Louisville census expert, said: "Figuring out a precise number is not impossible, but I don't know of anybody who has the time and resources to do it. I've never seen a real good study of how many dollars would be lost."
Dr. Hill conceded that the $39 million figure "was certainly an upper limit."
"My emphasis is not so much the number per se. It is more the ordinal relationship it has to other localities," said Dr. Hill, director of Morgan's Institute for Urban Research.
The study projected that Baltimore would lose more dollars in aid than larger cities such as Los Angeles, Chicago, Houston, Philadelphia and Detroit, but less than New York and Washington.
The Census Bureau has estimated that Baltimore's population was undercounted by 4.7 percent in the 1990 census, mainly because of the disproportionate number of blacks missed in the head count.
An adjustment of the count would increase the city's population from 736,014 to 772,000.
U.S. Commerce Secretary Robert A. Mosbacher must decide by July 15 whether to adjust the count. The final result will affect localities' political representation and their share of government aid that is based on population.
Dr. Hill said he supported adjusting the census and that he was "trying to time the report to influence" the secretary of commerce's decision.
"But I'm not presumptuous enough to think it will," he said.
Dr. Hill arrived at his $39 million figure by estimating government aid to Baltimore at $1,080 a person (based on 1988 figures) and multiplying that by 36,000, the city's estimated undercount.
"It's not that straightforward," Mr. Bounpane said. "I think people are deluding themselves a little bit if they think there will be big changes."
Dr. O'Hare said the formulas that govern how much aid localities receive "typically are very complex. They have grandfather clauses, floors, ceilings, minimums and maximums. And keep in mind, some programs give you money if you're losing population."
He added that "places that have the biggest undercount would probably be hurt the most."
Baltimore's estimated undercount was one of the highest percentages in the nation.
Mr. Bounpane and other experts said cities should not expect a windfall for two reasons:
* Most federal aid is doled out with no regard to population. When population is considered, it is usually only one of several factors.
* Federal and state grants generally come from a fixed pie of funds. If one locality gets more, another gets less. Baltimore would get a bigger slice of the pie only if its population was adjusted higher than other places.
Maryland could lose federal aid if the population count is adjusted because its estimated undercount of 1.8 percent was less than the national average of 2.1 percent.
Dr. Hill said the aim of his study was "to at least encourage some type of dialogue."
"The point is that even if the figure that you get is a half or a third of this, it is clear that places with the highest racial undercount certainly have the greatest need for these resources," he said.