A recently released study of the costs of growth in Loudoun County shows that medium density development would produce the most cost-effective residential community, confirming what county planners say they have suspected all along.

"It showed that a compact form of urban development in the towns and eastern Loudoun is probably best, which is what we've been trying to achieve with our master plan," said Loudoun County planner Milton Herd. "It supports the basis for much of what is in the proposed rural plan."

The Loudoun Board of Supervisors is currently considering adoption of a management plan for the rural sections of the county, and commissioned the study to determine what fiscal effect different kinds of growth would have on the yet undeveloped parts of the county.

The study, which was done by the American Farmland Trust, looks at the costs of providing county services to four different types of residential development and compares those costs with the amount of revenue generated through taxes by each category.

It compared the costs and benefits of rural sprawl development (defined as one house every five acres); rural clusters (one house every acre); medium density (an even mix of two detached houses per acre with four town houses per acre); and high density (a mix of all housing types, but very heavily town house and condominium.) Land for public services, including parks, streets and schools, also was figured into the space allotment for each development category.

What the study found was that, on the basis of the facilities already available in Loudoun, the least-expensive type of future development for the county was medium density. All types of development were found to cost more in services than the county would reap in tax revenues, but medium density produced the least shortfall per person.

When compared against each other, the study found that rural sprawl would cost the county $685 per person; rural clusters would cost $567 per person; and high density would cost $216 per person. Medium density would cost the county only $206 per person.

While medium- and high-density developments were found to bring in less tax revenue than rural development, the study found that the cost of providing schools, roads, school transportation and water and sewer for rural developments was higher than for higher-density developments. The study found that rural sprawl would cost 43 percent more than a high density development, with rural clusters costing 30 percent more.

Of particular concern to Loudoun farmers and planners is the recent proliferation of small estates, 10 to 15 acre mini-farms. In the past few years thousands of acres of farmland in the county have been cut up into small, single-house lots. Herd said that current figures show that fully a third of the county has been subdivided into mini-farms, a trend that the county would like to try and reverse if possible.

With the support of a study documenting how costly such low-density development could be, said one county board member, support for the rural plan could increase. The proposed plan would limit the pace of subdivision in the county and would channel growth to the parts of the county with existing service networks.

The Loudoun study differs from other land-use cost studies, however, in that it does not take into account that some kinds of developments attract different kinds of households.

The major cost for all the suburban counties is providing schooling for children. In the past 20 years, planners across the United States have determined that some kinds of development attract younger families and thereby may create a heavier burden on school systems than other kinds of developments.

Herd said, however, that the county wanted only to compare what the fiscal impact would be if the same kind of family moved to the different types of developments.

"We realize that that puts some distortion on the results, but the county felt that it would be socially unacceptable to say we don't want schoolchildren in Loudoun County," said Herd. "We eliminated that purposefully, and I think people have kept in mind the distortion it creates."

The authors of the report also noted that it is commonly accepted that there are a number of economic, environmental and social costs that differ according to the type of development. They authors said they limited their quantitative analysis to the public economic aspects of residential growth.

"Our limited analysis should therefore not be read as an attempt to prescribe or promote any particular pattern of residential development, but only as an attempt to provide more specific information on some density-related public costs than has heretofore been available," said the report.

Loudoun's chosen approach differs substantially from land-use cost studies done in the past for both Prince George's and Fairfax Counties. An extensive study on the costs of growth for Prince George's in 1970 was used by the county council as one primary defense for changing the county's planning policies and redirecting growth.

"The study done by urban planner C. A. Doxiadis helped us identify that there were some kinds of developments that would be of more economic value to the county than others," said Francis B. Francois, a former Prince George's County Council member. "We learned a lot from the study and made a lot of changes which can now be seen in the kind of development in the county."

Francois said that the Doxiadis study found that garden apartment development would be costly in that it would attract many young families, and school children, without generating much in tax revenues.

"What we decided was that we didn't want Prince George's to be just a bedroom community," said Francois. "We wanted to encourage economic growth, so we set aside land for that purpose. If we hadn't done that, all those interchanges around the Beltway now developing as commercial centers wouldn't have been possible. There would have been garden apartments on all that land," if the market had been left to work on its own.

A more recent study done on Fairfax County in 1981, comparing the costs of rapid residential development there with the growth of Montgomery, Arlington and Prince George's counties, showed that school costs were higher in Fairfax than the other counties for the level of service provided and that because industrial development had not kept pace, the percentage of taxes contributed by businesses was lower in Fairfax than in the other counties.

Fairfax County Supervisor Audrey Moore, who commissioned the study, said, however, that she doubts the study led to significant changes in county policies toward continued residential growth.

"Everybody, from the developers to the board itself, tried to attack that study because they didn't like the message," said Moore. "What they did instead was extend to commercial development the same" encouragement that had always existed for residential.

Moore, and developers in the county, estimate that by the year 2000 there will be between 20,000 and 30,000 additional homes in Fairfax County. "Moderation is what Fairfax County has always needed but never had," she said.