A sharp decline in low-income tenants' earnings over the last decade, coupled with inflation and rising rents, has led to serious deterioration in the nation's supply of rental housing, according to an expert.

"A person with $10,000 is not a person who has enough income . . . to sustain a rental unit at the level of maintenance" needed to keep the unit in a condition good enough to keep it in the housing stock for a long period, said William C. Apgar Jr., of the Harvard-Massachusetts Institute of Technology Joint Center for Housing Studies.

The number of households with incomes under $10,000 rose from 8.4 million to 11.9 million between 1974 and 1984 as the income of all renters in the United States dropped by 17 percent during the same period, he said. These figures, reported by the Census Bureau, include welfare payments and other subsidies.

When a tenant can pay only $150 to $200 a month in rent, he "can't pay enough to the landlord to heat the building, pay the property taxes, pay the mortgage and have anything left over for maintenance," Apgar said.

The result has been a downward slide in the quality of rental property. Federal and private counts of the low-income housing stock show that the number of units has remained about the same, but the statistics are misleading because they do not reveal what is really happening, according to Apgar. Many units "are dropping out at the bottom and being replaced by a lot of units that are coming from standard down to substandard," he said at a recent housing policy conference in Alexandria. The two-day meeting was sponsored by the Institute for Policy Studies, a Washington organization, and the College of Architecture and Urban Studies at the Virginia Polytechnic Institute and State University.

Apgar said the MIT-Harvard center's studies indicate "that as many as 20 percent of the 3.5 million units in metropolitan areas that were judged inadequate" in 1974 have been abandoned and are no longer counted as part of the country's housing stock.

Other speakers at the Alexandria conference argued for different interpretations of housing data.

George Sternlieb, of the Rutgers University Center for Urban Policy Research, argued that the "scrappage rate of rental housing has been declining in the last several years. When you're rich, you throw away used housing and you throw away used cities. When you're poor, you discover they have charm. We are now discovering they have charm."

"We are presiding over a decline in the housing standards of most Americans," Sternlieb added.

John Weicher of the American Enterprise Institute also said that Census Bureau surveys beginning in the 1940s show "steady progress toward eliminating substandard housing. Bad housing isn't as bad as it was 30 or 40 years ago."

Just how much better it really is, however, is hard to pin down because the Census Bureau, the Department of Housing and Urban Development and other organizations all have changed the questions and definitions they use with successive surveys of the housing stock, making accurate comparisons impossible, several speakers said.

Apgar argued that although the quality of housing has improved steadily since the early 1940s, when many homes, particularly those in rural areas, had no indoor toilets, piped-in water and electrical power, the pace of the improvement has slowed substantially since the mid-1970s. There has been "very little net improvement in the rental housing inventory" in the last 10 years, he added.

In his research, Apgar said, he has "looked at the absolute number of units that are deficient, and I found out that most of the improvement is the result not of a reduction in the number of substandard units, but . . . of the growth of the housing inventory. . . . The core of substandard units has remained largely unchanged."

A major factor in the deterioration of rental housing has been tenants' growing inability to take care of their apartments or pay their landlords enough rent to save buildings from ruin. While incomes declined sharply, rents remained the same in the 1970s and went up as much as 6 percent after 1980, he said. As a result, "the housing situation of many low-income households has taken a serious turn for the worse."

Households paying more than 35 percent of their income for rent have increased from 4.7 million in 1974 to 7.7 million in 1983. A recent report by the General Accounting Office said very poor families -- those whose incomes are less than half the median for the area where they live -- suffered the most serious setbacks in rent burdens. About 49 percent of these families paid more than half their income for rent in 1983, up from about 38 percent in 1975.

Attempting to explain the income decline among poor renters, the President's Commission on Housing said the apparent drop is a result of many middle- and upper-income people moving from tenant status to home ownership, leaving low-income people to make up a larger share of the tenant population.

Apgar said that although many former tenants have bought their own homes during the last 10 years, "that nowhere begins to explain this growth in absolute terms of the number of people with less than $10,000 in real income. Renters' income just didn't keep up with inflation as well as homeowners' income," which remained at about the same level over the 10-year period. The lag is not surprising, "since the long-term renter population includes many of the nation's most economically disadvantaged citizens," he added.

Nearly 16 percent of all renter- and owner-occupied housing units -- or 11.6 million -- were structurally inadequate in 1974, according to Harvard-MIT tabulations of data gathered in HUD's Annual Housing Survey for that year. This number had dropped slightly by 1981. Although more than 12 million new housing units were built in the United States between 1974 and 1981 and only about 4 million units were lost from the national inventory because of demolition or other reasons, the number of substandard units dropped by only 158,000, according to the survey figures.

"These statistics suggest," Apgar said, that "lack of maintenance and repair led to a deterioration of the existing inventory. . . . For every thousand inadequate units removed from the inventory from 1974 to 1981, 842 units of existing housing . . . moved from the adequate to the inadequate category as a result of lack of maintenace and upkeep."

Much of the housing that has dropped into the substandard category since 1974 is located in metropolitan areas, he said. The cities contained 54 percent of all the structurally inadequate housing in 1974, and the figure grew to 60 percent in 1981.

The count of substandard housing units would go up if survey techniques were changed, believes Anthony M. Yezer, an associate professor of economics at George Washington University. Most methods are flawed because "they are based on units, not people," he said. Using percentages is misleading because "as the number of units rises, the percentage of those that are inadequate may fall, but the actual number of inadequate units may rise."

The quality of vacant housing and its availability to homeseekers is not measured, he said. "Crowding, location, neighborhood characteristics" also are not taken into account.

Apgar agreed that neighborhood information should not be overlooked. "It's got to be true that when people are talking about the quality of their home in any real sense, they're talking about the structure and the schools and the public services that are provided . . . so it makes sense, if we're going to develop a quality measure we ought to think about the neighborhood attributes," he said.

The American Housing Survey -- known as the Annual Housing Survey until it was reduced to a biennual study by HUD in 1981 -- does include several questions to gauge residents' feelings about their communities. The answers indicate that "the condition of the neighborhoods in which much of the rental housing stock is located is deteriorating, in terms of more litter, more junk, more abandoned buildings . . . ," Apgar said.