Once upon a time, the Department of Housing and Urban Development made lots of low-interest, long-term mortgage loans to nonprofit sponsors for the construction of housing projects for the elderly -- the so-called Section 202 direct-loan program.

With rental assistance payments under another HUD program, nonprofit sponsors could provide housing accommodations for needy elderly tenants at below-market rental levels.

On Maryland's Upper Eastern Shore -- Cecil, Kent, Queen Anne's, Caroline and Talbot counties -- Upper Shore Aging Inc. (USA) managed nutrition and transportation programs for the elderly under a special state charter. In 1977, USA applied for and received a Section 202 fund reservation of $3.75 million to develop 130 units of housing, its first venture into building.

However, USA's approach was unusual. It wanted to distribute the units in six towns -- Perryville, Cecilton, Chestertown, Sudlersville, Greensboro and St. Michaels.

An article about the project published by the Urban Land Institute in its 1983 book, Housing for a Maturing Population, noted that "the heart of the sponsor's sociological and political strategy was to develop small, intimate minivillages in several existing communities to which elderly residents had strong familial, social and religious ties."

But the article also pointed out that "it was obvious to all, and especially to HUD, that this strategy posed serious questions of economic feasibility. Clearly, the most cost-effective and logistically simple plan would have been to construct all 130 units together in a few buildings on one site."

Nevertheless, the sponsor was convinced, and HUD initially agreed, that a typical 130-unit apartment building project would fit poorly into these historic Maryland towns. Detached houses -- some humble, some stately -- characterize these rural communities. Their streets are lined by homes clad in brick or clapboard siding with porches and gable roofs, dormers and chimneys, neoclassical cornices and low garden fences. The fabric of these towns, while venerable and charming, is fragile.

Fortunately, small and affordable sites were available in each town, and all had reasonable access to nearby commercial facilities for convenience shopping.

The architectural challenge was clear. On the one hand, buildings had to be designed to meet stringent, legally defined budget limitations and HUD "minimum property standards." On the other hand, there was a compelling mandate to create an appropriate aesthetic image, one that would be compatible with the existing townscapes and, at the same time, provide an accommodating domestic environment for the elderly.

The design concept envisioned a small, one-story prototype building -- a "quad" -- containing four side-by-side, one-bedroom apartments of about 575 square feet each. In turn, quads could be clustered around a kind of "village green." The size, geometry and orientation of the cluster arrangement could respond to the particularities of the sites in each town.

At all six sites, communal facilities -- laundry room, meeting room, manager's office -- were required. Sidewalks crisscrossed the landscaped village greens to interconnect quads and communal facilities; walks also led to limited on-site parking and abutted public sidewalks and streets.

The quad was designed to look like a big house. Each of the four apartments, stretching from front to back, had its own porch on the living/dining/kitchen side facing the green. Bedrooms on the opposite side had sliding glass doors leading to side and rear yard areas. A ridged roof, hipped at the four corners, served to unify visually the overall form of the quad building. In a reversal of customary material usage, and for budgetary reasons, the on-grade porches were framed by brick piers and lintels, while the building proper was covered with aluminum siding, the color of which could be varied.

A total of 27 quad buildings, plus five community buildings, were to be built. Perryville's site necessitated a special, two-story walkup building. Thus Upper Shore Aging had to build 33 buildings containing 130 units in six geographically scattered locations, all at the same time. The hope was that by using a repeatable prototype, the economies of standardization would offset the increased costs and inefficiencies of dispersal.

As it turned out, perseverance and extra money, in addition to hope, were needed; the project came close to dying several deaths before it was realized.

Although conceived by USA and funded by HUD as one project, it had to be treated as six separate projects with six mortgages, six sets of construction documents, six appraisals, six separate initial and final closings . . . six of everything. This greatly increased overhead costs and processing time, complicated scheduling and magnified design and construction efforts. Further, the scattered site approach, despite unified ownership, resulted in redundant property management operations at each project site. All of these continually threatened to make the project economically unfeasible both to build and manage.

Cost extras, especially for unforeseen site work, exceeded original budget allocations. Yet HUD balked frequently at allowing contingency funds to be used, and it refused to allow savings to be transferred from site to site or between previously approved line-item allocations. For example, Chestertown's project cost $30,000 less than budgeted, but this savings could not be reallocated to cover site work overruns at Greensboro.

Delays in processing by various governmental authorities continually delayed the start of construction. Labor and material prices were rising monthly, pushing the project's projected construction cost above statutory limits. At one point, when architectural and engineering drawings were nearly completed, the sponsor discovered that it had miscalculated the construction budget. As designed, the project couldn't be built. Only after the architects, engineers, contractor and sponsor agreed to make cost cutting (but concept conserving) revisions in the design and specifications could the project again proceed -- but delayed even further.

When finally completed, $4,009,500 had been spent, about $31,000 per unit. This included all development costs: land, construction, architectural and engineering fees, loan interest, permits, insurance, miscellaneous consultants and general overhead. Only $250,000 above the original 1977 allocation, this 6.7 percent cost increase over a three-year period of high inflation seems remarkable.

The units were rented almost immediately. Tenant waiting lists were needed. To choose occupants from among the pool of qualified applicants, a locally based selection committee was formed and included citizens and government officials, senior citizen group representatives, clergy and the developer.

Today, tenants continue to express satisfaction with USA's concept. Although small, apartments seem like "bright little houses" with their separate entrances, big windows and rooms filled with memorabilia. Residents and visitors enjoy the communal spaces, the intimate village greens shaped by the quads and the front porches on which they can sit and socialize. Many plant and care for flowers, shrubs and vegetables.

Passing through town you might not even notice these small-scale minivillages. The largest, at Chestertown, contains 30 units; the smallest, at Sudlersville, contains only 16. Yet densities are relatively high -- Sudlersville's four quads, community building and parking occupy less than an acre.

This project exists only because of the unrelenting persistence of everyone involved. But it was often frustrating. Some officials were so wedded to regulations and don't-trust-anyone thinking that, when faced with an atypical yet meritorious proposal, their first reaction automatically was to say "No!" Happily, those who believed in "Yes" prevailed.

NEXT: Middle-density housing.