.

Why is housing so expensive when televisions, computers and many other goods cost relatively less now than they did before?

Between 1970 and 1987, the consumer price index tripled, but the price of clothing and telephone services only doubled. The median sales price of a privately owned one-family house, on the other hand, jumped from $23,400 to $104,500 - a breathtaking increase of 447 percent.

Breathtaking and, for people trying to purchase their first home, heart-stopping. Only 20 percent of Americans now earn enough to purchase a new house at market rates without a trade-in, a dramatic drop from 50 percent two decades ago. For millions of people, the dream of owning a home has faded.

The main difference between houses and televisions, of course, is that houses require land, which is in fixed supply with rapidly escalating costs. Housing prices also are affected by interest rates, local business conditions and other factors that are hard to ameliorate.

But one factor that can and should be changed is the outdated way we build houses. Modern building techniques could reduce the cost of a new home from, say, $100,000 to $90,000, or even less. That is not a huge difference, but every dollar counts, particularly when one computes interest costs over the life of a mortgage.

Most builders in our country now produce houses one by one with conventional materials instead of taking advantage of mass production techniques and newer technology. They install bathrooms one fixture at a time rather than using prefabricated units with the lights, toilet, sink and tub already in place. They do the same for kitchens and make inadequate use of breakthroughs in composite materials, microelectronics and robotics.

The lowly two-by-four remains the primary construction material, even though a growing demand for wood products has caused it to become scarcer and more expensive. Few American home builders have thought seriously about replacing two-by-fours, a sharp contrast with the situation overseas, where many builders are experimenting actively with alternative materials and systems.

The failure of the construction industry to innovate threatens its own future in the same way that technological complacency hurt U.S. automobile and steel manufacturers. In some states, segments of the construction industry are now dominated by foreign companies.

For frustrated home buyers who lack the money even for modest "starter" homes, the situation is already critical. It will probably get worse so long as housing follows the characteristics of a service industry rather than a manufacturing industry. The aging of the baby boom generation and other trends may provide some relief, but low productivity will keep many Americans in rental units instead of their own homes. Those at the bottom of the economic ladder, in particular, will face rising rents and fewer viable options.

Home builders are not inherently averse to new technology. However, there now is little incentive within the industry to invest in technological innovations. Developing new technologies is expensive, requiring not only basic research but also material testing, construction of prototypes, code approval, tooling for production and marketing of the final products. Any one of these activities may take several years.

As things stand, the would-be innovator has no way of knowing what interest rates, the money supply and other conditions essential to success in the housing market will be like when the product is finally ready. As a result, over the past 15 years the building industry has tended to make minor changes to existing products rather than invest in true innovations.

For the sake of millions of would-be home buyers, this needs to change. One of the best ways the industry could become more innovative is through new public and private programs that spur fresh concepts and new products. Test beds should be established to try out appropriate ideas, facilitate testing and speed regulatory approval of innovations. To succeed, experimental programs would need to protect prototype designs and a limited number of housing units from frivolous lawsuits, and to disseminate their results widely. Instead of wringing our hands endlessly about housing costs, it's time we tackled each of the components of that cost and, with respect to technology, became more creative about supporting research and development. Americans need houses they can afford.