Large inflows of money are once again appearing in commercial real estate markets. The key question is: will this inflow of capital cause the industry to start another wave of overbuilding?

Up to March 1996, the evidence that serious amounts of new construction are beginning again is weak and anecdotal - except in retailing and apartments. In retailing, new construction did not take the prolonged nosedive that occurred concerning other types of commercial real estate.

In fact, by 1995, new building in retailing was back up to its mid-1980s highs. But that was mainly because of a technological change in the retail field: the advent of the low-margin "big box" retailers. These discount firms operate with free-standing units or clustered in so-called "power centers" that are in-line strip centers with very large, mainly rectangular stores in them. Because of the lower construction and operating costs of power centers, plus the lower margins achieved for other reasons by "big box" operators, they form a new type of retail outlet that attains very high volumes per store and per square foot.

But in other fields - namely industrial, office, hotels and apartments - no such major technological change has recently taken place. Therefore, when new development plummeted in 1990, it stayed at very low levels until recently. This permitted existing structures, which had been repriced to much lower levels through massive capital losses, to start raising their occupancy rates again.

By 1996, with five years of the general recovery gone by, in many markets, rents on existing spaces were beginning to rise notably too. Because new construction remained dormant, the increases in demands for space created undiluted improvements in the previously-distressed situations of existing properties. As I have said in earlier columns, this was the gradual absorption phase of the three-phase development cycle.

In recent months, there have been mounting reports of more new construction. It began with apartments, especially in the Southeast, which was booming economically. Apartment construction (units in structures with five or more units) hit highs of 576,000 units in 1985 and 542,000 in 1986, then dropped steadily to under 140,000 units per year from 1991-1993. But in 1994, apartment construction moved up to 233,000 units and in 1995 to 243,000 units. Recently it has been running at even higher levels.

New development of industrial space is also occurring more widely than in the past few years. Industrial vacancy rates have steadily fallen from a peak of 9.0% in 1992 to a recent national level of about 6.9%, according to CB Commercial. In metropolitan areas like Jacksonville, Minneapolis, Nashville and San Diego, vacancy rates are under 5.0%, and industrial rents have been rising. So speculative development of industrial space has begun to appear in many parts of the nation - though still on a relatively restrained scale in each market compared to past construction binges.

Office and hotel markets were the most overbuilt of all types, so new development has been slowest to reappear concerning them. Office construction began with isolated build-to-suits for large-scale tenants. Financing for purely speculative new buildings has remained difficult to procure because rents have not recovered enough in most markets to the levels that would make new space profitable. However, a few speculative efforts have begun to appear in those markets where rents are near reproduction-cost-supporting levels.

Will new building reappear on a massive scale again? For a long time, I predicted that there would be no nationwide development boom phase in this general economic cycle, unlike in the last two general cycles (in the 1970s and the 1980s). But the longer this general recovery continues, and the longer the capital markets remain swamped in liquidity, the more new construction is likely to start up again. As vacancies in existing structures fall, and rents rise, investors are bidding up the prices of those existing properties, thereby driving current yields down. Eventually, some investors will decide that they can get higher yields through new development than by buying by-then-expensive existing structures. So they will start financing speculative new developments.

I still doubt that there will be massive overbuilding in the 1990s like that in the 1980s. The next big development boom will come in the next general economic cycle, around 2000. But as long as investment officers get paid to make deals, and starve when they don't make any, the temptation to put money out in spite of evidence of increasing overbuilding will be irresistible. So a creeping cavalcade of commercial construction is underway, and will creep ever-faster as long as this general expansion lasts.

Anthony Downs is senior fellow at the Brookings Institution, Washington, D.C. The views in this article are those of the author and not necessarily those of officers, trustees or other staff members of the Brookings Institution.