The quality of the physical and biological capital used
in agriculture has improved greatly as a result of
technological advances in seeds, plants, animals,
machinery, buildings and agro-chemicals. Much of
today's capital substitutes for the horses and unskilled
labor used earlier. Other modern capital complements
the highly skilled farm labor generated by our general
and agricultural educational systems. Much of the
technology that has improved the capital used in
agriculture was produced by the agricultural research
institutions of the land-grant colleges and USDA.
However, in the early years, especially, innovative
farmers and artisans improved equipment, livestock,
crops and machines. At the same time, improved
policies, processing, storage and transportation
technology, and infrastructure have permitted increased
specialization in production and marketing among
agricultural regions, between the farm and non-farm
sectors, and among individual farms. There have also
been institutional improvements, private as well as
public, to provide marketing services, credit, transpor-
tation services and insurance. Again, the USDA/land-
grant institutions have done much of the research on
which these institutional improvements have been based.

 Not only must U.S. agriculture increase capacity to
 produce at a substantially higher rate, but it will also
 be required to maintain a fine balance between the
 supply of products produced and widely fluctuating
 demands for those products. If supply exceeds demand
 at what would be equilibrium prices by, say, 5 percent
 per year for four years, the results are disastrous: reduced
 prices in the absence of price supports or accumulating
 government-held stocks when prices are supported. In
 about 80 percent of the years since 1918, U.S. agriculture
 has suffered low prices, been subjected to production
 controls and/or has had burdensome government-held
 stocks. It is a dilemma: though U.S. agriculture is re-
 quired to expand production steadily, it often suffers
 from severe overproduction problems. Our problem is
 clearly one of creating the capacity to increase
 agricultural output while converting only as much of
 that capacity into actual production as we need.
 Nonetheless, it is much better to have an agricultural
 economy that tends to over- rather than underproduce.
 Though some have argued that the food shortages of the
 '70s ushered in a perpetual era of unlimited demand for
 and restricted supplies of food products, the events of
 1981 and 1982 indicate clearly that 1974 to 1979 were
 exceptions, not the rule (Johnson and Quance, 1972).
 Since the early '70s, U.S. agriculture has gone from
 surpluses to shortages and back to surpluses. Currently,
 some are declaring that U.S. agriculture, because of
 resource limitations, now faces more instability and
 uncertainty than formerly. That conclusion seems ques-
 tionable in view of the instabilities of the 1920s, the


economic collapse in the '30s, World War II and the
Korean and Vietnam wars, the activities of the
Organization of Petroleum Exporting Countries, fluc-
tuations in income levels and technology, and other
changes.
 The U.S. agricultural research establishment (ARE)
exceeds that in any other country in size, capacity and
accomplishments. In numbers of agricultural scientists
and comparative priority of governments for food, fiber
and forest products research, however, the Soviet Union
and the People's Republic of China may soon outpace
the United States (Wittwer, 1981a). Leadership in
research on technology for food production and
renewable agricultural resources will continue to reside
with the United States, however. The United States now
produces, consumes and exports more food than any
nation in all history. Sixty-one percent of the grain that
crossed international borders in 1979 was grown in the
United States (Batie and Healy, 1980). Agricultural ex-
ports for 1980 and 1981 exceeded $40 billion and offset
more than three-fifths of the cost of imported oil. We
have an important stake in keeping our agricultural
technology superior to that of our competitors.

 Our projections for the next half-century indicate
 needs for yield-increasing technology and technology
 that permits, first, additional and perhaps more fragile
 soils to be farmed, and second, more crops and more
 intensive mixtures of crops to be grown on the land
 farmed. We will require more labor-saving technologies
 and human development and institutional changes to get
 the new technologies used (Martin, 1983). Despite our
 current surpluses, we will have to continue to work hard
 and to invest more money, time and the services of
 highly skilled people in agricultural productivity.

 The USDA's National Inter-Regional Agricultural
 Production (NIRAP) model (NRED, undated) projects
 land-use requirements between 1980 and 2030. Current,
 unpublished baseline projections of land use from that
 model show that cropland and improved pasture will
 decrease from 543 million acres in 1982 to 523 million
 in 2030, with cropland increasing from 396 million acres
 to 424 million. This implies that pasture would decrease
 from 144 million to 96 million acres. Irrigated cropland
 will about double. Baseline projections show a change
 in the productivity index from 100 in 1977 to 193 in
 2030; thus, the NIRAP model presumes substantial ad-
 vances in yield-increasing technology and in technologies
 to permit existing land to be farmed more intensively.
 The NIRAP model does not allow the farming of
 substantially more land and, hence, indicates little about
 the need for technologies to farm additional less produc-
 tive, more fragile soil. In effect, the nature of the NIRAP
 model requires it to secure additional output per unit
 of land while leaving overall land use essentially un-