One of the most striking and central features of the development of U.S. agriculture has been the increased output per acre, per head of livestock, per farm, and per farm worker, especially during the twentieth century (Bonnen 1986). These are some of the more visible and tangible signs of productivity growth in agriculture arising from investments in the development and adoption of new methods, machines, plant varieties, livestock breeds, and agricultural production know-how.

In the earlier period of expansion of the agricultural land base, production growth came partly from increases in land and labor inputs, as well as from improved technology (Cochrane 1993). During the past 100 years, improved technology has allowed continuing growth in agricultural output from a given or shrinking area of agricultural land, while at the same time using many fewer hours of labor. Labor-saving technology has been critical. Much of the reduction of the agricultural labor force has come about through the consolidation of farms into fewer, more specialized, and larger units. In some instances, especially in intensive livestock and some specialty crops, vertical integration of farming with pre- and post-farm production has been an important element of structural change, facilitated by changes in technologies.

Through these means, U.S. agriculture has been able to supply such an abundance of food and fiber that real, that is, inflation-adjusted, farm-gate prices of food and fiber are much lower than they were 100 years ago, in spite of the growth in demand for most agricultural products that has accompanied the growth in population, per capita income, and trade (Schultz 1956). Productivity growth is the reason why the Malthusian nightmare, in which subsistence defines the entire economy, has not materialized.

Much of the dramatic transformation of U.S. agriculture over the past 100 years, as well as before that time, can be traced to the adoption of new technologies that allowed more to be produced with less (Griliches 1957; Smith and Roth 1990). To understand fully the implications of technological change requires considering and understanding all of the causes and impacts. Such understanding is elusive because the relationships are complex and ever-changing. A first step is to document those aspects that can be measured and more readily understood.

Back To Top

The data in this section provide a comprehensive picture of U.S. agricultural productivity during the twentieth century, and some partial but useful measures of productivity extending back into the earlier periods. Economists use the word "productivity” in a technical sense with a meaning similar to that found in common usage, only more specific. The general notion is to measure the quantity produced, compared with the quantity or the cost of the inputs used to produce it.

Some partial productivity measures express the quantity of a particular output relative to the quantity of a particular input or resource – output or yield per acre, or output per worker (usually per year). Other measures account for more of the inputs. A total productivity measure would express total output relative to the total quantity of all of the inputs used in production, but we rarely have all the data needed to measure the totality of inputs and outputs. More often, what is practical to achieve is a multifactor productivity (MFP) measure that expresses aggregate output relative to aggregate input – perhaps omitting certain outputs and inputs that are either difficult to measure or not sufficiently covered by available data. For example, the accumulation of highly localized, within-farm information on soil conditions or improved planting, weeding, and harvesting operations has important productivity consequences. Management skill is another type of unmeasured input that accounts for some productivity growth.

These measures or indexes of input, output, and productivity necessarily involve aggregating across different commodities, or different qualities of the same commodity, at a given place and point in time; they usually also involve aggregating to some extent over space and time as well. The measures themselves will depend on the decisions made about how to go about this aggregation, which depend to some degree on the availability of data (Griliches 1960, 1963).

As a related matter, the choice of indexing procedure may be important. The so-called index number problem arises when distortions in the aggregate quantity (or price) index result from the use of inappropriate price (or quantity) weights in aggregating the quantities (or prices) of individual goods. For instance, the aggregate price index for agricultural output was computed as a Laspeyres index, in which the series of prices of each of the individual commodities making up the index was multiplied by a weight equal to its individual share of the total value in the base period. This type of index overstates the rise in the cost of living over time because it puts too much weight on the goods that become relatively more expensive over time. Even when this index number problem, associated with the formulation of the index, is avoided or minimized (for example, through the use of Divisia indexes or approximations to Divisia indexes, such as a Fisher–Ideal index), similar problems can arise in aggregating and comparing inputs, outputs, and productivity over space or time. A more complete treatment of these conceptual issues and empirical applications to U.S. agriculture can be found in Acquaye, Alston, and Pardey (2003), AAEA (1980), Ball (1985), Capalbo and Antle (1988), Huffman and Evenson (1993), and Jorgenson and Gollop (1992).

Aggregation bias associated with heterogeneity of inputs and outputs is an issue in developing an accurate sense of the aggregate picture, particularly over long time periods. In aggregating quantities it makes economic sense to use prices as weights, but if the prices and quantities are not perfectly matched (for example, a single national price is used for all states, a single annual price is used for all months or all years, or a single price is used for all grades of a particular output, such as wheat), the picture may be distorted in that changes in quantities of lower- (higher-) value goods will be over- (under-) emphasized (Craig and Pardey 1996; Griliches 1963). For example, over time, the quality of labor and machinery used in agriculture has risen. When we ignore this fact, and just count hours of labor or the number of tractors, our measures overstate the gain in productivity. (Some of the additional output is a result of better inputs; equivalently, a greater quantity of inputs would have been required if input quality had remained constant.)

Back To Top

In 1990, U.S. agriculture produced, in aggregate terms, more than three times the quantity of output in 1910, implying a compound growth rate of 1.61 percent per year. Over the same time period there were substantial shifts in the mixture of output, reflecting both changes in the composition of demand for food and fiber products as incomes have grown and changes on the supply side, reflecting a host of technological innovations (Table Da1063–1081). In 1910 oil crops accounted for just 0.7 percent of the value of production; by 1997 their share had grown to 9.5 percent. Similarly, greenhouse and nursery products accounted for 5.3 percent of the value of output in 1997, well above their 1.8 percent share in 1924. Other crops have declined in importance. The replacement of the horse with the tractor, especially during the first half of the twentieth century, accounts for a significant shift to reduce production of oats (used to feed horses) and to increase both the production of other crops and the cattle herd.

The 1.61 percent per year increase in output over 1910–1990 was achieved with only a 0.06 percent per year increase in the total quantity of inputs, and, comparing those two figures, agricultural productivity grew by 1.55 percent per year (Table Da1117–1122). The comparatively small change in aggregate input use hides a good deal of variation across different categories of inputs, even for a national index. In 1910, labor accounted for 29 percent of the total cost of inputs, but by 1997, the labor input accounted for only 11.9 percent of total input costs (Table Da1082–1094). As a share of total input costs, energy grew rapidly from 1910 to 1932 and continued to grow gradually thereafter. Fertilizer, lime, and pesticide expenses have generally accounted for between 4.6 and 13.8 percent of total input costs. Beginning in 1950, purchased intermediate inputs grew from 14.5 percent of input costs to 18.7 percent in 1997, with a general decline in inputs purchased from farm origin.

These measured output and especially input trends can be misleading, because important changes in quality have not been taken into account in the measures. Simply counting machines does not capture the fact that machines are much more efficient than they were fifty or even five years ago. Similarly, the composition of the agricultural labor force in agriculture has changed to include more experienced and better educated farmers; hence "hours of work” in agriculture means something quite different today than it did in 1910 (Acquaye, Alston, and Pardey 2003). Nevertheless, labor-saving machinery represented an important element in the overall growth in farm productivity. Important innovations in cropping were introduced when tractors replaced horses and self-propelled combines replaced tractor-drawn combines. In earlier periods, of course, the mechanical reaper and binder replaced the sickle and manual shocking.

Much of the farm mechanization for cereal crops started in the 1800s, with important innovations continuing throughout the 1900s, especially in the first half of the century. In the cotton industry, mechanized picking systems did not appear until after World War II. Innovations to mechanize the harvesting of some other crops were much more recent; for example, mechanical harvesters for canning tomatoes and various tree crops were developed after 1960 (Kislev and Peterson 1981). Other important mechanical innovations include various irrigation technologies; technologies used off the farm to transport and process the harvest, including canning, refrigeration, and other food preservation technologies; and other processing technologies such as Eli Whitney's cotton gin. Electrification was an important development that enabled the adoption of these technologies, in particular in the dairy industry, which was revolutionized by the introduction of milking machines and refrigerated vats.

A significant factor in the aggregate productivity patterns has been the use of biotechnology to improve genetic traits, especially of crops. These changes have led to improved disease resistance; quality improvements such as more uniform grain and fruit, among others; better tolerance for drought, waterlogging, or shorter growing seasons; better adaptation to particular climates or soil conditions; or greater suitability for mechanical harvesting (including more uniform ripening and the ability of the plant or its fruit to withstand mechanical processes) (Olmstead and Rhode 1993). Figure Da-I and Table Da1095–1107 show that per acre yields of crops have grown dramatically, especially during the past fifty years, but with some significant variation among the crops. Genetic improvement is not the only factor, as there have been improvements in chemical fertilizers, irrigation, and weed and pest control, but it is the main factor in many cases, especially for wheat, rice, and corn (Hayami and Ruttan 1970).

Productivity has generally not grown as quickly in the livestock industries as in the cropping industries, partly because genetic improvement has been slower (see Table Da1108–1116). Nevertheless, there have been important innovations in terms of stocking rates, disease control, greater reproductive efficiency, and improved feed conversion efficiency. Such changes have enabled the development, for instance, of intensive large-scale, highly cost-efficient hog and poultry operations. To do so has meant combining elements of improvements in genetics, animal housing, feed, veterinary knowledge and medicines, and, importantly, livestock husbandry. Many of these changes have been somewhat controversial. People concerned with animal welfare have questioned the intensive livestock systems; people concerned about food safety have questioned the use of growth hormones (for example, rBST used to increase milk yield in dairy cattle was very controversial) and transgenic forms of biotechnology in animals or plants (for example, the introduction of a gene from a soil bacterium, Bacillus thuringiensis, into corn and cotton to confer pest resistance, or using similar methods to incorporate herbicide tolerance into soybeans).

Another perspective on productivity can be gleaned by considering the aggregate output per acre or per hour of work, as shown in Figure Da-J. Land productivity more than doubled between 1910 and 1990, but this was dwarfed by the nearly twenty-fold increase in labor productivity over the same period (Table Da1172–1174). This comparison reflects a more than trebling of output against a comparatively unchanging land base and a rapidly declining labor force in agriculture. The increases in land and labor productivity occurred mainly after the 1940s, with a slowing in the growth of both productivity indexes in the 1980s. Increases in the quantities of other inputs, such as fertilizers, herbicides, electricity, fuels, and irrigation, account for some of the growth in output per hour or per acre, but much of the measured growth in productivity reflects changes in technology.

Among the most often cited historical time series of agricultural statistics are the figures on labor requirements for various farming operations. These labor-requirement figures have been updated to the last year before the series were discontinued, and are included here in Table Da1143–1171. Like the measures of labor productivity in Figure Da-J, these labor requirements are partial productivity or factor intensity measures, reflecting the effects of technology and other changes on input–output ratios. For example, in corn production, the figures on the number of hours worked per acre fell from 35.2 in 1910–1914 to 3.1 in 1982–1986, and the number of hours required to produce 100 bushels fell from 135 in 1910–1914 to 3.0 in 1982–1986. The much greater saving in hours per bushel reflects an implied increase in yields from 26.1 bushels per acre in 1910–1914 to 103 in 1982–1986, broadly consistent with the average yields in series Da1096. In turkey production, the figures on the number of hours worked per hundredweight of turkeys fell from 31.4 in 1910–1914 to 0.2 in 1982–1986, a much greater improvement in labor productivity compared with corn, reflecting great changes in input combinations as well as technical improvements.

The labor requirement series were discontinued in 1986 because of concerns over their accuracy and interpretation. It is not known how the figures were derived in the past, but it seems that they were based on expert opinions about best or average practice, and not on any statistical sampling of actual data.¹  The main virtue in these figures is that the series extend back for a long time, into periods for which no alternative data are available on input–output relationships. They might be most useful as indicators of longer-term changes rather than as specific measures of average productivity at any particular time, and they illustrate the dramatic changes that have taken place, but they must be used with care.

In aggregate terms, in 1991 U.S. agriculture produced more than double the quantity of output produced in 1949. It did this with marginally less aggregate input, causing an index of MFP to grow faster than the rate of growth in output. In annual rate-of-change terms, we estimate that output increased by 1.71 percent per annum over the 1949–1991 period; inputs used in agriculture declined by 0.19 percent, and so measured MFP grew by 1.90 percent per annum. The comparatively small drop in aggregate input use hides a good deal of variation across different categories of inputs, even for a national index.

The quantity of labor used in U.S. agriculture in 1991 was only 40 percent of the labor used in 1949, an average annual rate of decline of 2.2 percent. Labor quality has changed, too. In 1949, 72 percent of the hours worked by farm operators could be attributed to operators with no more than eight years of schooling. In 1991, operators with this level of educational attainment accounted for only 12 percent of total operator hours. In 1949, less than 5 percent of total operator hours were attributable to individuals with any college-level education, but this share increased to 37 percent in 1991.

The amount of land devoted to agriculture also declined. At the national level, land use is reduced from 0.47 percent to 0.30 percent per annum when one controls for changes in the quality of land used in agriculture. This is, in part, the result of the doubling of irrigated acres in U.S. agriculture over the period from 1949 to 1991. There was a slow overall growth in the capital services used by agriculture, even after accounting for the substantial quality improvements in machinery. Purchased inputs such as energy, seeds, fertilizers, and agricultural chemicals represent the only rapidly growing category of inputs; U.S. farmers have more than doubled the quantity of such inputs used since 1949.

These national trends are broadly consistent with regional developments but there are some significant differences in productivity growth (Table Da1232–1243), and even greater variation in the growth of inputs and, especially, outputs among states. Although land inputs have generally declined, they have increased in quality-adjusted terms in some of the drier Western and Southern states, reflecting a more rapid increase in higher-quality irrigated cropland in those states. Disaggregation of the labor inputs reveals that the temporal pattern of change is not the same in every part of the United States nor is it particularly uniform over time. The reduction in total hours worked was more dramatic in the Southern states than in the rest of the country. But in the West the quantity of labor used in agriculture actually increased over the 1970s and 1980s against a national trend of decreasing labor use.

Back To Top

Over the centuries, much of the innovation in agriculture has been the result of tinkering by farmers and on-farm experimentation – including much breeding and selection of plants and animals and the development of improved methods, practices, tools, and mechanical innovations. Before the modern scientific age, great advances in American agricultural productivity resulted from the efforts of plant prospectors who imported new and improved varieties from foreign lands. Walter Bruling of Mississippi imported a new cotton variety from Mexico in 1806 that would become the mainstay of the early American cotton industry; Agoston Haraszthy helped transform the California wine industry by importing several hundred grape varieties from across Europe, North Africa, and the Middle East in the 1860s. In addition to the efforts of private citizens, the USDA sent its scientists to the far corners of the globe in search of better plant varieties. These efforts were particularly fruitful in introducing varieties of wheat and other crops suitable for the conditions on the Great Plains (Alston 2002; Pardey, Alston, et al. 1996).

Among the inventors who devised some of the more widely known innovations in U.S. agriculture, we can count Eli Whitney, who patented the cotton gin; Cyrus McCormick, whose mechanical reaper "made bread cheap”; John Deere, whose steel-tipped moldboard plows helped tame the prairies; and Hiram Moore, who built the first combined harvester (combining a reaper and a thresher in one machine). The list of biological innovators is less well known, but the legendary Luther Burbank, who developed scores of new and improved varieties, many of which still bear his name, is representative of thousands of farmer-scientists who by careful selection and, in some cases, hybridization, improved the plant varieties available to American farmers. The private sector has continued to emphasize more patentable inventions. In agriculture, in particular, however, it is difficult for individuals to appropriate fully the returns from their research investments, and it is widely held that some government action is warranted to ensure an adequate investment in research and development (R&D) (Alston and Pardey 1996).

In earlier years, agricultural innovation was encouraged primarily by state and local governments and farmer organizations, through the awarding of prizes and demonstrations of best practice at county fairs and such, but with relatively little organized public research. Until 1862, federal government action to encourage investment in agricultural (and other) research was primarily through patent law enabled by article 1, section 8 of the U.S. Constitution, ratified in 1788. Since 1862, which marked both the establishment of the U.S. Department of Agriculture (USDA) and the passage of the Morrill Land Grant College Act, state and federal governments have become progressively more involved through public investments in agricultural R&D.²

The first Commissioner of Agriculture – the non-cabinet post that originally directed the USDA – was Isaac Newton, who prior to the department's establishment had been the Superintendent of Agriculture at the Patent Office, with responsibility for the collection and distribution of seeds for new plant varieties. One of the first acts of the new commissioner was the appointment of a superintendent of the propagating garden, the USDA's first research facility, on what is now part of the Washington D.C. Mall.

Active intramural USDA research began immediately, with publication of the first research bulletin in 1862, describing the sugar content and suitability for winemaking of several grape varieties. The early years of the USDA were marked by the slow but steady expansion of the department's internal scientific activities, mostly devoted to "service” work rather than the discovery and development of new knowledge, but leading eventually to the establishment of the Agricultural Research Service (ARS). Under the Progressive Era leadership of James "Tama Jim” Wilson, from 1897 to 1913, the USDA budget grew dramatically (by more than 700 percent during Wilson's tenure), and, by 1904, employment of scientists within the USDA surpassed total employment of scientists in the state agricultural experiment stations (SAESs).

Publicly funded research outside the USDA grew out of the state agricultural experiment stations, developed first in Connecticut by Samuel W. Johnson, based on the prototypes developed in Germany in the 1850s. Following work prior to the Civil War to analyze chemical soil enrichers, Johnson was designated state chemist in 1869, with the formal establishment of the Connecticut Agricultural Experiment Station at Wesleyan University in 1875. At about the same time, experimentation directed to the problems of local farmers was beginning at the state land-grant institutions. USDA funding of external (that is, extramural) research followed the passage in 1887 of the Hatch Experiment Station Act. In the following years, agricultural experiment stations, supported by a mixture of federal, state, and some private funds, and generally located at the various land-grant colleges, opened across the country.

In the early years, extension activities provided local information and technology transfer services to farmers. Eventually the importance and popularity of extension activities led to legislative support, through the Smith–Lever Act of 1914, which created the Cooperative Extension Service and instituted a federal role in extension. During the 1930s, state support for the experiment stations fell sharply. In response, the Bankhead–Jones Act of 1935 provided additional federal support for research as well as extension. Several important features of the earlier Smith–Lever Act carried over to the Bankhead–Jones Act, notably the disbursement of federal funds to the states on a formula basis (20 percent of the funds distributed equally among states, 40 percent based on the state's share of the U.S. rural population, and 40 percent according to its share of U.S. farm population), provided the federal dollars were matched by state funds. In the ensuing years there were further legislative changes, as documented in Table Da-K and Table Da-L, but the main elements of combined federal and state funding institutions remained unchanged. Importantly, however, the total amount of funding for public agricultural research and extension, and the sources and disposition of those funds, changed dramatically, especially from the 1960s forward.

Table Da1244–1252 gives a long-term perspective on agricultural R&D spending. In 1889, shortly after the Hatch Act was passed, federal and state spending appropriations totaled $1,119,136. A little over a century later, in 1997 the public-sector agricultural R&D enterprise had grown to more than $3 billion, an annual rate of growth of 7.99 percent in nominal terms and 4.39 percent in real terms. Intramural USDA research accounted for an increasing share of the national system, until the late 1930s, after which the SAES share grew to 69 percent of total public spending on agricultural R&D by 1997. Of the funds spent in the SAESs in 1997, 31 percent was from federal sources; 47 percent from state government; and 21 percent from industry, income earned from sales and technology licenses, and various other sources.

Along with the growth in research expenditures, the number of research personnel grew substantially as well. In 1998, the SAESs employed about 7,300 full-time equivalent scientists in addition to the 2,270 researchers employed by the USDA.

In 1915, the first year in which federal funds were made available for cooperative extension between the USDA and various state extension agencies, approximately $1.5 million of federal funds were combined with $2.1 million made available from various state and local government sources for a total of $3.6 million (Table Da1260–1265). This total grew by 7 percent per annum to reach $1.6 billion by 1999. The public provision of extension services in the United States is essentially a state or local activity. Consequently funds from within-state sources accounted for 74 percent of the total funds for extension, with federal funds accounting for the remaining 26 percent in 1999, much less than their peak share of 62 percent in 1919.

In 1992, the private sector spent $3.4 billion on in-house agricultural R&D, about 31 percent more than the amount spent by the public sector. Private spending grew by 9.85 percent per annum over the period 1960–1992 (series Da1253), faster than the 8.67 percent per annum for public agricultural R&D spending over the same interval (series Da1244). As a result, for every dollar of publicly conducted research in 1992 the private sector spent $1.31 compared with just 96 cents in the early 1960s.

The long-term trend of rapidly growing total spending on public agricultural R&D (including extension) slowed in the 1990s. An important element of this was a reduction of federal support for SAES research. This has meant a shift in the balance between the private and public elements in both the funding and the conduct of the research, changes in the mix of research being funded, and changes in other aspects of agricultural science policy as well. Some of these changes have come about as part of a shift in funding for all science, some as a result of a shift of public support generally for agriculture, partly driven by changes in agriculture (Pardey and Beintema 2001). In addition, there have been changes in agricultural science encompassing new molecular biotechnologies and concomitant changes in intellectual property and related institutions.

Among the more striking changes has been the broadening of the scope of intellectual property protection to include inventions involving living things. In the United States, the first steps in this direction began with the Plant Patent Act of 1930, which protected asexually reproducing plants, that is, plants such as grape vines, fruit trees, and ornamentals, which are propagated through cuttings and graftings. The patent scope was expanded further in the 1970s and 1980s through the introduction of utility patents for life forms, including asexually and sexually propagated seeds, plants, and tissue culture. Patent protection was complemented by the introduction of the Plant Variety Protection Act in 1970, designed to strengthen intellectual property protection for nonhybrid varieties.

The public investment in agricultural R&D has been credited with much of the strong growth in productivity over the past 100 years, especially the past 50 years. A great many benefit–cost studies have been undertaken to measure the social payoff of the investment. These studies vary in their findings, but generally agree that it has been a highly productive and socially profitable use of U.S. taxpayer funds (Alston, Chan-Kang, et al. 2000).

Back To Top

Series Da1095–1096 and Series Da1103–1105.

Back To Top

Land productivity: series Da1172. Labor productivity: 1910–1984, series Da1173; thereafter, series Da1174.

The values have been reindexed to 1910 = 100.

Back To Top

[TABLE Da-K  Major legislation affecting federal funding of research in state agricultural experiment stations and other cooperating institutions: 1862–1996]

Back To Top

[TABLE Da-L Major legislation affecting federal funding of cooperative extension: 1914–1996]

Back To Top

Acquaye, A. K. A., J. M. Alston, and P. G. Pardey. 2003. "Post-War Productivity Patterns in U.S. Agriculture: Influences of Aggregation Procedures in a State-Level Analysis.” American Journal of Agricultural Economics 85 (1): 59–80.

Alston, J. M. 2002. "Spillovers.” Australian Journal of Agricultural and Resource Economics 46 (3): 315–46.

Alston, J. M., C. Chan-Kang, et al. 2000. A Meta-Analysis of Agricultural R&D Evaluations: Ex Pede Herculem? Research Report Number 113, International Food Policy Research Institute.

Alston, J. M., and P. G. Pardey. 1996. Making Science Pay: The Economics of Agricultural R&D Policy. American Enterprise Institute Press.

Alston, J. M., P. G. Pardey, and V. H. Smith, editors. 1999. Paying for Agricultural Productivity. Johns Hopkins University Press.

American Agricultural Economics Association (AAEA) Task Force on Measuring Agricultural Productivity. 1980. Measurement of U.S. Agricultural Productivity: A Review of Current Statistics and Proposals for Change. ESCS Technical Bulletin Number 1614. U.S. Department of Agriculture.

Ball, V. E. 1985. "Output, Input and Productivity Measurement in U.S. Agriculture: 1948–79.” American Journal of Agricultural Economics 67 (3): 475–86.

Bonnen, J. T. 1986. "A Century of Science in Agriculture: Lessons for Science Policy.” American Journal of Agricultural Economics 68 (5): 1065–80.

Capalbo, S. M., and J. M. Antle, editors. 1988. Agricultural Productivity: Measurement and Explanation. Resources for the Future Press.

Cochrane, W. W. 1993. The Development of American Agriculture: A Historical Analysis. 2nd edition. University of Minnesota Press.

Craig, B. J., and P. G. Pardey. 1996. "Productivity Measurement in the Presence of Quality Change.” American Journal of Agricultural Economics 78 (5): 1349–54.

Griliches, Z. 1957. "Hybrid Corn: An Exploration in the Economics of Technological Change.” Econometrica 25 (4): 501–22.

Griliches, Z. 1960. "Measuring Inputs in Agriculture: A Critical Survey.” Journal of Farm Economics 42 (5): 1411–27.

Griliches, Z. 1963. "The Sources of Measured Productivity Growth: Agriculture, 1940–1960.” Journal of Political Economy 71 (4): 331–46.

Hayami, Y., and V. W. Ruttan. 1970. "Factor Prices and Technical Change in Agricultural Development: The United States and Japan, 1880–1960.” Journal of Political Economy 78 (5): 1115–41.

Huffman, W. E., and R. E. Evenson. 1993. Science for Agriculture: A Long-Term Perspective. Iowa State University Press.

Jorgenson, D. W., and F. M. Gollop. 1992. "Productivity Growth in U.S. Agriculture: A Postwar Perspective.” American Journal of Agricultural Economics 74 (3): 745–50.

Kerr, N. A. 1987. The Legacy: A Centennial History of the State Agricultural Experiment Stations, 1887–1987. Missouri Agricultural Experiment Station.

Kislev, Y., and W. Peterson. 1981. "Induced Innovations and Farm Mechanization.” American Journal of Agricultural Economics 63 (3): 562–5.

National Research Council, Board on Agriculture. 1989. Investing in Research: A Proposal to Strengthen the Agricultural, Food, and Environmental System. National Academy Press.

National Research Council, Board on Agriculture and Natural Resources. 2003. Frontiers in Agricultural Research: Food, Health, Environment, and Communities. National Academy Press.

Olmstead, A. L., and P. Rhode. 1993. "Induced Innovation in American Agriculture: A Reconsideration.” Journal of Political Economy 101: 100–18.

Pardey, P. G., J. M. Alston, et al. 1996. Hidden Harvest: U.S. Benefits from International Research Aid. IFPRI Food Policy Report. International Food Policy Research Institute.

Pardey, P. G., and N. M. Beintema. 2001. Slow Magic: Agricultural R&D a Century after Mendel. IFPRI Food Policy Report. International Food Policy Research Institute.

Rasmussen, W. D., and G. L. Baker. 1972. The Department of Agriculture. Praeger.

Schultz, T. W. 1956. "Reflections on Agricultural Production, Output and Supply.” Journal of Farm Economics 38 (August): 748–62.

Short, Sara. 1986. "Measuring Agricultural Labor Hours and the Rate of Return to Equity in Farm Assets.”  Paper presented at the annual meeting of the American Agricultural Economics Association, Reno, July.

Smith, M. S., and D. M. Roth. 1990. Chronological Landmarks in American Agriculture. USDA-ERS Agricultural Information Bulletin. Revised edition. U.S. Government Printing Office, November.

USDA, Economic Research Service. 1987. Economic Indicators of the Farm Sector: Production and Efficiency Statistics, 1985. U.S. Department of Agriculture.

Weaver, R. D., editor. 1993. U.S. Agricultural Research: Strategic Challenges and Options. Agricultural Research Institute.

Back To Top