REGIONAL TRENDS IN THE PRODUCTIVITY - A i 7 OF AMERICAN AGRECULTURE Thesis for the Degree of 5’21. D. Mififiifim STATE UNIVERSITY L. 09M EAMEERT 1973 » 49 _ ‘a' L I B R A R Y : I'vliehfigan gtatc 3? ., Universxty y‘ .' This is to certify that the thesis entitled Regional Trends in the Productivity of American Agriculture presented by Leland Don Lambert has been accepted towards fulfillment of the requirements for Ph.D. degree in Agricultural Economics Major professor Date February 23, 1973 a {H ABSTRACT REGIONAL TRENDS IN THE PRODUCTIVITY OF AMERICAN AGRICULTURE by Leland Don Lambert This thesis describes the construction of new regional data series on agricultural inputs and their combination with existing regional farm output series to generate new regional productivity series. The produc- tivity estimates are global in the sense that all production inputs and all final products are included. Productivity is defined as total out- put per unit of total input. Input indices were built up from 87 ac- counts which were estimated separately. The land input was entered at the rental cost which removes appreciation from the analysis. That por- tion of feed and seed inputs, which is produced on the farm, was deducted from inputs to avoid double counting. Aggregation was done arithmetical- ly using a Laspeyres weighted aggregate formula with price weight peri- ods, 1947-49 and 1957-59 with a splice at 1955. The reference base was set at 1967 = 100. Regional estimates were made for the 32 year period 1939 thru 1970 and included seven major input subgroups: labor; farm real estate; farm power and machinery; fertilizer and lime; feed, seed and livestock purchases; taxes and interest; and a miscellaneous cate- gory. Regional indices for fixed and variable inputs were also con- structed. The new U. S. productivity series was spliced into an existing series at 1939 to make a continuous series from 1910 to 1970. This longer se- ries was utilized to examine the relation between productivity change and the adoption of major agricultural technologies. Economists commonly Leland Don Lambert assume that technological change is responsible for shifting production functions, yet few studies have attempted to link productivity change to the adoption of major technologies. The major technologies consider- ed were: mechanization, hybridization, fertilization and pesticides. Although multicollinearity limits the evaluation of individual techno- logies, the results indicate that a few major technologies were respon- sible for the bulk of productivity change during this century. Since most of the major technologies were fully exploited by the early 19603, productivity increase has slackened since that time. For the future, minor technologies will contribute a modest improvement in productivity but major change will have to await the discovery of new major technol- ogy. In the appendix is a detailed description of many USDA and other data series utilized as sources of data. There are candid references to the completeness and accuracy of some of these data sources. The states in- cluded in each farm production region follows: NORTHEAST: Rhine, New Hampshire, Vermont, mssachusetts, Rhode Island, Connecticut, New York, New Jersey, Pennsylvania, Delaware, Maryland. CORN BELT: Ohio, Indiana, Illinois, Iowa, Missouri. LAKE STATES: MiChigan, Wisconsin, Minnesota. APPALACHIAN: Virginia, West Virginia, North Carolina, Kentucky, Tennes- see. SOUTHEAST: South Carolina, Georgia, Florida, Alabama. DELTA STATES: Mississippi, Arkansas, Louisiana. SOUTHERN PLAINS: Oklahoma, Texas. NORTHERN PLAINS: North Dakota, South Dakota, Nebraska, Kansas. MOUNTAIN: Montana, Idaho, Wyoming, Colorado, New Mexico, Arizona, Utah, Nevada. PACIFIC: Washington, Oregon, California. REGIONAL TRENDS IN THE PRODUCTIVITY OF AMERICAN AGRICULTURE b Ely A”, L. Don Lambert A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Economics (455’ , l, (fir/i ACKNOWLEDGEMENTS The construction of the input series was a long and tedious process. It was a task which could not be accurately done by one person. The author wishes to acknowledge his gratitude to all who made the project possible, and especially to an outstanding few. Of the greatest assistence to the author was the patience of those who were willing to see the job "done and done right". Heading this list is Dr. Warren Vincent. Next is my wife who not only offered moral support but also put up with the rigors of "metropolitan living". Also on this list belong Dr. Glen Barton, Dr. Lester Mbnderscheid, Don Durost, James Vermeer and Warren Bailey. For technical assistence I am especially indebted to Dr. underscheid and Dr. Barton. I wish to thank Dr. Vincent and Dr. Barton for their adroit- ness in managing the research. I am indebted to Dr. Joachim Elterich for his assistence in straightening out the great feed data discrepancy. Finally, I wish to thank Mrs. Rachel Evans, who did the com- pilations. Many statistical manipulations were required to ar- rive at each of the input estimates. In some cases dozens of calculations were required. Estimates were made for 87 separate accounts for 10 regions for 32 years, a total of 27,840 final estimates. By taking an active interest in the project, Mrs. Evans was able to make accurate calculations with a minimum of supervision and understood the project thoroughly enough to detect several erroneous instructions. 11 TABLE OF CONTENTS Page Chapter I. Need for StUdy O O O O O O O O O O O 0 O 0 O O O l O 0 O 1 ObjeCtives O O l O O O O O O O O O O O O O O O O O O O O O 2 Productivity: Definitions . . . . . . . . . . . . . . . . 2 Survey of Literature . . . . . . . . . . . . . . . . . . . . 4 Theoretical Framework . . . . . . . . . . . . . . . . . . 7 Chapter II. Input Compilation . . . . . . . . . . . . . . . . . . . . . . . .13 Labor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 Real Estate . . . . . . . . . . . . . . . . . . . . . . . . .17 Interest on Real Estate Mortgages . . . . . . . . . . . . . . . .21 Land in Government Programs . . . . . . . . . . . . . . . . . . .22 Depreciation . . . . . . . . . . . . . . . . . . . . . . . . . .23 Insurance and Accidental Damages . . . . . . . . . . . . . .24 Maintenance of Service Buildings & Machinery. . . . . . . . . . .25 Grazing Fees. . . . . . . . . . . . . . . . . . . . . . . . . . .26 The value of Inventory Stocks . . . . . . . . . . . . . . . . . .26 Interest Added by Nonreal Estate Debt . . . . . . . . . . . . . .29 Fuel and Oil . . . . . . . . . . . . . . . . . . . . . . . . . .30 Custom Work . . . . . . . . . . . . . . . . . . . . . . . . . . .30 Fertilizer and Limestone. . . . . . . . . . . . . . . . . . . . .31 Feed and Seed . . . . . . . . . . . . . . . . . . . . . . . . . .32 Hired Trucking, Freight and Express . . . . . . . . . . . . . . .40 Poultry Purchases . . . . . . . . . . . . . . . . . . . . . . . .40 Real Estate Taxes . . . . . . . . . . . . . . . . . . . . . . . .41 Personal Property Taxes . . . . . . . . . . . . . . . . . . . . .41 Dairy Supplies. . . . . . . . . . . . . . . . . . . . . . . . . .42 Pesticides. . . . . . . . . . . . . . . . . . . . . . . . . . . .42 Cotton Ginning. . . . . . . . . . . . . . . . . . .42 Irrigation Operation and Maintenance Cost . . . . . . . . . . . .42 Chapter III. Indexing Procedure . . . . . . . . . . . . . . . . . . . . . . .44 Chapter IV. The Weight Period Profit Rate . . . . . . . . . . . . . . . . . .45 Input Indices. . . . . . . . . . . . . . . . . . . . . . . . .47 Adjusted Input Indices . . . . . . . . . . . . . . . . . . . . .47 Regional Trends in Input Use . . . . . . . . . . . . . . . . . .47 Input Use by Major Subgroup . . . . . . . . . . . . . . . . . . .52 Fixed and variable Inputs . . . . . . . . . . . . . . . . . . . .61 Chapter IV. (continued) Page Index of Farm Output. . . . . . . . . . . . . . . . . . . . . . 62 Output Trends . . . . . . . . . . . . . . . . . . . . . . . . . 69 Productivity . . . . . . . . . . . . . . . . . . . . . . . . 70 Outputs, Inputs, Productivity: A Comparison of the Three Measures . . . . . . . . . . . . . . . . . . . . . . . 73 Chapter V. The Mills-Boyne Procedure . . . . . . . . . . . . . . . . . . . 80 The Four Major Technologies . . . . . . . . . . . . . . . . . . 80 Chapter VI. sumry C O O O O O O O O O O O O O O O O O O O I O O O O I O O 95 P011cy Itnplications O O O D O O O O O O O O O O O O O O O O O O 97 Conjecture. . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Suggestions for Further Study . . . . . . . . . . . . . . . . . 98 Bib liogra phy O O O O O O O O O O O O O C O O O O O O O O O O O O O O 100 APPENDH O O O O O O O O O O O O I O O O O O O O O O O O O 0 O O O 0 1“ iv Table VOCDVOUI 10 11 12 13 14 15 16 17 18 19 2O 21 LIST OF TABLES The Weight Period Profit Rate . . . . . . . . . . Indices of Total Agricultural Inputs (including all land in fem) o o o s s s s 0 Indices of Total Agricultural Inputs (excluding land "voluntarily" retired by government programs). . Differences in Total Agricultural Inputs Caused by Adjustment for Land in Government Programs . . Indices of Farm Labor Input . . Indices of Farm Real Estate Input . . . . . . . . Indices of Mechanical Power and Machinery Input . Indices of Fertilizer and Lime Inputs . Indices of Feed, Seed, and Livestock Purchases. . Taxes and Interest Indices . . Indices of Miscellaneous Inputs . . . Indices of Fixed Inputs . . Indices of Variable Inputs . Total Farm Output. . . . . . . Indices of Total Agricultural Productivity . Allocation of Changes in Output Using the Mills- Boyne Procedure . . . . . Equation Equation Equation Equation Equation 4: 5 6. 7: 8 Regression for the Regression for the Regression for the Regression for the Regression for the C O O O O O O O O 0 United States . . . . United States . Lake States Region . Corn Belt Region . . Northern Plains Region . Page . 46 48 . 49 . 50 . 54 . 55 . S6 57 . 58 59 . 60 . 64 65 .67 .71 .81 .86 .87 .89 .90 .91 LIST OF FIGURES Figgre Page 1 The Productivity of American Agriculture, a comparison of various ‘8 timtes O O O O O O O O O O I O O O O O O O 8 2 Allocation of Output Change to Change in Inputs and Change in Productivity . . . . . . . . . . . . . . . . . 12 Labor Hours Comparison . . . . . . . . . . . . . . . . . . . l6 Hired Labor Hours Comparison . . . . . . . . . . . . . . . . 18 Derivation of "Product Added Plus Pasture". . . . . . . . . . 35 Derivation of Non-farm Value of Seed . . . . . . . . . . . . 37 Farm Inputs . . . . . . . . . . . . . . . . . . . . . . . . . 51 Fixed and Variable Inputs . . . . . . . . . . . . . . . . . . 66 Farm Output . . . . . . . . . . . . . . . . . . . . ._. . . . 68 10 Farm Productivity . . . . . . . . . . . . . . . . . . . . . . 72 QCDVO‘UIJ-‘w Output, Input and Productivity: 11 Northeast and Appalachian Regions . . . . . . . . . . . . . . 74 12 Lake States and Corn Belt Regions . . . . . . . . . . . . . . 75 13 Southeast and Delta Regions . . . . . . . . . . . . . . . . . 76 14 Northern Plains and Southern Plains Regions . . . . . . . . . 77 15 Mountain and Pacific Regions . . . . . . . . . . . . . . . . 78 16 The Four Major Technologies, United States . . . . . . . . . 83 17 The Four Major Technologies, Corn Belt Region . . . . . . . . 84 18 YClserved and Y Estimated, United States . . . . . . . . . . 88 19 Y Observed and Y Estimated, Lake States Region . . . . . . . 92 20 Y Observed and Y Estimated, Corn Belt Region . . . . . . . . 93 21 Y Observed and Y Estimated, Northern Plains Region . . . . . 94 22 Subdivision of the Livestock Feed Input . . . . . . . . . .154 vi CHAPTER I NEED FOR STUDY We can suppose that the first crude measures of productivity were simple measures such as milk production per goat. Such measures were valuable because they formed a basis for improving the herd. Down through the ages man has developed more and more sophisticated measures. Productivity indices are now used for a variety of purposes. Pre- diction of output forthcoming from a given quantity of inputs is one such use. Also, inputs required to produce a given desired level of outputs can be estimated with productivity measures. The economy's maximum capacity to produce is a measure desired during wartime situa- tions. Economic development is another area using productivity measures. The comparative rate of development between nations, geographical regions and over time periods are Of interest to development economists. Most previous studies of agricultural productivity have been limited to nationally aggregated indices. These indices are inadequate for solving many policy problems since interregional differences in invest- ment and productivity are unknown. Policy makers can more accurately formulate commodity adjustment programs if they know regional productiv- ities and their trends. Many problems requiring productivity estimates are of a regional or commodity nature. The application of national indices to these problems produce questionable results. Regional indices are needed for accurate regional estimates. The widening Spectrum of commercial vs non-commercial agriculture will eventually require another dimension in the matrix of indices, however, data for this separation are not yet available. OBJECTIVE The purpose of this study was to develop global or total productiv- ity indices for ten farm production regions in the 48 states for the period 1939 to 1970. Indices were constructed in such a manner that they could be Spliced onto existing U.S. indices which extend back to 1870. They were also designed to incorporate existing regional estimates of farm output. Thus the bulk of the problem was the compilation of regional input indices for the period 1939-70 and the necessary adjustments to make the input and output indices methodologically comparable. In view of the previous work already completed and the mass of data involved, a Laspeyres quantity index was deemed the most practical means of converting the various inputs and outputs to a common measure. 1947- 49 and 1957-59 were chosen as price weight periods with the index spliced at 1955. The reference base was set at 1967 - 100. In so far as the data permitted, factonsand products were entered in physical terms. The remaining items were deflated with the price in- dex or combination of price indices judged to be the most appropriate. A survey of the literature indicated that economists have arrived at differing estimates of productivity change. The discrepancies are caused by differences in assumptions, concepts and definitions. Thus a working definition is in order before proceeding. PRODUCTIVITY: Definition: Mach confusion surrounds the definition of the word "productivity". In an economic context, the term was first used to denote the average product of labor as per Peter Steiner: "Productivity is defined as output per unit of input. The almost universal measure of input is labor input, and physical output per man hour is the general measure of pro- ductivity and the one employed here." Peter Steiner, Review of Economics and Statistics, November 1950, pg. 321. Here Steiner is referring to a relationship such as equation (1) (1) 0t - f(Lt’Tt’ut) where t ' time 0t - index of output Lt - index of labor input Tc - technology ut - unexplained residual. This conception is deficient because it does not really measure the efficiency or effectiveness of labor since capital and other inputs are not held constant: e.g., we could replace one-half of the labor with machinery and output per unit of labor would double with no change in labor's effectiveness. This type of measure is more nearly a distri- bution function rather than a production function. Since the mid-fifties, economists have applied the term to other partial measures as well as to overall measures of output per unit of input: "The term productivity is frequently used loosely to denote the ratio of output to any related input or class of inputs. In this sense, there is a spectrum of productivity ratios, each of which indicates the savings achieved in particular cost elements over time as a result of changes in productive efficiency and factor substitutions. In order to attempt to measure changes in productive efficiency as such, however, output must be related to the aggregate of correspond- ing inputs. This is so because the proportions in which factors are combined usually change over time because of changes in relative factor prices or in technical knowledge." Here, Kendrick is discussing a relationship such as equation (2) (2) 0t . f(Lt’Kt’Tt’ut) where K - capital input. 1John W. Kendrick, "Productivity Trends: Capital and Labor", Review of Ecomonics and Statistics, Vol. XXXVIII, No. 3, (Aug., 1956). In a later publication Kendrick used the term "total factor productivity" to denote the overall measure. Currently, partial productivities other then labor are usually labeled as such by economists. However, the term productivity is still being used to denote both labor productivity and output per unit of input. Perhaps a better term for overall productivity would be "convert- ibility" or "transformability". For tltis study, the term productivity is defined as total output per unit of total input. SURVEY gr; LITERATURE The first global or overall measures of agricultural productivity were made by Barton and Cooper.2 These were annual U.S. estimates for the period 1910-1945. Barton and Cooper used a Laspeyres weighted aggregate formula. They experimented with two price weight periods, 1910-14 and 1935-39, concluding that the latter period had some advantages. Their study indicates that productivity of U.S. agriculture changed little during the period 1910-1922. There was a gradual but modest increase in productivity from 1922 to 1935. There was a more rapid increase during the last 10 years of the period, 1935-45. Loomis and Barton3 expanded and revised this series by extending the annual estimates through 1958 and by making decade interval estimates for the period 1870-1900. TwO price weight periods were used for this study: 1935-39 and 1947-49 with a splice at 1940. They found that productivity of U.S. agriculture divided roughly into four periods: 1870-1910 A period of extensification during which a large quantity of good cheap land was brought into production. Total inputs doubled during this period. Productivity increased 32 percent. 2Glen T. Barton and Martin R. Cooper, "Relation of Agricultural No. 2, (May, 1948), 117-126. 3RalphA. Loomis and Glen T. Barton, Productivity gqugriculture, Technical Bulletin No. 1238; U.S. Dept. of Agriculture, 1961. 1911-1940 All of the good cheap land was already in production when this period started. Most of the land added during this period was in the semi-arrid plains. The last decade was beset with the great depression. Total inputs increased only 18 percent during this period. Productivity increased only 15 percent. 1940-1950 This was a period of rapid adoption of technology and rapid increases in productivity. Total inputs increased only 4 percent. Productivity increased 18 percent. 1950-1958 During this brief period productivity increased even more dramatically: 23 percent. There was no change in the quantity of inputs used. Lave published several papers on the subject of growth models and technological change. Many of these papers were subsequently summar- ized in book form.4 He briefly outlines the growth models of Solow, Harrod-Domar, Abramovitz and Kendrick, Johansen and others. There is also a discussion of the aggregate production function and its relation- ship to growth models and indices of technological change. Lave attempted to measure technological change in American agri- culture at both the county, state, regional and national level. These were decade interval estimates for the period 1850-1960. However, the only inputs considered were labor and capital. GilbertS reviewed the problems of quality differences which came up in a previous study of international productivity. He distills out the relevant principles for application to time series studies. He concludes that an increase in output can only be defined unequivocally as an increase in the output of goods common to both the current and base years. He points out that "linking in" new goods should be done hLester B. Lave, Technological Change: Its Conception and Measure- ment. Prentice-Hall, New Jersey, 1966. 5Milton Gilbert, "Quality Changes and Index Numbers," Economic Development and Cultural Change, Vol. IX, No. 3 (April 1961), pp. 187- 294. with care due to the differences in production costs and prices between prototypes and mass production. Griliches6 studied the problem of quality change and its effect on productivity indices. He criticized the USDA index of automobile prices as not accounting for quality change. He also criticized the building cost index as being built up from separate indices of the cost of material and the cost of labor. This procedure misses the increased productivity resulting from improved technology. Griliches also points up the differences in wages reported depend- ing on whether you ask the buyer or seller and presents a method of separating out quality change in automobiles using multiple regression techniques. Ruttan7»8’9 co-authored three papers concerning agricultural pro- ductivity measurement. The other authors were Stout and Callahan. These three papers were apparently based on the same or nearly the same data. They are the most comprehensive attempt at developing a global productiv- ity index for U.S. agriculture by regions. The 1962 paper is vague as to what items were included in inputs and outputs. The 1960 paper states: (page 54) "This series is not an ideal measure of the value of commodities and services produced by farms since it in- cludes some double counting; interfarm sales of feed and seed, interstate sales of feeder and breeding livestock, and the value of inputs purchased from the nonfarm sector 6Zvi Griliches, "Notes on the Measurement of Price and Quality Changes", Studies in Income and Wealth, Vol. 28, Princeton University Press, 1964. 7Thomas T. Stout and Vernon W. Ruttan, "Regional Patterns of Technological Change in American.Agriculture", JFE, Vol. XL, No. 2, May,1958. 8Vernon W. Ruttan and Thomas T. Stout, "Regional Differences in Factor Shares in American Agriculture 1925-1957", JFE, Vol. XLII, No. 1, February,1960. 9V. M. Ruttan and J. C. Callahan, "Resources Inputs and Output Growth: Comparisons between Agriculture and Forestry", Forest Science, Vol. 8, No. 1, March,1962. which is of farm origin are all included in gross income. While adjustments for most of these factors can be made at the national level, it was not possible to adjust the regional data." This series of studies apparently included considerable double counting. The most comprehensive productivity measurements were made by Kendrick}o These measures were made separately and in total for 10 in- dustry groups of which one was agriculture, forestry and fisheries. Kendrick also made an extensive study of the problem of definition, alternative concepts, methods of measurement, data limitations, and the effects of business cycles. His estimates were Compiled with a Laspeyres weighted aggregative formula with weight periods: 1939, 1947-49 and 1954. Many of the meas- ures used were drawn or adapted from estimates originally made by Tostlebe, Strauss and Bean, and other researchers in the Department of Agriculture. Annual estimates were made for the Period 1869-1957. Comparison gg Estimates The agricultural productivity estimates by Loomis and Barton, Kendrick, Ruttan, Lave, and those made for this thesis were converted to a 1950 base and are graphed on Figure 1. THEORETICAL FRAMEWORK As indicated above, productivity means what it is defined to mean and there is more than one definition in current use. For this study productivity measurement hinges on changes in the rate of total output per unit of total input, both output and input being measured in constant dollars. It is concerned primarily with the relationship between factors and products. Such a measure is closely related to and is often used as a proxy for technological change. A more accurate measure would be of changes in the aggregate production function but here we run into problems. The 10JohnW. Kendrick, Productivity Trends in the United States, Princeton University Press, Princeton, 1961. _ o.=o_u .o:u=.x coco-.1 tone—on. eaten a 2:30.. use. as one. ov one. cu .zo. 00 can. on can. so coo. ~00. - Ono: .0 Les no _. anotu> Lo contuano e .o.=::o:o< eootoE . < .0 ar.>:o=voam o: . .P concept of the aggregate production function as advanced by Solow,11 has been a subject of controversy since its inception. The purists claim the problems of aggregating to the industry are formidable, aggregation to the total economy impossible. On the other side, there are economists that claim the aggregate production function is almost as legitimate a concept as the aggregate consumption function.12 Since Solow published his paper in 1957, there has been a wealth of literature relating productivity indices, indices of technological change and production functions.13 Most efforts to estimate productivity have centered on one of the following methods: I. Arithmetic aggregation: This approach has the advantage of being easier to compile, easier for the novice to comprehend. It implic- itly assumes a linear production function. II. Geometric aggregation: This approach weights the inputs with fac- tor shares rather than prices. It is implicitly related to the Cobb-Douglas rather than a linear production function. It appeals to many as being more realistic but is more difficult for the novice to interpret. Recently a refinement, the CES (constant elasticity of substitution) function has been developed. This function at- tempts to estimate the elasticity of substitution rather than assume it. This function is more difficult to estimate. For the purposes of this thesis, an index similar to that described by Domar14 (and used by Kendrickls) was considered to be most appropriate, 11Robert M. Solow, "Technical Change and the Aggregate Production Function". Rev. g£_Econ. & Stat., Vol. 39, OAug., 1957), pp. 312f. 12See Kenneth Arrow, H. Chenery, B. Minhas and R. Solow, "Capital- Labor Substitution and Economic Efficiency," 52!, 2;.Econ. & Stat., Vol. 43, Aug., 1961, pp. 225-40; Mirray Brown and J. Popkin, "A Measure of Technological Change and Returns to Scale," Rev. of Econ. & Stat. , Vol. 44, Nov., 1962, pp. 402- 11; Zvi Griliches, "The Sources —of Measured Pro- ductivity Growth: U. 8. Agriculture 1940- 60, " Jour. g£_Politica1 Economy, Vol. 71, Aug., 1963, pp. 331-46. 13See Lester B. Lave, Technological Change: Its Conception and Measurement. Prentice-Hall, New Jersey, 1966, pp. 13ff. 14Evsey D. Domar, "On Total Productivity and All That," Jour. 9; Political Economy, Vol. LXX, Dec. 62, p. 597-608. 15John W. Kendrick, Productivity Trends in the United States, Princeton University Press, Princeton, 1961. 10 in view of budget limitations, time constraints and the output work already completed. For a two input industry Domar's equation is (3) O " A(wa + 113K) where: A - Index of Productivity weight year or period I wage rate interest rate - Physical quantity of labor input “fit-“C Physical quantity of capital input This is known as the Laspeyres quantity index and it can be written as: 0 c 0b wac + ich where: c I current year Output per unit input is then compared to the reference base period in which both the numerator and denominator - 100. Outputs and inputs in the reference base will not be equal, the difference being profit. Thus the index, in a sense, compares the profit rate in the given year with the profit rate in the base period. In this abbreviated model only the two inputs labor and capital are indicated and only one price weight period is implied. The index compiled includes 87 separate items with two price weight periods: 1947-1949 for the period 1939-1954, 1957-1959 for the period 1955-1970. The method used to splice these series is discussed in chapter III. Measurement Problems In a strict sense, the index should endeavour to measure the ll productivity of that part of agriculture which transforms or creates goods on the farm. In a practical sense, it is not possible to measure the price of all items at the farm gate. Inputs are purchased at the market and products are sold at the market. Thus there is value added by transportation which is not within the confines of the farm as well as a host of other problems. The proliferation of integrated operations now adds another impedi- ment to price measurement: There is no clean cut transfer of ownership with a price consideration. Rather there is an agreement to share in some subsequent sale nearer the retail level in the marketing chain. Thus, the price received by farmers is, in the case of broilers, based on a diminishing proportion sold through the open market. Allocation gfIOutput Change £2 Change ig_lnputs §§g_Change i§_Productivity Following the procedure developed by Mills16 and Boyne}7 it is pos- sible to separate the change in output into two portions: that caused by changes in input use and that caused by changes in productivity. Refer- ring to Figure 2, let 11 I Input level at beginning of period 12 I Input level at end of period P1 I Productivity at beginning of period P2 I Productivity at end of period 01 I Output at beginning of period 02 I Output at end of period Then, if we hold the productivity level constant at P1 and increase in- puts from I1 to 12, the change in output caused by the change in inputs alone is PL£§I. If we hold the input level constant at 11 and increase productivity to P2, the change in output caused by the change in produc- tivity alone is II£SP. This leaves the residual [31(5P shown by the rec- tangle ab, which is caused by interaction between the change in inputs 16Fredreck C. Mills, "Productivity and Economic Progress," Occa- sional Paper 38, NBER (1952), pp. 32f. 17David H. Boyne, "Changes in the Real Wealth Position of Farm Operators, 1940-1960," Technical Bulletin 224, Michigan State University, 1964. 12 I b I 2 u ’7? . / I], a ——-T Figure 2 and the change in productivity. By arbitrarily dividing this residual into equal parts, we can allo- cate the total change in output between inputs and productivity. The part allocated to the change in inputs is then P1AI + AIZAP The part allocated to the change in productivity is 11AP + é—I-z-g- Alternatively we can reason that to move from a to b, we must start at a and end up at b. It is then assumed that the expansion path is a straight line. The next three chapters of this thesis detail the methods used to compute the many input accounts, the Splicing procedure, output computa- tion and productivity indexing. In chapter V are estimates incorporating the Mills-Boyne procedure and an effort to establish the relationship between productivity and the major agricultural technologies. The concluding chapter draws policy implications and outlines ex- pectations for the future. CHAPTER II INPUT COMPILATION A catalogue of the items included in the input index appears on appendix pages 104 to 106 . Details concerning data sources and the methods used to reduce these inputs to constant dollar expenditures also appears in the appendix. A discussion of associated methodological and conceptual problems follows: LABOR The labor input can be broken down into three general classes: Hired, operator, and other family labor. Since the labor of the opera- tor and other family members is not paid, it is difficult to establish a weight period price or wage for it. It seems reasonable to assume that the quality of hired labor and other family labor is roughly equal. How- ever, there is evidence which indicates that the operator's labor is of a higher quality than hired labor: a. Most operators who liquidate their farm operations take higher paying non-farm jobs rather than becoming hired men. b. Considering only the set including tenant operators and hired farm workers, a substantially larger proportion of tenants accumulate sufficient savings to purchase farms. These savings must come from one or more of the following sources: (1) More hours worked as a result of: (a) Longer work week, (b) Less unemployment. (2) Substantial returns to: (a) Investment in machinery and livestock, (b) Management, (c) Entrepreneurship. l3 14 (3) Higher returns per hour worked, or greater efficiency. c. Professional farm managers often have a choice of operating their client's farms as either tenant operations or "direct" operations. 0A direct operation is one run entirely with hired labor.) Although a professional management firm is selling more management with a direct operation, they seldom use this type of operation since it is usually less profitable. Many researchers apply the mortgage interest rate to the land value, account for the paid inputs, and label the residual as return to labor and management. This approach is defective since this residual includes profit or loss. Thus, the residual can vary with the quality and quan- tity of both labor and management held constant because of changes in weather or prices. If we rule out the residual as a measure of labor input, the only alternative remaining is the Opportunity cost approach. Here we have some difficulty in identifying the relevant alternative opportunity. Is it the hired farm*wage or the industrial wage adjusted for moving costs? The hired farm wage implies full employment. Thus, it seems likely that many underemployed farm operators would receive an average labor in- come less than that implied by the hired wage. At the other extreme, there are many operators who are capable of drawing industrial wages and do so when they leave agriculture. Few studies have been made to determine where the "leavers" go and whether they do better or worse in their off-farm alternative. Loomis];8 working with data from two counties in Southwest Michigan, found that part time farmers had a preference for farm work. These part time farmers were asked: a. How much annual income would you have to have from non-farmwwork before you would quit farming and b. How much annual income would you have to have from the farm before you would quit working off the farm? The difference between the two figures, $1,266, is an indication of the dollar value of the amenities which are associated with farming. 18Ralph A. Loomis, "Working in Two Worlds--Farm and Factory? Michi- gan State University Agricultural Experiment Station Research Report No. 32, 1965. 15 Because of these amenities, charging the family labor input at their industrial opportunity cost would result in an entrepreneural loss for many recent years. Thus, as a first approximation, a decision was made to charge the family labor at the hired farm wage. Inapection of the residual indicates little left over to justify a higher rate. Labor anntity There are three independent series measuring the quantity of farm labor. The Bureau of Labor Statistics measures the hours worked by those who work the majority of their time in agriculture. The Statistical Re- porting Service reports the employment of operators and hired men working one or more hours during the survey week and other family members working 15 or more hours. The Economic Research Service measures the labor re- quired to do the work performed. These measures are graphed on Figure 3. The differences between the series are believed to be caused by differences in concept and "stand- by" time. A detailed discussion of the labor data appears on appendix pages 107 to 110. For hired labor, it is possible to compute an additional measure by dividing the wage bill by the wage rate. These measures are graphed on Figure 4. The ERS quantities on this graph were derived with the proce- dure indicated in the appendix. It can be reasoned that the stand-by time of the family labor does not have to be paid and is thus not an in- put in the production function. It can also be argued that all committed inputs should be paid. For those items such as hired labor and machinery, this is a reasonable argument since the labor will not be available unless it is paid for standing by as well as working. The machine is paid for at purchase, thus it is paid whether it is used or standing by. However, this argument is not valid for the operator's labor. He does not pay him- self for his underemployment. The ERS series on labor required is the only complete one for the entire period. A decision was made to enter the family labor at the ERS level based on labor required. For hired labor, six percent was added to m muowwm ohma no 00 mm on we ~¢a~ lull II . unoamoneu a mum 0 II I l6 NH ca 0H ma ON Nu «use: .aaam aomwuumaoo mason momma 17 the ERS level to account for ”stand-by" time. This is the line labeled ERS + 6 on Figure 4. This general procedure is consistent with industrial indices since they treat labor as a variable input and thus do not pay directly for its unemployment. REAL ESTATE Two means were available to estimate constant dollar values for the real estate input: 1. Current dollar values of land plus buildings could be deflated with some appropriate deflator. II. The physical quantity of land could be multiplied by the weight period value per unit. The second method has theoretical advantages, but presents the problem that the only per acre values available by classes of land ex- clude building values. In either case, some means was needed to convert the constant dollar value of the stock of real estate into an annual flow of input. A decision was made to use method II, thus data were needed on: 1) The quantity of land, 2) The weight period per unit value, 3) The value of buildings, 4) Some means to convert the constant dollar value into an annual flow. The Quantity 2£,Lg§d_ A decision was made to use the Agricultural Census as a data series for land quantity since this was the only source available prior to 1950. Census land classifications have changed slightly during the period in- volved. There has been a general trend to carry through the sub-classi- fications with additional refinements and with changes in the headings and grouping of some classes. In order to develop the constant dollar value of the stock of land, it was necessary to develop land classifications which were consistent with the per unit values, and for the 17 western states, also consistent Bill. Hours Hired Labor Hours Comparison 3200 3000 2800 BLS Wage 2600 __ bill V\ | | ° A n o l 2400 t _ l V—+ ‘ ERS + 67. l ERS / D O 2200 A \ 1 \0—1 \0 ' SRS A 2000 J. 1965 1966 1967 1968 1969 Figure 4 19 with the rent/value ratios which were used to convert the stock into an annual flow of input. The method used and data series are discussed in the appendix. _Th_e_ 231413 gf Buildings In order to build up total land values by classes, it was necessary to use weight period per acre values which excluded buildings. The opera- tor's dwelling was not considered to be a production item, thus it was desired that it be excluded. However, the remainder of the farm buildings are production items and should be included as an input. Service build- ing values were available which could be deflated to obtain a constant dollar value. However, annual regional data were also available on the value of buildings as a proportion of total real estate value. These ratios permitted the calculation of a constant dollar value for land plus service buildings. An illustration of the equation used for this calcu- lation appears in the appendix. Conversion 2f the Stock g£_Land Into 32_Annua1 Flow 2; Input Many researchers use the mortgage rate of interest as a proxy for the earning rate of land. On a national basis this approach might be acceptable, however, for a regional index, the mortgage rate is defective because: 1. Both the seller and most prospective purchasers of farm land are dealing in a local market; however, the mort- gage rate of interest is determined in a national market. Loanable funds are quite mobile and go where the inter- est rate is highest. Hence, what little inter-regional differences there are in the mortgage rate can be attri- buted mainly to differential administration costs which are a function of the size of loans. Moreover, loanable funds do not confine themselves to the agricultural mar- ket in seeking the greatest return. Thus, the farmer is competing with industrial and commercial borrowers when he approaches the local lending agency. 2. It might be reasoned that a farmer can acquire land by borrowing the capital. Thus his cost of acquiring land is the mortgage interest payments. This would be true 20 if a farmer could borrow 100 percent of the purchase price. However, the lender discounts for risk, thus the bulk of the risk is borne by the owner's equity which should, ceteris paribus, earn a higher return for a greater risk. The mortgage rate of interest implies that the only means of ob- taining the use of land is to purchase it. Leasing is an alternative means of obtaining land services. The estimated rent is believed to be superior because: 3. The decision to buy or sell land may be influenced by factors other than value in production. Thus a buyer may pay more than the productivity value of land in order to: a. Gain status of landholder, b. Gain tax advantages, c. Retain control of birthplace, d. Speculate on: (1) General inflation of land prices, (2) Increase in value due to urban expansion, Live in the country, Live near friends and relatives, Be self-employed, D‘oov-nm . Gain job security. There are little data available on the aggregate rental value of all land. However, the rental can be estimated by multiplying real es- tate values by the ratio of rent to value. By this method we can not only convert the stock of land into an annual flow but we can also sort out some of the other factors which we wish to exclude. Since the tenant does not benefit from appreciation he will only pay rent on the productivity value (other things being equal), i.e., we let P I the productivity value and S I the extra payment made for speculation, then P + S I total land value. Let R I rent justified by productivity, then the tenant is willing to pay rent of R based on productivity value P or a rent/value ratio of R/P. However, the landlord has P + S invested in the land. Thus the ratio of 21 rent/value received by the landlord is R , a lower figure. Thus the P + S rent/value capitalization rate accounts for the extra investment which a buyer puts into land in order to speculate, or the income foregone by an existing owner to retain the property. In a similar manner the rent/value ratio accounts for the extra payment a buyer makes (or owner foregoes) in excess of the productivity value in order to: gain status of landholder, gain tax advantages, and retain control of birthplace. The rent/value ratio does a less perfect job of separating out ex- tra payments which an owner makes to gain some of the other amenities: country living, location near friends and relatives, self-employment, and job security, since, to a slightly less secure degree, a tenant can also enjoy these amenities and may pay a rent in excess of that justified by the productivity value. Another factor affecting land values is Government programs. A buyer might pay a premium because of the return possible from participa- tion in a Government program. However, a tenant would also presumably pay a higher rental for the same reason. If this factor has as equal effect, than both the numerator and denominator would be affected equally and the rent/value ratio would remain constant. However, there is no certainty that Government programs will continue in the long run. Since the buyer is concerned with the long run whereas rental contracts are typically for one year, it is likely that Government programs have a greater effect on rents than on values. This may be a partial explana- tion for the high ratios in the cotton and tobacco producing regions. Data sources and methods used to compile rent/value ratios are dis- cussed in the appendix, pages 117 thru 119. INTEREST ON REAL ESTATE MORTGAGES The procedures outlined for land above would give an accurate esti- mate of the real estate input providing the farmer owned a 100 percent equity. However, for that portion of the real estate which is mortgaged, the appropriate rate for converting from a stock to a flow is the mort- gage rate of interest. Thus, the constant dollar value of land plus service buildings was split into "equity" and "mortgaged" portions by 22 using the ratio of current dollar mortgages outstanding to the total current dollar value of all real estate. The equity portion was then converted with the rent/value ratio. Real estate mortgages outstanding were multiplied by the weight period average interest rate paid on real estate mortgages outstanding. The product is what it would have cost farmers to borrow the money if there were no change in interest rates. The interest which the farmers paid the lending agency declined in purchasing power with inflation. Thus the product above was deflated with the index of prices paid by farmers for items used in production, interest, taxes and wage rates. IAND ;§_COVERNMENT PROGRAMS Underlying the theoretical models for productivity measurement, are the usual assumptions of equilibrium in both the factor and product markets. Here we run into difficulty with government programs. We have chosen to use the Laspeyres quantity index: Output Index I Productivity Index I —— Cch Input Index I ---- Cbe where: the P's are output prices, the Q's are quantities, and the C's are input costs. Government payments pose a problem since they are an income but not a product. On the physical production function, there are inputs com- mitted in order to gain the subsidy but on the output side there is no physical counterpart. To the extent that Government payments are trans- fer payments, it would appear that there is no corresponding expense. However, there is a tendency for the potential profit from subsidies to be capitalized into the land values. Since there is no certainty that the programs will continue indefinitely, the potential profit will be heavily discounted for this risk. However, this discounting will have little effect on rents since they are for a relatively short term, 23 especially cash rents. Since the ratio of net cash rent to value was used to convert the stock of land into an annual flow, the rents will be inflated for most of the expected profit from the subsidy. Thus, for the weight period, the inputs will match the outputs if we include gov- ernment payments as outputs. Over the entire period, this procedure is defective however, since part of the payments have been made as price supports and part as direct payments and there has been a secular change in the proportions of each. It would be exceedingly difficult to separate out the benefits of the price support programs since they affected the free market prices. A decision was made to do the analysis both ways, with and without an adjustment for government programs. These results are discussed in chapter IV. The reasoning used for the government program adjustment follows: For part of the government programs, the farmer is required to idle certain resources in order to gain price supports. For these inputs, we can reason that the inputs committed to the programs are balanced by the increased value of the output. For some of the government programs, the farmer can choose voluntarily to retire additional resources (mainly land) in order to gain a direct government payment. This payment is generally in excess of the fair rental value of the land retired. Thus a weight period payment rate was computed by dividing the weight period payment (for the land voluntarily retired) by the acres so retired. This weight period payment rate was then applied to the acres voluntarily retired for the entire period. These estimates were then subtracted from the real estate input. DEPRECIATION Current dollar estimates of depreciation were obtained from the Farm Income Branch, Economic and Statistical Analysis Division, Economic Research Service, U.S.D.A. (subsequently abbreviated as F.I.B.). These estimates are compiled as the estimated outlay, in current prices, which would be required if farmers were to replace the plant and equipment used up during the year. The estimates are based on a "declining bal- ance method" in which a constant percentage representing the annual rate of depreciation of each type of capital is applied to its estimated 24 value at the beginning of each year» A more detailed discussion of these estimates appears in the appendix. The price indices used to deflate these current dollar estimates follow: Depreciation on service buildings and other structures: The F.I.B. index of service building construction costs. Automobile depreciation: The F.I.B. index of prices paid for new automobiles. Tractor depreciation: The F.I.B. index of prices paid for new tractors. Truck depreciation: The F.I.B. index of prices paid for trucks. Depreciation on other farm machinery: The SRS index of farm machinery prices. INSURANCE AND ACCIDJEfiNTAL DAMAGES 19 SERVICE BUILDINGS AND MACHINERY Referring to the diagram: Let A + B I insurance premium payments, B + C I accidental damages, B I accidental damages covered by insurance, A I premium payments less claims (or net insurance premiums), C I uninsured accidental damages. If farmers did not have insurance they would be out B + C. By buying insurance they are out an additional amount A since premiums (typically) exceed indemnities. . Current dollar estimates for this input were obtained from the Farm Income Branch, ESAD. These estimates were compiled as net insurance premiums and total losses. The net insurance premiums were deflated 25 with the SRS index of prices paid for building and fencing materials. Total losses were deflated with the FIB index of farm service building construction costs. MAINTENANCE Q; SERVICE BUILDINGS AND MACHINERY Current dollar estimates were obtained from the Farm Income Branch. These estimates cover repairs and maintenance of farm service buildings, and repairs and maintenance of motor vehicles and farm machinery. A detailed summary of the method of compilation appears in the appendix. The price indices used to deflate these estimates follow: Repairs to service buildings and other structures and land improvements: The SRS index of prices paid for building and fencing materials Automobile repairs, parts, and tires: A composite index made up of 0.66 times the BLS index of prices paid for auto repairs and maintenaangplus 0.34 times the BLS index of prices paid for auto tires. Tractor repairs, parts, and tires: A composite index made up of 0.91 times the BLS index of prices paid for auto repairs and maintenance plus 0.09 times the BLS wholesale tractor tire price index. Truck repairs, parts, and tires: A composite index made up of 0.70 times the BLS index of prices paid for auto repairs and maintenance plus 0.30 times the BLS wholesale price index of truck and bus tires. Other farm machines: Repairs, parts, and tires: A composite index made up of 0.98 times the BLS index of prices paid for auto repairs and maintenance plus 0.02 times the BLS index of prices paid for auto tires. 19During the war period and for some of the earlier years of the period, BLS figures were not available. SRS prices were used to bridge these gaps in BLS figures. About 1/3 of the expenditures for the automobile repairs, parts, and tires account was for tires, thus the indices were weighted 2/3 for repairs and parts and 1/3 for tires. Similar reasoning was used for weighting the indices for farm machinery. 26 GRAZING FEES Fees for grazing on public lands were compiled as physical quanti- ties times weight period prices per unit. For most classes of public lands, the unit was acres, for National Forest land, the unit was animal unit months. Mast of the fees on public lands are set at less than the fair mar- ket value. In most cases, the difference between the fee charged and the fair market value is a subsidy which is capitalized into the value of the adjoining ranch and thus shows up as land input there. Machinery: For farm machinery and motor vehicles, national total current dollar inventory values as of 1 January were available from the Farm Income Branch, ESAD. The current year 1 January figures were averaged with the 1 January figure for the following year to approximate an annual average. For tractors, trucks and the production share of automobiles, the nation- al total figures were distributed to regions with the number of machines or vehicles on farms. To date, there are only two observations on the distribution of other farm machines to regions: 1) the 1949 Census, 2) the 1955 Survey of Farmers Expenditures. The 1949 distribution was used for the years 1939-48. A straight line interpolation of the proportional distribution was used for the period 1949-55. The 1955 distribution was used for the period l956-date. The following price indices were used to deflate the current dollar estimates: Automobiles: The FIB index of prices paid for new automobiles, Tractors: The FIB index of prices paid for tractors, Trucks: The FIB index of prices paid for trucks, Other farm machines: The SRS index of prices paid for farm machinery. 27 Livestock and Grain Constant dollar inventory values for these items were compiled by multiplying current quantities by weight period prices per unit. For livestock, the total 1 January inventory value was compiled. For grains and forages, it was reasoned that inventories were near a peak on 1 Jan- uary, thus a decision was made to reduce these figures by one half to approximate an annual average quantity of grains on hand. Operating Capital A part of the demand deposits held by farmers are no doubt intended for consumption rather than production. It is reasonable to assume that this is likewise true of currency holdings. Since there are no data available on currency holdings, it was assumed that currency held for farm production purposes was equal to the demand deposits held for con- sumption. Thus the total operating capital is equal to demand deposits. Data on demand deposits held by farmers were obtained from the Agricul- tural Finance Branch, FPnfl. They were distributed to regions with the distribution of total production expenses from the Farm Income Situation. For the period 1939-48, total production expenses were not available. Thus, the 1949 relationship between production expenses and total cash receipts was applied to total cash receipts for the period 1939-48. The distribution of this derived production expense was used to distribute demand deposits. Opportunity Qp§£_p£_Capital Invested £p_Inventory and Operating {ppgg_ We were able to estimate the rate of return which farmers receive on their equity in land by measuring the ratio of rent to real estate value. Now that we have a return to land, is there any reason to sus- pect that farmers receive or expect a different rate of return on their other investments? In the past, researchers have customarily used the mortgage rate for converting land stock to flows and the short term in- terest rate for converting other forms of capital. We have rejected the mortgage rate for converting the land stock. Now how does the chattel mortgage rate fit as a means of converting inventory stocks into an an- nual flow of inputs? 28 Chattel mortgages carry a higher interest rate than land mortgages because: the loans are smaller, have a shorter term and thus the admin- istration cost per dollar loaned is higher; the collateral is movable, thus there is greater risk of default with little means of recourse. Now these factors that determine the Chattel rate have no relation- ship to the rate Of return the farmer receives on the investment in question. The figure we are trying to approximate is the rate of return that farmers receive on operating capital. We cannot estimate this di- rectly since farmers do not separate their return to: investment in real estate, investment in operating capital, labor, management, entre- preneurship. TO approach the problem by stages, let us compare, to an owner operator, a tenant operator who rents for cash. In discussing the land input, we concluded that a tenant operator may pay a rent greater than is justified by earnings in order to gain some Of the amentities that acrue to a farm operator, i.e., the status that results from being a self employed businessman, country living, a location near friends and relatives. A farmer who owns his own farm gains additional amentities: landholder status, control Of birthplace, additional job security, full control Of management Of farm operation and farm residence, land as a vehicle for speculation. (A tenant operator may speculate on feeder cattle, however, he cannot expect a long term capital gain such as a landowner can.) Thus we can reason that a farmer would expect a somewhat lower rate Of return on his investment in land than he would expect on his invest- ment in operating capital. However, an investment in operating capital is more flexible than an investment in land, i.e., a tenant farmer has less asset fixity than an owner and can get in and out of business with less transfer costs. Thus the rate for which we are searching should lie somewhere be- tween the rent/value ratio and the chattel mortgage rate. In view of the data available, a decision was made to use the interest rate on real estate mortgages as a proxy for the Opportunity cost of equity capital invested in inventory. 29 Interest on short term debt was handled differently from real es- tate mortgages in order to simplify computations. For the real estate input, it was fairly simple to divide this single input into an equity and a mortgaged portion. But for non-real estate capital, there were 18 items and it would be both laborious and fictitious to allocate non-real estate debt to each item, so the problem was approached indirectly by reasoning: Let L I the cost to the farm of acquiring the use of equity capital. The real estate mortgage rate was assumed to be a good proxy for this opportunity cost. interest rate paid on non-real estate loans. total value of non-real estate capital. quantity Of short term capital borrowed. )< *6 >< to I - P I equity portion of non-real estate capital. r: A >< I P) I Opportunity cost on equity portion. SP I interest paid on non-real estate loans. L(X P) + SP I total interest on non-real estate capital. We can more easily Obtain the same results by taking the long term rate across the total non-real estate capital and adding on the difference between the long and short term rates times_the principal borrowed, i.e., LX + (S - L)P I L(x - P) + SP The term (S - L)P was then labeled in the input list as "Interest added by non-real estate debt". Data were not available on P, the quantity of short term capital borrowed. The Farm Income Branch did have data on the current dollar interest paid. The principle can be estimated by dividing interest paid by the interest rate, i.e., P I SP/S Data were also not available on the interest rate being paid on short term loans outstanding. An assumption was made that the current short term rate lagged one year would be a reasonable estimate of the rate for loans outstanding. The rate of interest was defined to be a price, thus additional manipulation was required to get S on a weight period basis. A mathe- matical note illustrating the method used appears in the appendix. Also 30 in the appendix is a discussion of the Farm Income Branch's estimates on interest paid. The method described above derives the amount Of interest that the farmer would have paid the loan company if there had been no change in interest rates. It was still necessary to make an adjustment for changes in the purchasing power Of the money paid, thus the figures derived were deflated with the SRS index of production costs. FUEL AND OIL Physical data on fuel and Oil consumption were available only for the years 1947, 1948, 1953, and 1959. For that reason, a decision was made to deflate expenditures available from the Farm Income Branch. Inspection of regional price movements indicated that a national price index would not be appropriate since prices moved in different directions in different regions. Thus, regional price indices were constructed based on the tank truck price Of regular gasoline as re- ported by the SRS. Price data were skimpy for the early part of the period. The method used to fill in the void is discussed in the appendix. CUSTOM WORK Except for cotton ginning, there was no series available on custom work. TO the extent that custom work is done by farmers, the input has already been compiled by its components, i.e., the labor, fuel, depreci- ation, maintenance, etc., have been accounted for. However, that portion of custom*work done by non-farmers is an input which should be considered. The data were Sparse on this input. A detailed summary of the means used to build up the estimates appears in the appendix. OTHER INPUTS IN_THE MECHANICAL POWER.AND MACHINERY SUBGROUP Expenditure data were available from the Farm Income Branch for the remaining input items in this subgroup. These items were deflated as follows: Vehicle licenses: The SRS index of prices paid by farmers, including interest, taxes, and wage rates. 31 Vehicle insurance: The SRS index of motor vehicle prices. Blacksmithing, hardware, and small hand tools: The SRS index Of prices paid for farm supplies. Electricity: An index constructed from the price per KWH paid by farmers for electricity. Harness and saddling: An index constructed from the price of horse collars. Details concerning these compilations appear in the appendix. FERTILIZER AND LIMESTONE Fertilizers are sold as either straight materials, such as ammonium nitrate, superphOSphate and potash, or as mixtures of straight materials which are called commercial fertilizers. Price data are available only on some of the more popular analyses Of commercial fertilizers. Price data were available on most Of the straight materials. The Production Resources Branch, ERS, maintains a data series on the physical quanti- ties Of primary plant nutrients (N,P,K) applied. Thus a decision was made to value all of the straight materials, whether they were applied as such or as commercial mixtures, and add on the cost of mixing the part that was mixed. A preliminary attempt was made at determining the weight period cost Of mixing the primary ingredients into commercial fertilizers by taking the price Of straight materials (weighted by quantities used) times nutrient quantities. This gave the cost at the farm level, that is, what farmers would have paid had they purchased all their fertilizer at a straight material price. This was then subtracted from the Farm Income Branchs' total expenditures to get at the difference which would represent mixing costs. Substantial negative mixing cost in the Corn Belt and Lake States, indicated that this method would not be appropri- ate. The difficulty was believed to lie in the method by which the Cen- sus asked the question concerning fertilizer expenditures. The Census indicates that some Of the answers may have been net of government pay- nents. This was apparently the trouble since an analysis by states re- vealed that the discrepancies were in those states which use large 32 quantities Of rock phosphate. The method used was as follows: The composite fertilizer grade was calculated for each region, then a grade as nearly the same as possible was found for which a price was available. Mixing cost was then assumed to be the difference between what farmers paid for this grade and what they would have paid if they had purchased the equivalent quantity Of nutrients at straight material prices. Nonfarm use was estimated and subtracted from the total. Details concerning the procedures used for fertilizers and limestone appear in the appendix. FEED AND SEED Special Problems Associated with Farm Produced Inputs Of all the inputs compiled, the feed and seed inputs presented the most problems both: a. conceptually b. concerning gaps in the data or the lack of data entirely c. and concerning major descrepancies in the data. Conceptual Problems A small portion of the crOps produced is subsequently used for seed. A large portion Of many crops is used for feed for livestock. These items require special handling since they are both outputs and inputs. For example, suppose the utilization of the oat crop is: 2 percent is used for seed 80 percent is used for feed for livestock 18 percent is used for food and industrial uses. If we account the livestock produced as an output, and also the oats produced as an output we sum to more Output than was available for consumption. To avoid this double counting we have to deduct the inter- mediate products. That is, the livestock and crop output available for consumption is equal to livestock output plus crop output minus the crOps used for feed and seed. Another complication which enters the feed and seed input is the 33 processing by "middlemen". In the case of Oilseeds, the farmer sells the beans to a processor who separates the Oil from the meal. The proc- essor will in turn sell the meal to a formulator who mixes the meal with other ingredients and sells this back to the farmer. The value added by processing and formulating is a non-farm input. However, an additional complication arises from interregional trade. The oilseeds may be pro- duced in one region but used for feed in a different region. To measure directly the value added by transportation and handling to grains that move in interregional trade, we would have to know where the grain came from. Since complete data on interregional grain movements were not available, a search was made for an alternative means Of measurement. Two alternatives were available: 1. By making numerous simplifying assumptions, a crude approximation to interregional grain movements might be simulated with a linear programming tranSportation model. Fox20 made such an analysis for the period 1939-1950. The coefficient of determination between regional prices estimated by the model and actual prices averaged 0.88 for the period 1939-48. How- ever, the r2 was only 0.28 for the drought year (feed- ing year) 1947-48. 2. The second alternative assumes equilibrium in the feed and seed supply industry. Using feed as an example, as diagrammed in Figure 5, suppose 100 bu. of grain was shipped from the Corn Belt region to the Northeast region. The price received by farmers for grain in the Northeast region must be equal to the price re- ceived by farmers in the Corn Belt plus the cost of transporting the grain from the Corn Belt to the Northeast region. If the price in the Corn Belt were higher, then the Northeast feeder would purchase grain locally and there would be no movement. If the price in the Corn Belt were lower, then it would be 20 Karl A. Fox, "A Spatial Equilibrium Model Of the Livestock-Feed Economy in the United States". Econometrics, Vol. 21, NO. 4, pp547-566. 34 more economical for all feeders to import their grain and locally produced grain could not find a market at the higher price. Thus, if there is equilibrium, when we deduct the value of feed fed from livestock produc- tion, if we value the feed at the price received by farmers in the region where the imported feed is fed, we are deducting the exact amount paid to Corn Belt farmers for the feed plus the cost of transporting it to the Northeast region. Of course the feeder in the Northeast region may pay more for grain then the price received by farmers for grain. He would not if he bought his grain from neighboring farmers, however, if the grain moves through the local elevator, there will be a local handling charge. Interregional trade in seed introduces yet another complication: differences in quality. Much of the leg- ume seed produced in the midwest comes from hayfields that were accidently not needed for hay production. Much of the high quality certified seed is produced in the Pacific region. Thus the price received in the Corn Belt region for common seed is lower than the price received in the Pacific region for certified seed plus the cost of transportation to the Corn Belt. This problem is taken care Of in the process of computing the value added if we compute this as the dif- ference between the price received and the price paid in the region where planted. For example, assume the Corn Belt region plants three bushels of alfalfa seed (Figure 6), one bushel being certified seed imported from the Pacific region, the remaining 2 bushels being produced locally. Assume further that all the seed is processed in the Corn Belt. Thus the value of the certified seed after shipment to the Corn Belt is equal to the price paid to Pacific re- gion farmers ($30) plus the cost of transporting it to the Corn Belt, ($5). Assume further that the price paid to 35 :Ousummm wean momma uosmoum: modded osao> voosmoua daemon HO O5Hm> .eouwom m2 one nu magnum Hana one Ou uouo>uao Odom nuoo HOOOA .>OHO .auoa me $382. Eu eouunuuoamewuu mo eases new 1+1 03am> com» Assamese u .=m\na.aw um eouwom Hz Ou ueoamwem uuuww madam mo .an cod 08mm onu mo xuoo Commando Boom ousuw one mom Ou moufiavou nuoum>oao amnuauou he newsman: use nouuouuomoemuu we once sou +.uaom nuoo emu ow nuoauwm an mo>wooou oouum osu Ou Hoses mu m2 one a“ muoaumm he mo>uouou sound an» away meadows mash .muoaumw he mo>uooou Duane an acumen mz ea ewmum madam Auz cue unsound mo umoo .on\oueou ca as mass Imam: owmum HmOOH mo umoo .Ouo ..umeam .Oo> .uoeoa mo umoo eowumuomo weamoom co uawoum n ouswwm usaa> ace” I .=a\aa as Damn cuoo Boom macaw .sn 00H 36 Corn Belt farmers for common alfalfa seed is $20/bu and the cost of processing is $10/bu. Thus the total cost of the three bushels of seed is $105 or an aver- age price paid per bu. of $35. This cost is broken down as follows: 2 bu. of locally produced common seed $40. priced at the price received by farm- ers ($20/bu) l bu. Of imported certified seed priced 30. at the price received by farmers in the exporting region ($30) Cost of transporting imported seed 5. Cost of processing and local handling 30. at $lO/bu. $105. Of this total, the intermediate products produced by Corn Belt farmers are the 2 bu. of common seed pro- duced locally with a farm value of $40. Thus from the viewpoint of the Corn Belt, the value added by non-farmers is $105 - $40 or $65. We can arrive at this $65 value added in- directly as follows: Call the value added the difference between the price paid and the price received by Corn Belt farmers: $35 - $20 I $15/bu. times the quantity planted (3 bu.) gives a simulated value added of $45. When we mark Off the inter- mediate products, we use the quantity plant- ed rather than the quantity produced. Thus we mark off 3 bu. at $20/bu. or $60. Since this is $20 more than the value of intermedi- ate products actually produced in the Corn Belt, the total charged to Corn Belt input is $45 + $20 I $65 which is equal to the $65 which we arrived at directly. For the purposes Of this analysis, it is not nec- essary to know in which region the seed was processed 37 .=e\mn A mos» ca» m A omw oew om» u once ammuo>< I HQUOH cow maaaecue Amoco one maumuoooum we ON OH n one come nauseoum maawOOH new maidens: awOoH one wcwnmoooum moon vouuomaa now weaamewe HOOOH use moumoooowm moon vouuomaa wouuuoaonmua moon mouuomaq pom uauahom we» a as - nos» as» u oNA on .se N "aaomxmoum “mooamoum moon needs mean you nousuamoonxu "eowuusvoum mo oaum> "moon now mousuqmooaxm m» u mewuuogaa mo Deco mod» I mm» um .3n n UGOO on» umoo ueouoouwca weauuomaw mw 1+ on« Hanamauo u ufiom euoo Ou uuonuemuu nouns as» umOU uauwoouwCH .oea u .=e\o~» on mean suammas coesOO couao .oo N uOu muosusm Odom euoo he po>qouou Ooaum moon mmauwam ooawwuuoo .:o a mean no OOHm> .omw u .=e\onw um moon emammam moamuuuoo anode .:o H you dogmas ouuwomm aw museumw he mo>aooou ouuum ON” I .=n\o~w um madamemn HmOOH mom weummoooum mHHmOOH mousvoum moon :OESOO Duos maosnso N eowwwm Damauwm any aoum nuances“ moon moamguuoo ass on H "Damn cuou onu ea woodman mums moon endowed mo .3n n o ouawwm "Casan< 38 since we wish to charge the processing costs to the region which plants the seed. To be accurate a local handling cost was added since the above analysis prices the seed deducted from production at the price received by farmers. A discussion of the data available and the method of compilation appears in the appendix. THE FEED INPUT Expenditures figures were available on feed purchases, however, they could not be used because: 1. Until the 1964 Census, the Census questionnaire did not separate: a. Grains from supplements b. formula feeds from feeds fed as such c. purchases from neighbors from purchases made through commercial channels. e.g., the question from the 1950 Census was worded as follows: How much was spent last year for FEED for livestock and poultry? (include cost of grain, hay, mill feeds, concentrates, and roughages; also amounts paid for grinding and milling feed.) 2. Even if 1 were available, it would still be necessary to reduce the purchased feed input to physical terms in order to determine the farm value Of the raw materials. The Data Gap After an extensive survey Of data available on the feed input, it was obvious that an exhaustive data search would have to be made to ar- rive at regional feed input figures having any acceptable degree of ac- curacy. Thus, approximately 180 publications, including all those in the National Agricultural Library with feed statistics were searched for data. Cardex files were made for 104 of these publications to facilitate data retrival. 39 This exhaustive search turned up reasonably complete regional data which could be used in the form presented only for the one year period 1 October 1949 to 30 September 1950. A discussion Of the means used to simulate the missing data follows: Feed Grains, proportion going through commercial channels. An abortive attempt was made to estimate these proportions by starting with the following procedure: Let q I Feed grains fed (from Production Resources Branch, FPED, Statistical Bulletin 337 and unpublished worksheets). w I Feed grains fed on farms where grown (from SRS, Statistical Bulletins 115, 208, 311 and 404). Then q - w I Feed grains purchased from all sources. However, this procedure produced negative purchases for the 1957-59 period for two regions, the Lake States and the Northern Plains. In an effort to pinpoint the source Of the discrepancy, a tabula- tion by states was made for the 1964 crop year. This resulted in nega- tive purchases for the states Of Michigan and North Dakota and positive purchases which were Obviously too small for many remaining states. A conference with the Feed Grains Branch, SRS, indicated that the figures for "Feed grains fed on farms where grown" were obtained by subtracting what farmers report as sold from what they report as produced. These data were obtained from questionnaire C.E. 2-308, a copy of which ap- pears on appendix page 140. Labeling this residual as "Feed fed on farms where grown" is sus- pect for the following reasons: I. No account is taken Of change in stocks. II. There is an implicit assumption that farmers report grains under CCC loan or purchase agreement as "sold or to be sold". The discrepancy suggests that this assumption is unwarranted. Legally, liquidation Of grains under CCC loan is a foreclosure Of chattel mortgage rather than a sale. Purchase agreements also lack one Of the two requirements Of a legal sale contract: The CCC offers to buy but the farmer does not accept the Offer at the time the purchase 40 agreement is signed. Thus the purchase agreement resembles an option to sell rather than a sales contract. III. If a cash grain farmer is asked "How much grain was sold or will be sold?" This is tantamount to asking him "How much will your income be?" If this question is asked by the USDA, a Federal agency, how is the farmer to know that the Internal Revenue Service, an- other Federal agency, will not review his report? Thus it is prudent to expect some grain farmers to make a conservative report, especially as to their anticipated future sales. A Statistical Bulletin 268, "Grain Transportation Statistics for the North Central Region" estimated the sales of feed grains by country elevators to farmers for the Calendar year 1958. Thus this year was chosen to attempt an adjustment that would arrive at a plausable estimate of feed grains used for feed on farms where grown. A discussion of the means used to manufacture the missing data appears in the appendix. HIRED TRUCKING, FREIGHT AND EXPRESS A data series was available on the cost of hauling milk but no series were available for other items. Data for the single year, 1955, were available from the survey of farmers expenditures. Thus for all items except milk, an assumption was made that the 1955 relationship be- tween hired transportation and the value Of farm output was a constant that could be applied to all years. The method of computation is detailed in the appendix. POULTRY PURCHASES Physical quantity data series on the numbers of baby chickens and turkeys purchased were Obtained from the Farm Income Branch. The aver- age weight period price paid was obtained by dividing expenditures by the quantity purchased. Both the chickens and turkeys were divided into two classes: Laying chickens and broilers and heavy turkeys and light turkeys. 41 OTHER INPUTS l§_THE FEED, SEED AND LIVESTOCK SUBGROUP Expenditure data were available from the Farm Income Branch for the remaining input items in this subgroup. These items were deflated as follows: Milk hauling: A composite index including farm wage rates, auto repairs and truck tires. Livestock marketing: The SRS index of items used in produc- tion, interest, taxes and wage rates. Details concerning these compilations appear in the appendix. REAL ESTATE TAXES Data on this input are for taxes levied rather than taxes paid. A discussion with Thomas Hady, Economic DevelOpment Division, ERS, indi- cated that there was a secular trend for the states to require payment the same year the taxes are levied. Thus this input was compiled for the year levied although it must be recognized that for some states for some years, expecially the early years, the taxes may not have been paid until later. The tax figures available included taxes on the dwelling. To re- move dwelling taxes, the ratio of the values: land + service buildingg: land + all buildings was applied to the total tax figures. These current dollar estimates were deflated with the Commerce De- partment's implicit price deflator: "State and Local Government Pur- chases of Goods and Services". Data sources for this input are discussed in the appendix. PERSONAL PROPERTY TAXES Data for the years 1939-1959 were obtained from the Farm Income Branch. For subsequent years, the data were obtained from the Community Facilities Branch, Economic Development Division. These current dollar expenditures were deflated with the same index used for real estate taxes. 42 DAIRY SUPPLIES Current dollar expenditures for this item were obtained from the Farm Income Branch. For a deflator, for the period 1939-1954, an index was constructed based on a simple average of prices paid for milk pails and milk cans. For the period 1955 to date, it was reasoned that most dairy farmers were using bulk cooling equipment, thus a decision was made to base the price index for this period on the price of laundry detergent. PESTICIDES Because of the many kinds of pesticides and a lack of adequate quan- tity data, a decision was made to deflate expenditure figures. Dr. Shepard, ASCS, was consulted regarding the construction of a price index. Data sources and details of computation appear in the ap- pendix. COTTON GINNING The following figures were obtained from the Farm Income Branch: a. Expenditures for cotton ginning for the weight period years. b. The number of bales of cotton ginned for all years. From these figures were derived the weight period cost per bale by re- gions. These rates were then applied to the number of bales ginned to obtain expenditures in weight period prices. guano»: OPERATION AND MAINTENANCE cos'r Current dollar expenditures for this item were obtained from the Farm Income Branch. For the period 1949 to date, the index of "Irriga- tion Operation and Maintenance Cost" compiled by the Bureau of Reclama- tion, was used as a deflator. Since this index was not available for earlier years, a decision was made to use the Engineering News Record index of construction costs for that period. These data sources are discussed more fully in the appendix. 43 OTHER INPUTS IN_THE MISCELLANEOUS SUBGROUP Expenditure data were available from the Farm Income Branch for the remaining input items in this subgroup. These items were deflated as follows: Crop insurance: The index of crop values. Containers: An index of prices of selected items including baskets, bags and crates. Binding materials: An index constructed from the cost of baler twine. Veterinary services: The SRS index of prices paid for farm supplies. Telephone services: An index constructed from the base rate for local service. CHAPTER III INDEXING PROCEDURE After the various inputs were compiled into constant dollar esti- mates, the two periods, 1939-1954 and 1955-1970 were Spliced at 1955 to put the entire series on a 1957-1959 price weight equivalent basis. The splicing procedure is illustrated as follows: Let P = price Q = quantity 3 = splice year c = current year E = early period L 8 late period Then: QBPL . Q P a Q P ---- c E L QsPE c These Splices were made separately for each Farm Production Region. For the indices of individual input items, the splice was made at the indivi- dual input level. For the total input index, the Splice was made for all inputs combined, i.e., ___§gs:r. - w. , w. s E For the indices of major input subgroups, the splice was made at the sub- group level. The year 1967 was used as a reference base, i.e., the 1967 expendi- ture was arbitrarily set equal to 100 and other years were eXpressed as a percent of the 1967 expenditure. The output values were indexed in a similar manner. 44 CHAPTER IV sEs_UL_'.r_8_ In order to gain a longer historical perspective for part of the analyses which follows, the new national series was spliced onto the existing national series compiled by Loomis and Barton and maintained by Don Durost. This series is weighted with 1935-39 average prices. The Splice was made at 1939 using the technique outlined in chapter III. Ihg_Weight Period Profit Rate As a broad check on the accuracy of the compilations, a comparison was made of the total income and total expenses during the weight periods. If equilibrium prevailed, we would expect income and expenses to be equal assuming a normal profit. A normal profit implies opportunity cost on equity capital and family labor and management. The indices were constructed assuming the average rental rate on real estate and the mortgage rate of interest on non-real estate capital. Labor pres- ented a problem in that it was difficult to identify the relevant alter- native opportunity. As a preliminary estimate, a decision was made to enter the family labor at the hired rate. Inspection of the residual would then suggest whether a higher rate was indicated. The results were compiled both on an aggregate basis by regions and on a per farm basis. The results appear on the page following. The early weight period results (1947-49) indicate a modest profit for five regions, a meager profit for four regions, and a loss for the Pacific region. For Pacific region farmers this indicates that, on average, their labor returned them $164 less than their hired help. Of course they could live partially on their return to capital and this is appar- ently what many Pacific region farmers did do. Over the entire period, there was also substantial capital gains which this analysis does not include. For the 1957-59 weight period the results were even worse. The Northern Plains farmers made a modest profit, Corn Belt and Southern 45 445 mNN- um»- mm- AN mNN- SAN- now. «No mc~ use- sass- assasoa .a».« sea sauces uuuuo>< mom- was- we- as Hos- mas- can- mom «NH sad- ass. a .aaa ..u>¢ an-~na~ ".uauoum Nao.on mean anau amAN nuns swo~ om- mama Hess ss~m oasm » .Haa ..a>- an-hnm~ "asses“ aso.m~ wn~m sNHN Hana «has Essa oosu szn sass omen ms- m .Haa ..o>a an-~mma unassuso coauom u «03 anasnoa oak ass- sum was ~o~ mus ass «mun sews o~oa and suaaaos .auum sea sauces owuuo>< n~m.¢ as- ems nan DNA mm and mum hams new cm A .aaa ..u>a as-~¢aa "assuage Hom.m~ NNoN Hose NONN Hams sass wnmw NHMN owes a~s~ maNN w .Haa ..o>s as-~s¢~ ”susaaH omo.- o~o~ Anna oonN Nana onus nasN Hman Nuns somu o~n~ » .Haa ..o>o as-hsaa "assasao nousum oauwoem auuuaaoz seasam asuuum unmonuaom cmanuwasaa< unawam uaom noumum panda: coonusom wagon anusuuoz choc use; ummunuuoz «use “among season sews»: one .a «sane 47 Plains farmers made a meager profit and the remaining regions lost money. Losses were substantial in the Northeast and Pacific regions. Input Indices As indicated in chapter II, the input indices were compiled both with and without an adjustment for the land in government programs. The input indices including all land in farms are to be seen in Table 2. The indices with the adjustment for land in government programs appears in Table 3. Also, in Table 4 are the differences between the two indices. The greatest difference nationally was nearly 3 points in 1962. Most re- gional differences were greater with the greatest, 6.4 points, in the Delta region. The difference between the indices also exceeded 5 points in the Corn Belt, Northern Plains and Southeast regions. As discussed in chapter II, there are theoretical advantages in making the adjustment for the land in government programs. For this reason, the balance of the analysis is based on the total input index figures in Table 3. Adjusted Input Indices The adjusted index for the United States, 1910-1970, is graphed on Figure 8, page 66. The regional indices for the period 1939-1970 are graphed on Figure 7. The total use of inputs in American agriculture has remained re- markably constant on a national basis. The range from 1939 to 1970 was only 11 index points. The low point for the U.S. was 1962, the high at 1951. At 1944, the use of inputs ended an upward trend whid: began in the mid-thirties. There was a downward trend from 1952 to 1962. From 1962 to 1970 the use of inputs increased every year. Regional Trends i2_Input Use There has been somewhat more variation in input use when we look at the regions. Inputs in the Northeast changed little from 1939 to 1951, but from 1951 to 1964 there was a 28 point decrease. Since 1964 the Northeast index has changed only 2 points. 8 4 o¢.NoH so.HoH am.aoH oo.oo~ m~.mm mm.om «n.0m ~5.mm oo.mm n¢.¢m ww.¢m mo.om mm.¢¢ aa.¢m ~¢.nm Nm.am o¢.mm ma.ooH om.oo~ H¢.oo~ mm.m¢ mn.wm om.o¢ mm.m¢ oo.ma Ho.mm mm.mm mo.mm oo.mm wn.qm m¢.¢a mm.mm museum sauna: 0N.No~ ~¢.~oH o~.~o~ oo.ooH mm.wm m¢.mm mo.mm m~.mm om.nm m~.oa No.mm No.¢a m¢.~m mH.Hm qm.~m ~¢.om No.aw om.oa oo.~m qm.~a No.mm «n.5m m¢.om em.mm om.mw sa.nm Hm.mw on.~m oo.m~ om.mu m¢.wh on.nn uamaomm ma.¢od No.¢oH om.HoH oo.oo~ mm.wm m~.om m¢.om HN.mm om.¢¢ mm.mm mm.mm mm.¢¢ sa.mm o~.ma ah.mm ~5.¢m mw.mm Ho.¢m Ha.¢¢ ¢~.qa om.Ha mn.~a mm.mm NH.nm mm.mm mH.mm cm.ww mw.mm m~.ow an.mw ow.~m mo.om cumucsoz Aooa u asaav oo.moA mN.¢Oa oh.NoH oo.oo~ mm.hm q¢.mm 05.0m mm.oa oa.mm mo.om ma.om mo.na qo.oa mn.¢¢ sm.na om.oo~ mm.HoH oo.¢oH na.moH mq.moH wo.mo~ Nm.moH m~.¢oH mm.moH mo.~oH om.¢oH om.nH~ w¢.NHH wm.¢fia 00.0”” Hm.~aa om.wo~ adamam cuosusom mm.on om.~o~ #m.¢o~ oo.oo~ mn.na m~.h¢ mo.o¢ n¢.nm om.0¢ 0H.Nm ¢¢.Ha an.Na ~¢.mm mH.H¢ sw.na mo.HoH o~.NoH Nm.moH Na.on mH.moH m¢.moH mn.wod ao.oHH mm.ooH om.na mN.HoH Na.noH H¢.~o~ qo.moH mm.moH wo.oo~ mm.moH mmumum muaon h¢.noH eH.Ho~ as.ooa oo.oo~ hn.om mm.mm oH.~m ~o.om oh.mm a~.am sm.mm H~.om ~a.~m o~.mm mm.~m om.mm NH.mm om.o¢ om.~a ma.ma ae.~m mn.am mm.am H¢.~m ma.am m¢.~m -.sa om.na so.~m no.mm mo.mm mm.ma ummfl -euaom Amaumw ca name Adm movnaocwv mn.No~ Nm.ooH ~m.mm oo.oo~ mm.wm ha.wm o¢.ma «w.am mm.wm ma.hm wa.wm n~.oo~ ¢¢.mm cw.wm mw.¢oH wN.hoH o~.wo~ oo.o- HN.MHH ¢¢.MHH oo.m0a «n.0AA w¢.mo~ mw.noa mm.50a H¢.moH em.ooH oo.mo~ do.aoa sm.wm ao.mm mm.~o~ Guano nuamcu Hmuauaaofium« hm.¢oH Nn.mo~ oo.NoH oo.oo~ on.wo mm.¢a an.na Ho.~m mm.am ms.¢w oo.om 0H.Hm o~.oa «m.aw «n.0m m~.¢m m~.¢¢ m¢.nm on.nm hn.mm o~.om an.wm «0.0m w~.mw hm.oa NH.¢¢ no.~¢ on.om mm.mm 05.nm an.mw oo.~m madman noamam< anonuuoz nm.am «a.mm as.ooH oo.oo~ mw.ha NH.¢m o~.¢m m~.mm ~0.Ha eh.om ma.~m ¢O.¢m -.~¢ om.~m wn.mm ~¢.mm oo.¢m NN.¢¢ mu.¢a ~m.~m m¢.om ~¢.oa od.nm m¢.mw so.ow cm.~m om.om mo.cm mm.mw mn.mw oo.nm mm.mm uaom auou Haney mm o¢.Ho~ oo.HoH mo.HoH oo.oo~ No.wm No.5m oo.wo m¢.ha aa.mm ah.wa om.mm dw.ooH o~.ooH ~m.~o~ wo.mo~ mm.¢oH No.m0a mm.No~ an.NOa Nm.HoH mm.ooH Hm.ooH hm.n¢ ew.wa mn.HoH o~.noH Nm.ooH m~.oo~ mm.ooH ow.NoH w~.moH o~.ooH mmumum mama moouucu ~o.am chad mm.aa seas mn.aa woos oo.oo~ hem“ m~.HoH wood a~.~oa moms ms.aoa sea“ am.moa mead o~.soa «sad ma.moa Head m~.noa coma oa.wo~ mass ~m.aaa mass wa.naa “mad o~.~aa snag om.maa mama mm.a~a «was m~.m~a mmaa om.m- «was o~.-a Amos ma.m~a omma m~.-d mass as.n~a mama #m.¢- head mo.-a cams wa.oma mama ~o.~m~ sass mo.oma mama mm.oms mead oo.a~a dams o~.w~a osmfi mo.e- mmoa umww umow -nsuoz .N magma 49 N¢.NOH «N.HOH Nm.ooA co.oo~ ma.hm do.mm on.¢m mm.mm ¢~.Nm om.No oo.¢m Ho.om ow.¢m NN.m¢ No.mm n¢.~0~ mm.ao~ mm.NoH ao.mo~ OH.moH Ho.ooH mm.ooH ow.wa oo.nm wo.hm fiH.mm ¢N.HOH n~.ooH Nn.mm no.0m on.em mm.nm museum amuse: AmsmuNOum ucoacuo>ow an wouauou :zaaumuasao>: name movaaoxmv m~.~o~ hm.~oH aH.NoH oo.c¢~ «h.wm No.mm ao.mm Ho.ha Nw.om on.mm mm.¢m m¢.¢m Hm.Nm hm.om ON.Nm Ah.oa Hm.mm no.0a om.Ha ¢~.Nm om.wm mo.mm an.om Nw.mm wa.ow ~N.qm mm.mw No.aw N¢.mm mm.ms s~.wm Ho.nn H¢.¢o~ nm.m0a oo.HoH oo.oo~ nm.nm om.¢m cw.¢m w¢.m¢ ~m.~a om.Nm om.~m mn.mm om.mm «H.Nm hm.ma an.ma Nm.om mm.mm am.ma nn.nm Hm.Nm ao.N¢ Hm.mm No.mm nn.om wo.mm m~.mm s~.mw mo.nm oo.¢w mo.~w om.ow oamwomm cumucsoz nn.oo~ qw.¢ca mm.NoH oo.oo~ no.0a no.mm mo.¢a Nn.¢a No.na no.¢a mm.ma mm.h¢ Am.mm mm.¢m om.mm Hm.moH mo.moH om.moH n~.moH mn.Haa mm.ooH -.maa m¢.noH cH.moH NH.¢OH q~.moH 55.0HH n¢.oHH Nm.hfla o¢.m- om.oHH mm.~HH magnum anonusow mm.oHH No.oo~ mo.ooH oo.oo~ m¢.ma Nn.mm mu.mm mm.¢¢ mo.~m mm.mm om.ma mm.mm aw.mm nm.¢m am.mo~ oe.~o~ ma.on mH.nHa Ne.¢HH ¢¢.¢HH h¢.moH mo.qa~ mH.NHH m¢.oo~ om.~oA ¢H.moH o~.¢aa co.MaH No.0HH co.HHH mN.NHH mm.maa mmumum muaon mm.mod Nh.NoH n¢.oo~ oo.oo~ Ho.mm mo.~¢ on.~m um.¢w an.mw n~.am wN.om oo.Nm n¢.wm om.om oo.hm H¢.ooH ea.ooH hm.aoH mo.~o~ co.mod mm.nm mm.am om.om om.nm mm.ma om.nm ma.mm Hm.mm mm.em «M.M¢ ma.nm ¢~.mm UQQQ -eusom ow.~o~ no.oo~ mm.mm oo.oo~ on.sm mN.wm «w.mo m~.wa ¢H.na a¢.s@ oo.oo~ on.~o~ Hu.ooH Nm.Ho~ o~.wo~ HM.HHH an.~a~ mH.¢HH n¢.n~a 05.5HH Hw.mda «m.¢aa mm.~aa «m.HHH hm.HH~ mm.mod om.o- ma.oo~ sw.¢o~ wN.~oH q¢.moH nn.mOa cmano Aooa u heady om.n0a m¢.Hc~ wh.~oa oc.oca mm.om um.aw nw.nm mm.hm do.nw mu.mw 0H.mw em.oa om.aw mm.ow Ho.0¢ om.nm Ho.om o~.na an.nm om.¢m «w.H¢ m¢.oa mm.~m Nm.am Na.Nm sa.mm ma.w¢ mm.wm o¢.ma o¢.mm a<.mm mm.mm sausam usdmaa< anonuuoz no.mm ow.nm mw.mm oo.ooH ma.om ma.oa oc.om -.mm m~.ow om.mw mo.Na ~¢.¢m ~m.~¢ ¢N.Nm Nm.mm ¢¢.om ha.om Nu.mm mn.mm mN.¢m mw.~m ow.~¢ m¢.wm «w.ow No.mm mo.mm ah.am n¢.Hm «n.0w 0H.nw N¢.om HN.mw “Hum cuoo hw.oo~ mo.oo~ m¢.oo~ oo.oo~ no.5m mm.nm on.o¢ mu.mo oo.mm w¢.om NN.wm o¢.oo~ mo.ooH mo.HoH Na.¢oH Nu.mo~ wa.¢cH w~.¢o~ mw.moH ou.noH do.HoH m~.NoA oo.mm ma.ooH oo.mcH #m.ooH Hm.noH nm.noH ma.moH em.moH mm.¢oH hn.HoH moumum oxmA .muancH HouSuasowuw< HauoH mo mooancH No.ma ohms ma.mm acma h~.aa mead oo.oo~ head «m.ooH coma mh.ooH mcaa mm.ooa swag mm.NoH mead m~.mo~ Nooa Nh.¢c~ acme N¢.oo~ ooaa oo.moH amma o~.~HH mama «n.nHH snag m~.waa snag co.o- nmma ao.m- smma mw.¢N~ mmma «0.5Na Nmma Na.mNH Hmma oo.m- omma mm.w- mead -.o~a mama no.mNH ncma em.mNH mama on.~ma mama N~.mmH «sad Hm.HmH mead no.~ma Nsaa ma.omH Head N¢.¢NH Gama NN.~NH mama ummw you» -nusoz .m manna SO wmumum wouwcu Aooa I moaav mm. as. mm. no. mm. mm. mm. Na. No. HN. om. am. am. am. on. mm. mm. mm. mm. mm. mm. RN. RN. om. 0N. mm. 0N. mN. ouwaomm camuosoz am. I mm. I am. 0 mw.H mn.~ h~.~ mo.N mw.~ no.N ed. on. I nu. ma. No.aI HH.mI m~.mI «N.MI m~.mu mm.mI Hm. I mm.MI NN.nI RN.MI ~a.nI mN.MI Hm.mI m¢.mI em.mu o¢.nu m¢.MI nm.nI oaamfim unusuaom am.¢I Nw.nu HN.HI o ¢N. h¢.NI mN.NI N¢.HI mm. I m¢.~I om.~I Ho.mI m¢.HI NN.mu m¢.mI Nm.mI mm.mI mm.oI om.oI mN.oI No.0I Hm.cI ¢¢.ou om.mI co.mI mm.mI mw.oI mu.oI wn.eI ¢H.oI NH.0I mm.oI mmuwum wagon o¢.NI om.HI no. 0 om.H qm. as. no. ma. no. I ¢¢.aI mm.NI mo.HI oh.NI No.¢I mo.mI No.mI HH.mI mH.nI ma.mI ww.¢I mm.¢I om.¢I aw.¢u om.¢I ww.¢I om.¢I m¢.¢I mw.¢I no.¢I oa.¢I N¢.¢I ummw -susom .memuwoum .u>oo cw pawn now uaoaumsmv< «w. Nb. «0. Hd.~ mn.H Nm. mm.HI m¢.NI NN.HI o¢.NI nm.m- mo.¢u mo.¢I ma.¢u 0N.¢I o~.¢I NH.¢I NH.¢I no.¢u oo.¢I qo.¢u om.MI No.¢I hw.mI om.mI an.mI mn.mI «w.nI Guano ~¢. sm.~ NH.H o ah.a mm.¢ NN.m N~.m Hm.o Nm.¢ on.~ we. ow. om.~ mm. mm.HI Nm.HI Hm.HI Hw.HI mn.HI ¢N.HI NN.HI mn.HI c~.~u mn.HI Nw.HI ww.HI mw.HI Hm.HI oh.HI No.HI om.HI magmam Imamam< cuosuuoz um. I mm. co. ea. I Nm.HI Hm.HI om.~u om.HI w¢.HI ¢¢.HI ¢¢.HI mm.HI om.HI wM.HI mm.HI ¢¢.HI ¢¢.HI a¢.HI nm.~- on.HI ¢M.HI “Ham :uoo mm. mm.H oa.H ha.N mm.~ MN.N an. mm. ma. um. I ¢O.HI hm.~n om.HI mn.HI «M.HI ¢M.~I Hm.HI mm.HI m~.HI a~.~u mm.HI mm.HI mm.HI mm.HI o¢.HI em.HI mm.HI am.HI mmmmmw mama ma vomsmo mooavcH usacH Hmuoa a“ moocouommwn o Chad dN. $0¢H NA. wo¢d o Read mN. ooaa mm. momfi on. #cad ow. meme mo. Noma hm. Head 0N. coma OH. I ¢noH mm. I wmma on. I nmmfi no. I on¢~ wO.HI mmmH OH.HI @mma NH.HI mmmH #A.HI Nmad OH.HI Hmmd NH.HI ommfl MH.HI aéaa MH.HI m¢ma ma.dI N¢mH mH.HI ©¢¢H ®H.HI m¢ma ON.HI ¢¢mH wH.HI m¢0a wH.HI N¢¢H NA.HI ~¢ad 0H.HI o¢mH #HofiI omad ammo new» -nsuoz .s magma PERCENT OF 1967 51 FRRM INPUTS NORTHERST LRKE STRTES O 3:.“ °- W' '. ,fiw‘wu M p. O L 1 l L L J L 1 L J L L L In CORN BELT NORTHERN PLRINS O 0‘ OI. WV" w V“ O l J l l L l l L J_ J L J I In RPPRLRCHIRN SOUTHERST 0 °‘ ‘ W— " ”M‘s-0" sat- "" O l l L L J l l L l J l J I In OELTR STRTES SOUTHERN PLRINS 8-, W O L l J _l_ L i l l L J J L L In NOUNTRIN PRCIFIC O 31 wfi‘w" [port—no“ O L l J J 1 l l l l J LL 1 J In I r 1 T 1 w IOHO 1950 1980 1970 1840 1950 1980 1970 --"'"' Uns- -- REGION Figure 7 52 In the Lake States and Southeast regions input use has followed along with the U.S. average. In the Corn Belt and Mountain regions, there was an upward trend from 1939 to 1955. Input use in these regions stayed on a plateau for the next 10 years. The trend has been upward since 1965. In the Northern Plains region, there was a 15 point increase in in- puts from 1939 to 1944. From 1944 until 1962, Northern Plains inputs followed along with the national'trend. Since 1962, input use in this region has increased every year for a total of 19 points. In the Appalachian region, inputs rose 15 points from 1941 to 1951, then fell back 17 points by 1958. There has been little change in input use in this region since 1958. In the Delta and Southern Plains regions there was a general de- crease from 1939 to about 1959. Since 1959 input use in these regions has increased along with the national trend. In the Pacific region there has been a general nearly straight line increase in input use since 1939. The total increase was 25 index points. The greatest change in input use was in the Northeast region with a decline of 34 points from 1944 to 1968. The Pacific region showed the greatest increase with the low point at 1939 and the high at 1969. 1939 was also the low point in input use for the Corn Belt, Northern Plains and Mbuntain regions. The low point for the Southeast and Delta States was 1958. For the Lake States, Appalachian and Southern Plains, the low point in input use was 1962 or 1963. The Northeast region hit a low point at 1970. Peak input use came at 1969 or 1970 for the Northern Plains, South- east, Mountain and Pacific regions. In the Corn Belt the maximum inputs were used in 1967. In the remaining regions peak input use came early in the period, 1951 in the Appalachian region, 1948 in the Delta, 1944 in the Northeast and 1942 in the Lake States and Southern Plains. The range in input use was greatest in the Northeast with 34 points. The Lake States region had the least range with 13 points. Input Use 21 Major Subgroup The inputs were indexed by the major subgroups indicated by the 53 catalogue on appendix pages 104 through 106. These index numbers are tabulated in Tables 5 through 11. The farm labor input group changed the most, a 179 point decline nationally. Farm labor declined the most in the Delta region, by 338 points. The least labor was used in 1970 in all regions. The input group changing the least was farm real estate with a range nationally of only 31 points. The national peak came in 1955, the low just 7 years later in 1962. There was considerable divergence in the trends by regions. The Northeast, Lake States and Corn Belt hit peaks at 1940. The real estate input peaked in the Northern Plains and Appalachian regions in the late forties. The peak came in the early fifties in the Southeast, Delta and Southern Plains regions. In the re- maining regions, the Mbuntain and Pacific, the real estate input is still rising with the highest observation at 1970. For the remaining groups of inputs, there has been a general secu- lar increase. For the three groups: Mechanical Power and Machinery; Fertilizer and Lime; and Feed, Seed and Livestock Purchases, the in- crease began at 1939 which was the low point for all regions. For all regions except the Pacific and Northeast, the use of inputs in these three groups peaked at either 1969 or 1970. In the Pacific region, the input of fertilizer and lime peaked at 1968. For the Northeast region, power and machinery inputs peaked at 107 in 1942, feed, seed, and livestock purchases peaked at 113 in 1951, and the fertilizer and lime input peaked at 1966. For all regions except the Northeast, expenditures for taxes and interest increased generally throughout the period. In the Northeast the low point was at 1966, the high at 1945. The low point for the Del- ta region was 1948. For the remaining regions the low point for taxes and interest expenditures fell in the period 1939-1941. The high point was 1970 for all regions except the Northeast. The use of miscellaneous inputs was lowest generally in the early part of the period and highest in the late sixties. For the Delta re- gion the low was 1958, for the Southern Plains, 1946. In the remaining regions the use of miscellaneous inputs was lowest in 1943 or earlier. In the Northeast region the high point came in 1963. For the remaining 54 aa.~m wh.mm aa.um am.am wa.qm Nq.a0a Na.mm an.mm m¢.am a~.~oa aq.wm ma.mm oo.ooa oo.o0a oo.ooa oo.ooa mm.aoa a~.m0a ow.nm aN.mm on.m0a m~.aaa ao.o0a aw.aaa ac.aaa a¢.wm was . aa.aaa aa.aaa aa.aoa aa.aaa aa.mm am.oa aa.mm ma.am am.am ma.aa mm.aa am.aa aa.aa ma.aa mm.ma am.ma aa.aa aa.ma ma.aa ma.aa mm.am ma.ma am.mm aa.mm aa.am ma.am aa.mm am.mm aa.mm aa.am mm.am aa.am uwwfl -eusom noon .aoom ma.aaa aa.aoa aa.m0a oc.oca Om.am aa.mm mm.aa am.ma ma.aa aa.aa ma.aa mm.ma am.mm am.mm ma.mm ma.ma am.aa aa.aa mm.ma ma.aa aa.aa aa.aa aa.aa am.aa aa.aa om.ma aa.ma aa.ma aa.aa ma.ma aa.am am.mm asses o mooaaca m manna aa.aaa aa.aaa aa.aaa oo.OOa aa.moa aa.am aa.mm aa.mm aa.am mm.ma am.aa aa.aa am.am am.ma am.mm aa.mm aa.ma aa.aa mm.aa ma.ma aa.aa aa.aa ma.ma mm.aa am.aa mm.aa aa.aa am.aa aa.aa aa.ma mm.ma mm.am measam Iuaumma. anoaunoz am.mm mm.noa mm.am aa.aaa aa.am aa.am aa.am ma.am mm.am ma.mm aa.aa aa.am aa.ma mm.aa mm.ma ma.mm am.am am.mm am.mm aa.Om am.aa am.aa am.aa aa.aa ma.aa am.aa mm.aa aa.aa aa.aa mm.aa mm.ma aa.aa aama :uoo ma.aoa am.aoa am.aOa aa.aaa aa.mm mm.am am.mm aa.mm am.mm aa.am ma.ma mm.ma aa.aa mo.na aa.aa ma.aa aa.am am.mm ma.am ma.ma ma.aa aa.aa am.aa am.ma am.ma aa.aa ma.aa ma.aa aa.aa aa.ma ma.aa am.aa nououm oxoa am.aOa mm.ooa aa.aaa O0.00a aa.aaa aa.mm aa.mm aa.mm ma.mm aa.am aa.mm aa.am aa.mm mo.mm aa.mOa aa.moa aa.moa aa.aaa aa.maa aa.maa aa.mca ma.aoa mm.am aa.am aa.aa aa.aa am.aa aa.ma ma.am aa.am mm.ma am.aa umwfl Iauuoz omma mama aama mama aama aama aama mama mama aama oama mama aama mama aama aama aama mama mama aama oama mama aama mama aama aama aama .mama mama aama aama mama ume» S9 mm.mOa aa.aaa am.moa aa.aaa aa.mm aa.am ma.mm mm.am aa.am am.aa am.ma am.aa am.aa aa.aa aa.aa mm.aa aa.mm aa.aa am.mm ma.mm aa.am aa.am mm.am aa.am aa.am ao.am aa.am mo.mm aa.aa am.ma aa.ma ma.ma moumum wouacb mm.maa mm.aaa aa.aOa co.OOa am.mm ma.aa mm.ma am.aa ma.am aa.am mm.am mm.aa am.aa aa.ma mo.qa aa.ma aa.ma oa.aa aa.ma am.aa mm.aa oa.aa am.aa ma.aa aa.aa ma.aa am.ma mm.aa aa.ma am.aa mm.aa aa.ma oawauwm Aooa I asaav ma.aOa aa.mOa mm.aOa oo.00a aa.mm aa.am mm.am ma.mm aa.aa ma.aa aa.aa ma.aa mm.aa aa.ma ma.aa am.aa aa.aa ma.aa ma.aa ma.aa ma.mm am.mm am.am ma.am ma.am aa.mm ma.am am.mm aa.ma am.ma ma.aa aa.aa camucaoz am.m3a ca.aoa mm.aoa 00.00a aa.am am.am mm.am ma.mm ma.am mm.ma am.aa ma.aa aa.mm mm.am ma.mm am.am aa.am ma.am mm.am aa.am ao.¢m mm.am ma.am ma.am ma.mm ma.am aa.am aa.am aa.am ma.am aa.aa aa.ma acamam cuesusom ma.aoa mm.moa mm.aoa oo.OOa aa.am mm.am am.mm ma.aa aa.ma ma.aa mm.aa am.aa aa.aa am.aa ma.ma mm.aa mm.aa aa.aa aa.aa ao.mm am.mm ma.am am.am am.mm aa.ma am.aa ma.aa ma.aa am.ma aa.aa aa.ma aa.ma magnum usage ma.aoa aa.mOa am.ooa oo.OOa aa.am ao.am aa.ma mm.aa ma.ma ma.mm aa.am aa.am am.mm aaymm aa.aa am.mm ma.mm ma.am aa.mm aa.am mm.am mo.ma mm.ma ma.aa am.mm ma.ma am.aa am.ma am.ma ma.ma 0m.ma mm.ma ummm -nuaom ma.mOa ma.aoa aa.aaa 00.00a aa.mm mo.am ma.am am.am aa.mm am.mm am.am am.ma aa.ma aa.am ma.mm mm.am aa.am ao.mm mo.mm mm.am am.mm ma.mm aa.am mm.mm ma.mm aa.mm aa.am aa.om ao.aa am.aa aa.aa ma.aa Guano I ma an; mooaaca umououca acm.mmxme aa.aoa mm.aOa aa.00a 00.00a am.mm mm.am aa.mm aa.am ma.ma aa.aa aa.ma aa.aa aa.mm mm.am mm.am om.mm ma.am aa.mm mm.am aa.mm ma.mm mm.am Om.ma aa.mm am.ma aa.mm am.aa am.aa aa.aa aa.ma am.aa ma.ma aaamam anonuuoz aa.mOa aa.aoa aa.aaa aa.aaa am.mm am.am ma.am aa.mm aa.mm aa.om aa.am mm.om aa.aa am.aa mo.¢a aa.ma ma.mm ma.aa am.mm am.mm mm.mm aa.am am.Om aa.am mo.mm am.0m ma.mm aa.am ma.aa ma.aa aa.ma mm.aa uaom auoo Oa oaamfi aa.mOa aa.aaa m¢.qoa oo.o0a ma.OOa am.mm aa.mm aa.am am.mm aa.am ma.am ma.mm ma.am mm.am mm.am mm.am aa.ma aa.ma ma.ma am.aa ma.aa mm.aa ma.aa aa.aa aa.aa ,ma.¢a mm.ma ma.aa aa.mm ma.am am.mm mm.am moumum mama aa.mOa aa.aaa ma.aca aa.aaa aa.mm ma.mm ma.¢0a aa.aaa aa.aaa am.m0a aa.moa mm.¢¢a mm.mm mo.a0a am.qu aa.aaa ma.aoa am.aOa oa.m0a aa.aaa mm.c6a aa.mOa ma.mOa ma.oaa aa.aaa am.aaa mm.aaa aa.moa aa.aaa om.m0a aa.aaa mm.maa ummfl -auuoz 0mma mama mama mama aama aama aama mama mama aama oama mama aama mama aama aama aama mama mama aama oama mama aama mama aama aama aama mama mama aama Gama mama snow leanI-I 60 mm.oca am.aoa am.moa aa.aaa aa.am mm.mm mm.mm am.mm am.am mm.ma am.mm aa.mm am.am mm.ma mm.ma aa.aa aa.aa ma.aa aa.ma am.ma aa.ma ma.ma mm.ma aa.aa am.ma ma.ma ma.ma mm.aa am.ma aa.ma ma.aa am.aa moumum susaaa Aooa I usage am.am aa.mOa ma.00a aa.aaa aa.mm aa.am aa.am ma.mm aa.am mm.am am.am aa.mm ma.mm mm.am ma.mm ma.am mm.ma mm.ma am.am mm.aa aa.ma aa.ma aa.aa aa.aa ma.aa ma.aa mm.ma ma.aa am.aa ma.ma ma.ma ma.ma oauaosa am.mm amAmOa am.aOa aa.aaa aa.mm mm.mm aa.am am.am mm.am am.am ma.om am.ma aa.ma am.mm aa.mm mm.am aa.ma aa.mm mm.mm mm.am ma.ma mm.aa ma.aa oc.aa am.aa ma.ma ma.aa ma.aa am.ma aa.ma aa.aa am.ma camuasoz ma.aca ma.am mm.aaa 0o.00a am.mm aa.aaa am.am aa.am mm.am ma.am mm.ma aa.mm am.aa am.am mm.mm aa.am ma.mm am.mm ma.am mm.ma aa.ma am.mm oa.aa mm.aa ma.aa aa.aa ma.aa mm.ma ma.aa aa.ma mm.aa ma.ma mcamam anonusom am.aaa aa.aaa am.ama aa.aaa aa.mm ma.aOa aa.aaa ma.ma ma.am aa.am ma.ma am.aa mm.ma am.aa aa.am am.am ma.ma am.mm am.mm aa.am am.ma am.am ma.am am.aa aa.ma aa.ma am.aa aa.aa am.aa am.ma am.ma am.aa moumum auaon aa.am ma.mm aa.aaa aa.aaa am.am am.am ma.am am.am aa.mm mm.aa ma.aa ma.ma ma.aa am.ma aa.aa ma.ma ma.ma am.ma mm.aa aa.aa ma.ma ma.aa am.ma aa.ma aa.ma aa.ma am.ma mm.aa mm.aa ma.aa ma.aa mm.ma ummfl -ausom muammw msoocmaaoommm.wm aa.am ma.aOa aa.aaa aa.aaa am.am aa.am aa.mm mm.ma aa.mm am.am aa.mm am.mm oa.ma mm.ma mm.aa am.ma aa.aa am.ma ma.aa am.aa ma.ma ma.ma aa.aa ma.aa ma.ma mm.aa am.ma aa.ma ma.ma aa.ma aa.ma ma.ma awaao Imammm< mooaaca Ha «aama mm.maa am.moa aa.aaa aa.aaa ma.am ma.mm mm.aa am.mm am.mm mm.am am.mm ma.mm aa.mm mm.aa ma.ma ma.ma ma.aa aa.ma aa.ma aa.ma mm.ma am.ma aa.ma mm.aa am.ma mm.aa am.aa am.aa aa.am aa.aa aa.am am.aa acaaam auuauuoz am.mm aa.mm mm.maa aa.aaa ma.mm am.ma am.ma mm.mm ma.am mm.am aa.am am.mm am.ma mm.aa am.ma ma.aa mm.ma aa.ma mm.aa. am.aa am.aa ma.ma aa.ma am.oa aa.aa aa.aa mm.aa ma.aa aa.aa ma.aa aa.aa aa.aa same :uoo am.mm aa.aaa aa.aaa OOIOOa am.am aa.mm ma.mm mm.aa aa.am aa.mm aa.am aa.am mm.mm aa.ma ma.ma mm.aa am.aa ma.ma ma.aa ao.oa aa.ma ma.aa ma.aa mm.ma am.aa aa.aa ma.ma aa.ma aa.aa am.ma am.aa mm.ma moumum mesa am.am aa.aoa aa.am co.OOa am.mm aa.mm ma.moa aa.aaa aa.aaa aa.aoa mm.am aa.am mm.mm am.aa am.mm ma.aa am.am aa.am ma.mm ma.am am.mm aa.mm aa.mm ma.ma ma.mm aa.am am.aa mm.aa aa.aa mm.aa mm.aa aa.aa umwfl -auuoz ohma. mama aama mama aama aama aama mama mama aama oama mama mama mama aama aama aama mama mama aama oama mama mama mama aama aama aama mama mama aama oama mmma umow 61 regions the high point in miscellaneous input use fell at 1968 or later. Fixed and variable Inputs The inputs were also indexed by fixed and variable categories. The length of run considered was one year. That is, if the input could not be varied within a production season, then it was classified as fixed. All other inputs were considered as variable. These estimates are tabu- lated in Tables 12 and 13 and graphed in Figure 8. There has been considerable decline in fixed inputs since 1939. The high point in fixed input use was 1942. The decline was moderate, 12 points, from 1942 until 1955. From 1955 until 1962 the rate of de- cline was greater, the index dropping 25 points during this period. From 1962 until 1970, the use of fixed inputs increased by 7 points which fol- lowed the trend of the variable inputs. The regional pattern of fixed input use was similar to the national pattern for all regions except the Mauntain and Pacific. Fixed inputs were relatively constant in the Mountain region with a range of only 13 index points. In the Pacific region fixed input use was a minimum at 1939 and maximum at the end of the period. Variable Inputs Nationally there has been a general secular increase in the use of variable inputs. These inputs increased in 26 of the 31 years of the period. Of the five declining years, the greatest decline was only 1 index point. As indicated on Figure 8, the variable inputs are highly correlated with productivity. Regionally the trends tend to follow the national average except for the Northeast region. The low point for use of variable inputs was 1939 for eight of the ten regions. In the Southeast the low point came in 1941, and for the Delta States the low came at 1946. In the Northeast region, the high for use of variable inputs came at 1951. For the remaining 9 regions, the high was at 1968 or later. The range in variable inputs varied from 21 points in the Northeast to 62 points in the Northern Plains. The range for the U.S. was 42 points. 62 Index p£.Farm Output The index of farm output is constructed as follows:21 In calculating farm output, two major indexes of gross production are computed--total livestock production and total crop production. Subindexes are computed for the principal commodity groups that compose total livestock production and total crop production. The farm output, livestock production, and crop pro- duction indexes are calculated annually for each of 10 farm production regions beginning in 1939, as well as for the united States from 1910 to date. Farm output estimates are calculated at decade intervals from 1870 to 1900. These indexes are calculated by the familiar constant pricedweight method. Weighted average prices received by farmers in a given region are used as weights in constructing the indexes for the regions. The quantity-price aggregates for the 10 farm production regions are summed to obtain the quantity-price aggregates upon which the index for the united States is based. (Table 14 and Figure 9.) The reference period used for the indexes is 1967. Two weight periods are used for the regional indexes, and three weight periods for the U.S. indexes. Average 1957-59 prices received by farmers in each farm production region are used as weights for 1955 and subsequent years; average 1947-49 prices are used for 1939 to 1955. In the U.S. indexes, average 1935-39 prices are used for the years prior to 1939. Conceptually, farm output does not include the production of producer goods. These are goods produced on farms and used in further production of farm products for human use. Pro- ducer goods include such items as seed and farm-produced power of horses and mules. These products are included in the gross farm production index, which is not published but is available for research purposes. The current year's indexes of farm.output, crop production, and livestock production are based on preliminary and some- times incomplete data. Thus, the current year's indexes are subject to revision in the following year after all the basic data are available. A general revision of the series is made after each agricultural census for all years back to the pre- ceding census data. SRS calculates a preliminary index of crop production for the current year based on its monthly forecasts of crop pro- duction beginning in August of each year. The index is pub- lished in the monthly Crop Production reports. These prelimi- nary indexes for the current year are prepared for the united States only, but they are directly comparable to the historical 21From USDA Agriculture Handbook No. 365, pp. 15-17. 63 indexes for the United States built up on a regional basis. Limitations of Series The indexes of farm output and the component indexes of crop and livestock production do not adequately measure changes in quality of products over time. These indexes re- flect the changes in production caused by changes in quanti- ties of individual items included. The failure of the series to measure quality change is an inherent problem in most indexes. In theory, any crop grown for seed should not be included in farm output. Hayseed, pasture seed, and covercrop seeds are not included. Because of the lack of necessary data, no deductions are made for other types of seeds. Also, because of the lack of data, several minor products are not included in farm output or in its component series. The main item of production omitted in the farm output series is production from farm forests. This, plus other minor items omitted, probably accounts for less than 5 percent of the total farm output in recent years. Alaska and Hawaii are not included in the current series because of the lack of data through the weight period, 1957- 59. These States will be included in the farm output and component series after the next new weight period is adepted. As more than one set of price weights is used in computing the indexes, the series are spliced at 1955 for the regions and the United States. The U.S. series is also spliced at 1939. The Splicing is necessary to convert the indexes based on the various quantity-price aggregates to one final series of index numbers with 1967 as a reference base period of 100. Official reports of SRS are the chief sources of data for crop and livestock production and prices. Farm output includes crops produced during the crop year exclusive of hayseeds, pasture seeds, covercrop seeds, and hay and concentrates fed to horses and mules on farms. Farm out- put also included the ”net" production of livestock other than horses and mules, and production of livestock products. Net livestock production is gross livedweight production of live- stock on farms during the calendar year minus the constant- dollar farm value of hay and concentrates fed to livestock. Thus, the value of pasture consumed by livestock is included in net livestock production. Hatching eggs for broilers and chickens raised also are excluded in calculating net livestock production. These calcu- lations are made to avoid counting feed crops in both livestock and crop production, and to avoid counting hatching eggs as part of poultry meat production as well as egg production. 64 am.moa mm.ooa Om.OOa 00.00a aa.mm aa.am ma.am mm.am ma.am am.am mm.aoa aa.aOa mm.aOa mm.moa ma.aaa am.maa mm.0ma mm.mma am.mma mm.mma ma.ama am.mma mm.ama am.ama ma.mma am.ama mm.oma oo.ama aa.ama mm.mma aa.mma mm.mma moumum awuaca ma.mma ma.mma am.ama am.ama am.mm mm.mm am.mm mm.am aa.am mm.am mm.am am.am mm.am am.am mm.mm am.mm am.mm mm.mm mm.mm ma.mm ma.am mm.am am.mm aa.mm ma.mm aa.am ma.om aa.mm aa.mm ma.am mm.am aa.am oawaomm AooH I eoaav ma.mma ma.aOa aa.aaa am.moa aa.mOa mm.moa ma.aoa mm.aoa mm.aOa ma.mma am.ama ma.mma am.mm am.am mm.mm am.am ma.mm aa.aoa ma.am am.am ma.mma ma.am ma.am am.mm aa.mm mm.am am.mm mm.am aa.mm mm.aOa mm.am ma.mma mm.aOa aa.mm am.ama aa.aaa mm.mm aa.aaa mm.aoa ma.mm ma.mma ma.aaa mm.mOa mm.maa ma.mma mm.aOa mm.ama ma.mma ma.aoa am.mma ma.mma ma.aoa mm.ama am.maa ma.aOa mm.ama aa.aaa mm.aOa mm.mma mm.maa am.mOa mm.ama ma.aaa mm.m0a ma.mma ma.maa ma.mma am.mma ma.aaa mm.mm mm.ama mm.maa mm.mm am.ama o~.mma ma.mma ma.mma mm.aaa mm.aOa am.maa aa.aaa mm.aOa ma.aaa ma.aaa aa.00a mm.maa mm.maa ma.mm am.aaa am.aaa mm.am am.aaa am.aaa am.am mm.maa ma.aaa camuaaoz mcamam moumum cuoausom muaoa aa.moa am.ama mm.aoa am.mm am.ama mm.aOa ma.mm mm.am ma.mma ma.mma ma.mm am.mm am.ama ma.mma ma.mma ma.mma mm.am aa.mm mm.am am.am am.mm mm.mm mm.mm aa.am mm.am am.ama am.mm mm.am am.am am.ama am.mm ma.mm mm.am mm.aOa mm.am ma.mm ma.mma mm.mOa am.mm am.mm mm.aOa ma.mma mm.am mm.moa mm.maa ma.aaa aa.mm mm.mOa ma.mma mm.maa aa.mm am.ama ma.maa am.maa am.am am.mOa ma.mma ma.mma ma.mma mm.aaa aa.aaa ma.mma aa.aaa am.maa ma.maa aa.mma ma.maa am.maa am.aaa mm.aaa ma.aaa mm.maa am.maa ma.aaa aa.aaa ma.maa mm.maa mm.maa am.aaa aa.maa am.maa am.aaa mm.moa am.aaa ma.aaa aa.aaa ma.mma mm.aaa am.maa ma.aaa mm.mOa mm.maa ma.aaa mm.aaa mm.mOa mm.maa aa.aaa mm.aaa aa.aaa am.aaa am.aaa ma.aaa am.maa mm.ama ma.maa am.maa aa.ama aa.ama mm.maa ma.aaa aa.ama mm.mma ma.mma mm.maa ma.mma ma.mma ma.maa mm.maa ma.aaa ma.ama ma.aaa am.maa aa.aaa am.ama mm.maa mm.maa ma.maa mm.ama ammo amass maamam uaom Iausom Imammma cuoauuoz cuou muamca aoXah.Mm mooaaca NH «Home mm.mm ma.aOa mmma am.mm am.00a mama ma.OOa ma.0¢a mama oo.ooa am.ama mama aa.mm am.mm aama ma.mm mm.aOa aama am.mm aa.aOa aama am.mm am.ama mama mm.mm aa.aOa mama mm.ooa mm.moa aama ma.aOa mm.maa oama mm.mOa mm.maa mama am.maa am.ama mama aa.aaa am.mma mama mm.maa mm.mma aama am.mma mm.mma aama ma.ama mm.maa aama aa.ama ma.aaa mama mm.ama aa.aaa mama am.ama mm.maa aama ma.mma mm.maa oama aa.ama aa.aaa mama ma.ama mm.maa mama ma.mma mm.aaa mama ma.mma am.maa aama mm.mma mm.aaa aama am.aaa mm.mma aama am.maa mm.ama mama mm.aaa ma.ama mama am.maa aa.mma aama mm.maa ma.ama oama aa.oaa ma.ama mmma moumum ammo ume» mama Isuuoz 65 mm.moa mm.aoa aa.aOa mm.aoa am.moa am.aca ma.aoa aa.moa am.ama am.ama oo.o0a 00.00a am.mm ma.mm ma.mm mm.mm aa.mm mm.mm mm.mm am.ama ma.am ma.mm 0a.ooa ma.mm mm.mm ma.mm mm.mm am.mm aa.am ma.mm aa.mm am.am mm.mm mm.am mm.mm ma.am am.mm mm.mm am.am am.mm aa.am am.mm mm.mm mm.am ma.am aa.mm am.mm am.mm aa.mm ma.am mm.am oo.om aa.mm ao.am mm.mm ma.am aa.am aa.mm ma.am am.am mm.am aa.mm ma.mm am.am mm.mm am.mm mm.am ma.mm mm.am am.mm aa.mm ma.mm aa.mm ma.mm mm.mm mm.ma mm.am mm.mm am.mm aa.am am.am ma.ma ma.am aa.mm am.aa mm.mm ma.ma aa.aa mm.mm am.aa mm.ma mm.am aa.ma mm.aa mm.mm ma.aa moumuw aouacm uauaoom camucsoz AooH I essay am.moa ma.maa ma.aOa .mo.a0a ma.aoa aa.aaa aa.aaa am.ama am.mm mm.am aa.mm mm.am am.mm mm.am ma.mm ma.am am.am ma.am ma.mm ma.am mm.mm aa.am mm.am mm.mm aa.am ma.am mm.mm aa.am am.am aa.mm ma.mm am.mm .mmemm mm.mm am.am ma.mm ma.mm mm.mm mm.mm ma.am aa.am ma.am aa.am ma.am am.mm mm.am aa.mm ma.mm mm.mm mm.mm am.mm ma.am ma.am mm.mm mm.mm am.mm ma.mm mm.am aa.mm aa.mm mm.am am.mm am.am ma.mm whamam moumum caucuSOm muaun am.mOa aa.OOa mm.00a am.ama aa.am aa.mm ma.mm am.am ma.mm aa.am mm.mm mm.mm mm.mm mm.am ma.mm aa.mm mm.am mm.am am.am am.mm mm.aa aa.aa ma.aa am.ma mm.aa am.aa am.ma ma.aa mm.ma am.ma mm.ma mm.ma ammo -ausom mummaa oaamaum>_wm aa.aOa mm.mm ma.mm ma.mma aa.mm aa.mm ma.am am.am ma.mm aa.mm am.om mm.mm mm.am mm.mm ma.am aa.mm aa.mm mm.mm am.am ma.mm ma.mm ao.om aa.mm am.am mm.mm mm.ma mm.mm mm.aa aa.aa am.aa mm.aa ma.ma moans Imammm< mooamca ma oaama mm.mOa mm.mm mm.aOa mm.mm am.ama Om.oca ma.mma am.ama mm.mm mm.am am.mm ma.mm mm.mm mm.mm am.mm am.am am.am ma.am mm.mm mm.mm mm.am mm.mm mm.mm am.mm aa.am aa.am mm.am mm.mm ma.am am.mm mm.mm mm.am am.am am.mm mm.mm mm.ma ma.am mm.ma aa.ma ma.aa ma.aa am.ma aa.aa ma.ma aa.aa aa.ma mm.aa oo.ma mm.ma mm.aa mm.aa ma.ma aa.ma ma.aa mm.aa mm.aa am.aa ma.ma mm.ma am.aa mm.ma mm.aa ma.aa mm.ma meamam uaom muonuuoz :uoo mm.mOa ma.am mmma ma.mma am.mm mama am.ama aa.mm mama am.ama am.ama mama am.am am.ama aama aa.mm aa.aaa aama aa.am om.ooa aama mm.mm am.mOa mama ma.mm ma.mma mama am.am am.ama aama aa.mm aa.mOa oama mm.mm mm.aOa mama mm.am ma.mma mama mm.mm am.ama mama aa.mm ma.aOa aama ma.am am.ama aama mm.mm mm.aOa aama am.mm ma.mma mama mm.am ma.maa mama am.am mm.maa aama ma.mm ma.maa oama ma.mm mm.mOa mama aa.mm am.ama mama mm.ma am.mca mama am.ma ma.moa aama am.ma aa.mOa aama ma.aa ma.aOa aama mm.ma am.mm mama ma.ma mm.am mama mm.ma ma.am aama mm.ma am.mm oama mm.aa mm.mm mmma moumum ammo poor aama Iauuoz 66 2.3:. m case: 1.3:; 2:. 3:". ¢ a On,- 67 am.mOa mm.mOa am.ama 00.00a aa.am am.mm am.am am.am ma.am aa.mm am.mm am.mm ma.am ma.mm am.am aa.am mm.mm mm.mm mm.mm ma.am ma.mm aa.mm ma.am ma.ma mm.mm aa.ma mm.mm ma.ma ma.ma mm.aa am.ma ma.ma mwumum Sousa: am.mOa am.mOa aa.aaa oo.OOa am.ama ma.mma aa.aaa mm.am aa.am ma.mm aa.am mm.am aa.mm aa.mm am.mm am.mm am.am ma.mm am.am mm.mm mm.mm mm.am am.ma aa.ma mm.mm mm.aa mm.aa mm.aa am.ma am.ma ma.aa mm.ma oamaumm ma.mma mm.aOa am.mOa ma.mma ma.am am.am ma.mm mm.am mm.mm ma.am ma.mm mm.am am.mm am.mm am.mm aa.mm mm.am aa.am ma.am am.am ma.ma ma.ma mm.ma am.ma am.ma mm.ma aa.aa aa.aa ma.ma ma.ma am.ma ma.ma camucaoz mm.aaa mm.aOa aa.aaa ma.mma am.mm aa.aaa mm.mm mm.mm ma.am mm.mOa ma.mma am.am aa.am mm.am am.mm mm.mm am.am am.am ma.mm aa.am ma.am am.mm mm.am am.mm mm.mm ma.mm ma.am am.am am.mm mm.mm ma.am aa.aa mcamam cuoausom ma.aaa mm.aOa aa.maa am.ama mm.mm mm.aoa aa.aaa mm.aOa mm.om mm.mm aa.mm am.am am.mm aa.mm mm.mm mm.mm aa.mm mm.mm mm.am mm.mm aa.ma mm.mm mm.mm mm.aa aa.ma ma.aa mm.am ma.ma mm.mm mm.ma mm.ma mm.aa moumum auaoa mm.mm mm.am aa.mm oo.OOa ma.mm am.mm ma.am am.mm mm.am mm.mm mm.mm ma.mm am.mm am.ma mm.mm ma.am am.ma aa.am aa.ma am.mm ma.ma mm.ma ma.ma mm.ma am.ma am.aa aa.aa am.ma ma.ma ma.aa ma.ma mm.aa Haw” -eusom flood I Nessa mm.aOa mm.aOa am.am am.ama mm.mm mm.am mm.mOa aa.m0a mm.mm ma.mm ma.am aa.mm mm.mm ma.am am.mm am.am am.am ma.mm mm.am ma.am am.mm ma.mm am.mm mm.mm mm.am ma.am ma.am am.mm aa.mm ma.mm am.am am.am amano mm.aOa am.maa mm.aOa ma.mma mm.am mm.mm ma.mm am.mm am.mm am.mm mm.mm ma.am mm.mm mm.mm mm.ma am.aa am.ma aa.ma aa.mm mm.ma mm.ma ma.ma am.mm ma.ma aa.ma am.am ma.mm am.ma mm.am am.ma ma.aa am.mm oaamam Imaamm< cumnuuoz .uamuso aama amuoH mo mooaaaa am.mm am.mm mm.mm am.ama ma.mm am.am am.mm am.mm am.am aa.am am.am aa.am am.am am.mm ma.mm mm.mm ma.mm ma.am am.mm am.ma mm.ma mm.ma am.am mm.aa ma.ma am.ma ma.ma ma.aa aa.aa mm.ma mm.aa mm.aa aama :uoo ma.moa mm.00a mm.mOa oo.o0a mm.mm mm.am am.mm am.mm am.am am.mm ma.am mm.am am.mm aa.mm ma.mm aa.mm mm.am ma.mm am.mm ma.mm ma.am ma.mm mm.am mm.mm aa.am am.am am.am ma.am ma.am am.aa am.ma mm.ma mmuaum mesa am.mm mm.mm mm.am oo.oma mm.mm am.am aa.am ma.am ma.mm mm.mm ma.am am.am aa.mm aa.mm aa.am ma.mm am.mm mm.mm ma.mm am.mm ma.mm mm.am aa.am am.mm ma.am mm.mm mm.mm mm.mm ma.mm ma.mm am.mm mm.ma ummm -ausoz mmma mama mama mama aama aama aama mama mama aama oama mama mama mama aama aama aama mama mama aama mama mama mama mama aama aama aama mama mama aama oama mmma meow .aa manna PERCENT OF 1967 68 FRRM OUTPUT NORTHEAST 4W h‘T ,,~af~* ,1 a: J, LJ L J __J L_ CORN BELT film ,ase U? LL J. L. L, J 1 I__ N BPPRLRCHIRN Uh” !4 4"”; h MW ,1, U! .J_LL1 1_ L, 1 L N DELTA STRTES m1 l~ w In 1 4L 1 LL 1 I IL_ N HOURTRIN I”. h’ r g::e +1 a: +1 I J % 19W 1950 1980 1970 ""'""" UsSs _ REGION LRNE STRTES _I__LL LLLL J L NORTHER% I I— _L 1 IL I_ LL L SOUTHERST Loads, :§;:=:§C7~""r J ._J L _L IL I SOUTHERN PLRINS ?ve£E£yaL‘E’./:::3!LJQC._J "I L I I I L I L PRCIFIC J I l L _L T Tw IL I {Sun 1950 1930 Th? Figure 9 69 Feed, other than pasture, is calculated as a constant proportion of gross production of each kind of livestock or livestock product. The prOportion varies among regions, and is changed each time the basic price weights are changed. The feed factors are in terms of the proportion of total value (in constant dollars) per unit of livestock productimm such as per 100 pounds of beef production and per 100 eggs. Feed value excludes the nonfarm inputs such as processing and minerals that are included in the value of commercial mixed feeds. Except for broilers and turkeys, the same feed factors were used each year for each class of livestock in each region. Available data indicate that, except for broil- ers and turkeys, efficiency of feed use by livestock has shown little change. Output Trends The national trend in farm output has been upward. The low point was 1939, the high 1969. During this period output increased 45 index points or at an annual rate of 1.9 percent. The increase was fairly uniform with output increasing 22 of the 31 years. Regional movements in output have been somewhat more erratic. In the Northern Plains region output jumped 32 points from 1939 to 1942. Output in this region then changed little for the next 15 years. Begin- ning in 1958, there was another surge in Northern Plains production which carried the index up another 32 points by 1968. Poor growing conditions may have been reSponsible for the 15 year plateau in Northern Plains output. The Southern Plains region experi- enced a similar plateau from 1944 until 1957 and a similar rise of 30 points in output from 1957 until 1968. The development of hybrid sor- ghums may have been responsible for the beginning of the surge in pro- duction in these regions in the late 1950's. In the Northeast and Appalachian regions output tended to follow the national average until about 1956 when output in these regions lev- eled out. There was a somewhat similar pattern in the Lake States re- agion although the "leveling out" period did not begin until later, about 1961. Year to year variations in output was greatest in the Delta States. Here the trend followed the national average until the 18 point decline from.1955 to 1958. Delta States output has increased rapidly, 44 points, Since 1958. 70 Also somewhat erratic was output in the Southeast region although the long term trend followed along with the national average. In the remaining three regions, the Corn Belt, Mountain and Pacific, output trended along close to the national average until the late 1960's. Corn Belt production has declined since 1967. In the Mountain and Pacific regions output has pushed above the national average during the same period. Productivity Indices of productivity appear in Table 15. Regional figures for the period 1939-1970 are graphed on Figure 10. The dashed line repre- senting the national figure. The U.S. figures for the period 1910-1970 are graphed on Figure 8 (pg. 66). In an effort to average the effect of weather, a three year moving average productivity for the U.S. was plotted on Figure 18 (pg. 88). This reveals three periods of productivity advance separated by brief plateaus or declines. The first period began about 1922. The rate of increase was modest and the rise ended about 1933. Productivity then declined for three years and began another advance about 1935. Between 1935 and 1947, productivity increased rapidly for a total of 21 index points. Productivity then plateaued for another three years before be- ginning a spectacular advance about 1950. During the period 1950-1964, productivity increased 28 points. Productivity has been on another pla- teau since 1964. Regionally there has been considerable divergence in productivity trends. In the Northeast there was a general increase at a somewhat greater rate than the national average. In the Northern Plains produc- tivity shot up 28 points in the first four years of the period. There was little change in productivity in this region for the next 14 years, from 1942 until 1956. In the next two years, from 1956 to 1958, the Northern Plains index shot up another 28 points. Since 1958, the index in this region has fluctuated around the national average. The pattern in the Southern Plains region was also somewhat erratic. Here produc- tivity fluctuated around the national trend until 1957. In 1958 a rise began which carried the index up 23 points in 4 years. Since 1961 71 am.maa ma.aoa mm.ooa oo.ooa aa.mm aa.moa ma.mm am.ama mm.mm mm.mm am.am mm.mm ma.am, mm.am am.mm ma.am mm.mm am.mm mm.am mm.mm ma.mm am.mm mm.am am.mm ma.mm am.ma mm.ma am.ma ma.ma am.aa am.aa mm.ma moumum mooaom mm.aoa am.ama am.ama oo.o0a ma.mma ma.aOa aa.aaa ma.mm am.mm mm.am ma.am aa.mm mm.am ma.am mm.am ma.mm mm.mm mm.mm ma.mm aa.am mm.mm mm.mm mm.mm mm.mm am.am am.am am.am am.am am.am ma.mm mm.am ma.ma aamausm ma.aOa mm.aoa am.maa am.ama am.maa mm.mOa am.ama 00.00a am.mm am.ama ma.mma am.maa ma.am mm.aoa am.mm ma.mma mm.am mm.mOa ma.mm mm.mOa am.mm am.ama aa.mm mm.mm mm.am aa.mm am.mm mm.am aa.mm am.mm mm.mm aa.mm mm.mm mm.mm mm.am mm.am mm.mm mm.mm ma.am ma.ma ma.am am.mm am.mm mm.am mm.mm aa.mm ma.am mm.am mm.mm ma.ma ma.mm mm.aa ma.mm ma.mm mm.mm am.aa mm.am am.ma mm.ma ma.ma mm.ma mm.ma ma.ma am.ma mcamam camuczoz sumausom am.maa mm.¢0a ma.aoa am.ama am.ama ma.mma mm.mOa mm.aOa ma.mm am.am ma.mm Om.om mm.mm am.am ma.mm mm.mm aa.am mm.am am.aa aa.ma am.aa mm.aa ma.am mm.ma mm.ma mm.aa am.ma am.ma am.ma am.ma am.aa aa.ma museum muaom aooa I Roomy 3.3 3.3 3.3 8.2: 2.8 2.8 2.3 3.8 3.3 3.3 3.3 3.3 2.3 3;: ~38 3.8 2.3 3;: 3.33 3.2 2.3 8.3 3.3 2.3 8.3 3.33 3.3 2.3 3.3 3.3 2.33 3.33 ummfl -583 mua>auoaaoum amuauasoauma aouoa mo mooamaa am.mm am.ama am.ama am.maa aa.am am.maa am.ama ma.mma ma.am mm.mm mm.mm ma.mma ma.mma mm.mm mm.mOa aa.aaa mm.moa aa.aaa am.mm mm.am ma.am ma.mma mm.am mm.am ma.mm mm.mm mm.mm ma.am mm.am mm.am mm.am aa.ma ma.am am.mm am.mm am.mm mm.mm am.am mm.mm am.am mm.mm mm.am am.am aa.ma am.am mm.mm mm.mm ma.am ma.mm mm.am mm.mm am.am aa.mm am.am mm.mm mm.ma mm.am mm.am am.mm mm.ma ma.am aa.ma mm.am am.aa amaze moamam Imaumma cuonuuoz ma.am am.ama am.mm mmma am.ama mm.mm am.mm mama aa.aaa am.maa mm.am mama am.ama am.ama am.maa mama aa.mm mm.aOa ma.am aama am.ama mm.00a mm.am aama aa.mm aa.aaa ma.mm aama mm.aOa am.ama ma.mm mama am.maa ma.mma mm.mm mama aa.aaa am.maa mm.mm aama ma.mm am.am ma.am oama ma.mm am.am ma.am mama aa.mm ma.mm am.am mama mm.mm am.mm am.mm mama ma.mm mm.am am.mm aama aa.mm mm.mm aa.am aama ma.am am.am am.am aama am.am mm.mm ma.am mama aa.am am.mm am.mm mama am.am mm.mm ma.mm aama am.am mm.am mm.ma oama ma.am am.am mm.aa mama mm.am mm.am mm.aa mama mm.aa ma.mm ma.aa mama am.am mm.mm aa.aa aama am.ma ma.mm ma.aa aama mm.ma mm.aa mm.ma aama ma.mm ma.aa am.ma mama mm.am am.ma am.ma mama am.aa mm.ma ma.aa aama mm.ma mm.aa am.aa oama ma.aa mm.ma ma.aa mmma uaom mououm ammo ume» auoo aama Iauuoz .aa oaamh PERCENT OF 1967 72 FORM PRODUCT I V I TT NORTHERST LRKE STRTES 031 I J L L I l I I I I L J I N CORN BELT NORTHERN PLRINS 91 " IDJI I I I J I I I J I L I I RPPRLRCHIRN SOUTHERST W W :21» -- ATM-d, ' ‘ ’ - ' ’ gLL J LJLJ I _L J L L I I DELTR STRTES SOUTHERN PLRIN m0! p. 3L 4L1 LL I L I L I I J I I HOUNTRIN PRCIFIC . "I" V’. U“ mI L I I L I I L I I I I I L NI I I I I I I I 19u0 1950 1980 1970 1940 1950 1980 1970 ---- U.S. -——- REGION Figure IO 73 Southern Plains productivity has fluctuated around the 1961 level. The pattern in the Appalachian region is rather odd. The trend followed the U.S. until 1956. In the following year Appalachian produc- tivity dropped 9 points. After 1957, the Appalachian region again fol- lowed along with the U.S. average. In the Southeast and Delta regions productivity advance followed a similar pattern. The trend followed the U.S. until 1957. During the period 1957-1963 productivity increased more rapidly than the national average. Since 1963 productivity has been on a declining trend in these regions. The 1957-1963 advance was greatest in the Delta States, 32 points compared to 23 for the Southeast. In the Pacific region, productivity followed along with the national trend until 1954. After 1954 productivity in this region was still on an upward trend but it was moderate, the total change being only 12% points in 16 years. In the remaining regions, the Lake States, Corn Belt, and Mountain, productivity followed along closely with the national average. Output, Inputs, Productivity: A_comparison prthe three measures A better perspective of the trends of the three measures; output, inputs, and productivity, can be gained by viewing them as displayed on Figures 11 thru 15 which appear on pages following. On these figures the dashed line indicates the national average. On a national basis output has been steadily increasing, input use has increased very slightly resulting in a steady increase in produc- tivity. Looking at the regional figures, in the Northeast output lagged be- hind the nation but inputs were reduced drastically resulting in produc- tivity increasing more rapidly than the U.S. In the Appalachian region output also increased less than the national average but inputs did not decrease sufficiently and productivity increased less than the 0.8. In the Lake States outputs increased slightly less than the U.S. but inputs were essentially constant resulting in productivity following closely along with the nation. In the Corn Belt outputs increased a little faster than the U.S. but it was necessary to increase input use 74 ._ o.:m.u Ohm. 0mm. 00m. O¢m_ Ohm. 0mm. Oom— O¢m_ a 1 . a 1 R a a d a J 1 d J Ohm. 000. 00m. O¢m_ a . 3 J a L _ \ OD a gap .. %I 00. Zn: rec-Emma ZCazucI—CLLC mu. zcazo¢4¢mmc Ohm. 0mm. 00m. O¢m. Ohm. 00m. 00m. O¢m_ _ a d 4 1 a a a a H J a a I 98. com. one cam. 4 . « Jl]. a T a CD EPIC... III 2. , I , .00. hmcmzhcoz . haw-=80: DN— hmcwzhcoz >._._>_._.oaoomm Fonz. ban—p.30 . m .a IIIIIII scans. 75 2.9 89 one can. A R a 4 a a _— Y pawn 2cou Ohm. cam. Cam. Gam— a J q a a a a mmhchm wxcq >._._>.._.0:oomn_ m. o..:u.u Ohm. owm. 00m. O¢m_ - q a d a N - IE. hduo 2‘00 Ohm. 0&9 Oom— O¢m_ 'd d J J a 4 d .E awash» mega HDQZ. cam. omm. 03. ova. - 4 u a - a J- on I... . 2. oo. pawn 2:00 3. 03m... com. 08. oem. 1a a a J 4 la‘ I! on 2. oo. mm.aa...» mama mm. PDQPDO 76 n. o.=_.ru 0N2 0mm. 0mm. O¢m_ . a .1 a a 1 4 Ohm. J O@Q_ 00m. OVQ. . a J a a d mm.achm 5.4mm marsh chm—mo Ohm. 00m. 0am. Dam. - a a 1 a 4‘ Ohm. 0mm. Cam. O¢m_ a 4 a J A1 d a IIIIIImmmmmmwwu-Lmnummmmu ._.mmmzhaoa hmcuzhaom >._._>_._.0300mm Pan—Z. 05. owm. one cam. - a d a J a J on LI ..n~. oo. «mach» can»: am. 2.9 com. com. cam. 0a a» nxu_ amcuzpaoo am. .hamhno 6.: ....... Samoa 77 Ohm. 0am. 0am. a a Gaga. _¢_ ._:.o.u Ohm. 00m. 00m. O¢m_ _ . 1 a a . . Zacqm zcmzhaom Ohm. cam. 0am. Dam. a a a ‘\ ‘L azacqm zmmzhcoz >._._>.._.oaoomm 1;. ma.a-3.. 2:95.50» Ohm. 0am. a 41 a a IE azacaa zcuzhcoz k3m2_ 0am. Dam. a 03. 0mm. 03. can. d a d d 4 i d L 2.. DN 00. mzacaa zcuzpaom_ mm. 03. 89 0mm. cam. I- N a J1 '1 d1" on ON 00— mzacsa zcuzpcoz 0N. HDQHDO .m .D IIIIIII seamed 78 2.9 8.9 a. a a '1 03.9 a 3.9 n. 23:. uamaumm Okra. 0001. 00.0— 0N9 d 1 fl q H u‘ — uamaucm can. cow. N 4 1 8.9 ow9 1 a a zachzaoz >h.>_._.oaoomn_ 2.9 8m. SW. 3.9 4 a J! H -I a 1a E Zachzsoz FDQZ. cam. 8.9 8.9 3.9 L OD mh 00. Damaumm mm. Ohm. 00m. 000. O¢m. Al. 4 WT J. “V a a On . mm 00. zaChzaox 1 mm. PDn—FDO 79 more rapidly resulting in productivity following closely along with the nation. In the Southeast region outputs followed the U.S. trend. Inputs in this region also followed the national trend until about 1955 when input use slackened until the early sixties. Since that time input use in- creased resulting in a decline in productivity. In the Delta States out- put followed the national trend for the first half of the period and then increased more rapidly than the 0.8. Input use declined until about 1960 and then followed the national trend for the remainder of the period. Productivity fluctuated along with the national trend until about 1960. During the decade of the 1960's, Delta States productivity has risen above the U.S. level. In the Northern Plains region there was a large increase in outputs during the first three years of the period. This was accompanied by a moderate increase in inputs and a substantial increase in productivity. For the next 15 years there was little change in outputs, inputs or pro- ductivity in the Northern Plains region. In the Southern Plains input use decreased in the first half of the period, output increased slightly and productivity followed the national trend. In the middle of the pe- riod, from 1957 to 1970, there was a rapid increase in Southern Plains output, a 19 point increase in three years. During these three years, input use changed little resulting in a great increase in productivity. Since 1960, both output and inputs have increased at about the same rate and productivity has changed little. In the Mountain region both outputs and inputs increased at a slightly more rapid rate than the U.S. The resulting productivity index followed along close to the national trend. In the Pacific region out- puts increased a little more rapidly than the national trend but input use increased considerably and the productivity gain was less than the U.S. average. CHAPTER V ANALYSIS The root cause of productivity improvement is assumed to be techno- logical innovation. These innovations can be input reducing; output in- creasing; or innovations which increase inputs but which increase the value of outputs by more than the value of the added inputs. This chapter summarizes an effort to relate productivity to the adoption of some of the more important agricultural technologies which were adopted during this century. The Mills-Boyne Procedure The Mills-Boyne procedure, discussed at the end of chapter I, was applied to changes in outputs to estimate changes due to change in in- puts as differentiated from changes in outputs due to changes in produc- tivity. The various time periods considered are charted in Table 16. The results point up the dramatic increase in productivity from about 1935 to 1964 and the general lack of change for the other periods, es- pecially the period 1964-1969. This raises the question: What caused productivity to level out in the late 1960's? There is an ancillary question: What caused the dramatic increase in productivity from the mid-thirties to the mid-sixties? A backward look suggested considera- tion of several major technologies. ng_Four Major Technolggies There have been many technological developments which fueled the productivity increase from the mid-thirties to the mid-sixties. Of these, four seem to stand out as candidates to explain part of the in- crease in productivity: 1) The replacement of the horse by tractors enabled the farmer to pull heavier loads for longer hours. 2) Hybrid seed produced superior yields. 3) Fertilizers further enhanced yields. 80 81 Allocation pf changes 33 output using the Mills-Boyne procedure P e r c e n t C h a n g e i n O u t p u t D u e t o c h a n g e i n annual Period Tota1* Inputs Productivity average 1911-1937 21.4 12.4 900 0.35 1937-1950 36.9 6.1 30.8 2.37 1950-1964 28.6 -6.5 35.1 2.51 1964-1969 7.4 7.1 0.3 0.06 1911-1922 6.9 9.4 -2.5 -0.23 1922-1932 13.3 1.4 11.9 1.19 1932-1935 -12.4 -5.8 -6.6 -2.20 1935-1950 56.7 15.4 41.3 2.75 1950-1964 28.6 -6.5 4 35.1 2.51 1964-1969 7.4 7.1 0.3 0.06 . 1922-1935 -0.7 -4.8 4.1 0.32 1935-1964 101.5 9.6 91.9 3.17 1911-1969 129.6 28.1 101.5 1.75 * Output data are three year moving averages for this analysis. Table 16 82 4) Pesticides prevented yield decreases due to pest damages. These four major technologies were graphed against productivity for the United States and the Corn Belt region, in Figures 16 and 17. We see a correlation between the adoption of these technologies and the change in productivity. As one would expect from theory, the technologies had their greatest effect in the early and middle stages of adoption.22 By about 1964, all of the major technologies had been adopted by the bulk of farmers. Could this be the reason for the apparent leveling of productivity? To more accurately measure this correlation, a polyno- mial regression analysis was run as follows: Y I dependent variable: A three year moving average produc- tivity index. A three year average was used in an effort to average out part of the weather variation. X1 I interest on tractor inventory. X I X2 2 1 X3 I percent of corn seed which was hybrid. X I X2 4 3 X5 I percent of grain sorghum seed which was hybrid. 2 x6 -x5 X7 I index of fertilizer and lime input. 2 x8 -x7 X9 I index of pesticides input. X I x2 10 9 The graph for the U.S. (Figure 16), indicates a strong correlation and an R2 in the neighborhood of 0.7 to 0.8 was expected. The results of the regression are labeled equation 4 in Table 17. The resulting R2, 0.994 is puzzling. Averaging productivity over three years should remove part but not all of the weather variation. Considering the myriad of data involved, we would expect measurement error alone to reduce the R2 22A3 a new technology such as fertilizer approaches full eXploita- tion, returns diminish and the gain in productivity can also be expected to diminish. 23Regional data on tractor numbers were not available back to 1939. The index of interest on the tractor inventory was used as a proxy variable. 83 m. 2.3.... v.3»: 2.3.2. This...” 1:23-.. _ \\ <_ 32:5“. I \. (\ use... .22...» cm on L .3»... 2.3.2. .59... ca 7‘ not... :3 a a. Ca . ({<<, 3.2.263“. 0a o» \\.L..I.V\A\ om ax... \\ m ‘1.” Om \ .0../\AV 00. RI 0.. D K . ID 2.9 no 39 an 39 no 39 on 39 n 89 99 no.2...“ ooze: .ao.oo.oczooh Loan: .30... 0...... 84 The Four Major Technologies, Corn Belt l94O 45 I950 55 |960 65 IIO I00, Corn.__percentl hyvb’r‘ifliww ___. e/.’. 90 /. 80 Productivity o 70-- . A O / D 60 / o-,/ 50 r] 1. / Fertilizer . /, Tractor Index °‘° 40* . —- Index/ 0 n o- o’ A /o/ °\°/ // 30 /_—/ /° 0 , \/ o / o—O/ 0’0 20 / / Pesticides 0’0 n n—. ' o, /\n/ Index o—o/ D / / IO /o’° n—D'oxur“ 0’0 Bin -0-°,u—-D"° ’Dw/n O 85 by more than .006. These results also leave nothing to be explained by the minor technologies unless they are highly correlated with the in- cluded variables. In addition, the maximum of this function is 101.9. Can it be that agricultural productivity cannot rise above this level without new major technology? It was hypothesized that the minor technologies occur more or less at random and thus could be captured with a trend variable. For this reason and to avoid the maximum ceiling, the equation was rerun with a trend variable. Three of the variables which were not significant in 4, X7, and X8. equation 5 in Table 18. A graph of Y and Y as predicted by this equation the first run were dropped, X The results are labeled appears on Figure 18. The largest residual, -3.2 index points occured at 1935 reflecting the poor weather for the three years contributing to the moving average output for 1935. Similar analyses were made for three of the farm production regions, the Lake States, Corn Belt, and Northern Plains. The results of these regressions are labeled equations 6 through 8, Tables 19 through 21. Y observed and Y estimated for these farm production regions are plotted on Figures 19 through 21. The time trend carried a higher coef- ficient in these regressions, probably reflecting the later time Span involved. A priori, it was believed that poor weather conditions were reapon- sible for the flat trend in Northern Plains productivity from 1942 through 1956. The close correlation between Y and Y tends to negate this hypothesis. The correlation is so close as to lead to a suspicion that there is some feedback mechanism operating. Aaoa.a I ma monogamov mmm.a ouusauao mo nouns .aum 86 «ma.o I mm Anmm.o I a saumaaaa. haa.o coaumaouuou manages: mam.oa umoououoa essence amass Iguanasmsm oom.o- ooa.a- .oo.o ~oo.o- Na cassava xmuca usuaaasuma oka.o- Nma.o- Noo.o .oo.o- Ha cassava Sauna: sausage .masamsom ~3~.o- aam.~- .oo.c Hoo.o- o. smumaam saunas sasussa .auoo .m~.o- omm.a- Hoo.o Hoo.o- m voumsvm mucucw>ca nouowuu co amououca amo.oI mmm.oI ooo.c ooo.oI m xmsca mmaaoasmsa snm.o maq.m co“.o asm.o 3 amen“ uauaaausaa amN.o «ma.a mo~.o “ma.a 3 Sean»; samossa .msanwsom 3mm.o o~o.m mao.o ms~.o a sauna: samusma .csoo 303.0 Non.m 3¢¢.o sm~.o m mCOucm>ca acuomuu co amououaa mao.o amm.o mao.o mmo.o m .aaooo muom u mouameoo .mooo nonum .aum .aooo .mom mammauo> .a acaumdmm .moumum mouac: .xomCa mua>auoaaoum ommuo>m mmma mouau u » moumum mouac: man now ceammoumom ma oaama 87 33 o I NW... moumnmm xoaCa momaoaumom mouosmm manna: ucoouom .nasamuom mouosmn mHOuco>Ca noboouu co umououca womca moaaoauaom aama»; “coouom .uEsnwuom manna: ucoouom .cuoo muouco>cH Logomuo co umououca acouu oEaH .moumum mouacm .xoaca mua>auosaoum ommuo>m Loom oousu AaoN.H .mam.o 0mm.0I ama.0I mmm.0 mmm.0 a0a.0 amaso aam.0I aam.0 .mooo muom moumum mouac: onu uom scammoumom I mm eusasaes. I a eusmaaeav mma.mI 0aa.aI mma.m mma.m ama.a aam.ma ama.mI m0m.a moummEoo ma wanes 0ma.a mmm.0 000.0 000.0 000.0 am0.0 ma0.0 ma0.0 aa0.0 ma0.0 .asoo Noose .asm oumaaumo mo nouuo a00.0I a00.0I a00.0 mmm.0 mma.0 mam.0 maa.0I mmm.0 .amoo .wmm n w .aum coauoaouuoo mamauasz umoououca magmass> .a scaumsmm 88 ma ouawam 0m 0a 0a 0a 0m 0m 0m 00a 0aa oNa. no osa. mm one. we oqa. mm 0333 3N ONm. m. oaaa magnum amuacs .aoumsaumo a mom ao>uomao w 89 aamm.a I mm susaaasuv ama.a Isqaaumo as Nouns .esm moa.o I am Aomm.o I a Issuance. Nao.o ceasIHIssoo Iaaasaaz m0a.aaI umoououca guesses Image IIS.UNOIIa Nam.o- .3o.m- aaa.o moo.o- o. assess» amuse ama.aNuNIa 3N3.o .aa.N Hoo.o Noo.o a cassava sauna: scsussa .csoo mmm.0- 3m3.o- soo.o moo.o- m flown—.60 NAHOuCO>Ca .aOuomhu CO umwufluCH maN.OI Nmo.al a00.0 AOO.OI m xaaa. savageusum Nwa.o NNN.H as..o om..o m xmea. souNaasuoa Nam.o- mm..N- ama.a aoa.o- a cause; scaosua .csoo mNN.o mNm.o Nam.o aNN.o m muoucosca Logomuu co uoououca 0m0.0I maa.0I ama.0 aa0.0I m ecous mesa mmw.a NNN.a ao3.o Nmm.N H .msoo Iowa 3 sousaaoo .wsoo Noose .sOm ImmmmIqumm Iansaua> acamom moumum mama .aua>auosa0um ommuo>m you» wanna cOamom moumum mama oau now acamuoumom ma maamH I a .a a0aumomm 90 .mm o I Nm acumSVm momma moaaoaunom moumomm xoaoa nouaaauuom consume mauams unmouom .cuoo moumomm muouco>ea uOuomuu co umououaa xoaca momaoauuom xoaca Nonaaauuom aama»: unmouom .cuoo mLOuco>ca u0uomuu co ueououca momma mEaH .COawom uaom cuoo .mua>auoaaoum owmuo>m umom oouau onm.a .maa.o mmm.0I mmm.0 mam.aI aaa.0I mma.0I ma0.aI ma0.a maa.0I aam.m .wooo muom :Oamom uaom snoo ago you SOamwoumom 0m maamH I mm amuusaamv Nom.. I m moumsaamv mmm.0 m0m.maaI 0ma.aI a00.0 aam.a m00.0 0am.aI m00.0 ama.aI a00.0 a0a.0I mma.0 mam.mI m0m.0 mma.a 0mm.a mma.0I aaa.0 amm.aa a0m.0 u mouamaoo .mooo Nouum .aum a00.0I m00.0 aa0.0I m00.0I aa0.0I maa.0I maa.m mm0.0I mma.m .amoo .mmm u M ouQEaumo mo nouns .aum aOausaouuoo mamauasz umoououCH O I—I HNMQONWO‘ maaaaum> .m coauosmm .333.N I 33 3333=a330 300.3 33333333 30 30333 .333 91 333.0 I NM. .N33.0 I 3 333303330 333.0 33333333303 3333333: mma.maI umoououca 3333003 033333 3333330 .35033303 N3N.0- 300.3- 300.0 N00.0- N. 3333003 33003 3303333333 33N.0- 033.0- N00.0 300.0- 33 3333003 33333 3333333333 3N3.0 N0N.3 300.0 300.0 03 3333003 333333 3333330 .3300 330.3- 333.N- N00.0 300.0- 3 Umumdvm kHOucmaua uOuUm-au CO umwuquH amm.OI Boa-HI NCO-O NOO.OI w 033333 3333330 .3a033303 0N3.3 333.3 330.0 333.0 N 33333 333333333a 30N.0 N33.0 30N.0 330.0 3 33033 3333333333 333.0- 333.0- 33N.0 NoN.0- 3 333333 3033333 .3300 330.3 N30.3 N3..0 N33.0 3 mug-aama...“ uOuomuu co umOuOUCH Naa.o Nam.o mmm.0 aOa.O N 00333 3539 333.0 303.3 333.0 303.. 3 .wmoo 30m 3. 835800 .moou .30-3.3m .mum fig sang-33> .cOamom mcamam anonuuoz .aOa>auo=moum ommuo>m amok mounu n w .m acaumamm :Oamom mcamam camsupoz 353 now scammoumom am oaame 110 100 90 80 70 60 50 4O 3O 20 10 92 Y observed and Y estimated, Lake States Region 1940 1950 55 1960 65 1970 Figure 19 110 100 90 80 7O 60 50 40 30 20 10 Y observed and Y estimated, Corn Belt Region 1940 45 93 1950 55 1960 65 1970 Figure 20 94 Y observed and Y estimated, Northern Plains Region 1940 45 1950 55 _ 1960 6 1970 'U‘i 110 ._.—._- 100 90 80 70 6O 50 40 30 20 10 Figure 21 CHAPTER VI smear Productivity was defined as the ratio of total farm output to total farm inputs. The estimates are global in the sense that all productive inputs and all final products are included. Aggregation was done arith- metically using a Laspeyres weighted aggregate formula. The new series differs from the previous work of Loomis and Barton in that separate estimates were made for each of 10 farm production regions. The regional quantity-price aggregates were then summed to obtain the United States index. This procedure captures compositional difference which were missed in the earlier procedure of aggregating nationally. Loomis and Barton computed the land input by multiplying the real estate value by the mortgage rate of interest. Their method includes investment made for speculation. The new series entered the land input at the rental cost thus excluding speculative value. Another difference between the old and new series is the adjustment for land in government programs. The older series included all land in farms. The new series was computed both with and without an adjustment for land in government programs. On a national basis, the use of inputs, for the period 1910-1970, has remained remarkably constant. The regional input analyses, made for the period 1939-1970, indicate considerable diversity in input trends varying from a drastic reduction in the Northeast region to a steady in- crease in the Mbuntain and Pacific regions. The analyses of input subgroups indicate a general decline in labor input, a relatively constant real estate input and general increases in the remaining input categories. The use of variable inputs increased generally from the early 1930's to the end of the period, 1970. Over this period, fixed inputs generally declined until about 1962. Fixed inputs have increased along with variable inputs since 1962. The output indices were constructed in a manner similar to the 9S 96 input indices. These indices follow an "accrual" concept in that they endeavour to measure the quantity of output produced during the calendar year even though marketing may have taken place later. On a national basis, the trend in farm output has been a remarkably smooth increasing function since the mid-thirties. Regional trends have been similar for eight of the ten farm production regions. In the two Plains regions there were plateaus during the period 1942 through 1957. This thesis assumed that farm operations were charged for land serv- ices at the rental cost, for other equity capital at the mortgage rate of interest, and labor was charged at the hired farm wage. Under these assumptions, management return for the weight periods, 1947-49 and 1957- 59 was found to be meager. Farm owners did realize a return from appre- ciation but that was excluded from this analysis. The study indicates four general trends in the productivity of American agriculture during the period 1910-1970. There was little change in productivity during the two periods: 1910-1922 and 1964-1970. There was a slight improvement in the period 1922-1935. There was a dramatic improvement in productivity in the remaining period, 1935-1964. The regional analyses covered the period 1939-1970. During this period, the regions generally shared in the productivity increase and the recent plateau. In the Plains regions there was also an earlier plateau. In the Pacific regions the plateau was reached earlier. Only in the Northeast and Mountain regions does productivity appear to be still increasing in 1970. The regression analyses tend to indicate that four major technolo- gies have been reSponsible for most of the increase in productivity of American agriculture. These technologies were: mechanization, hybridi- zation, fertilization and pesticides. However, these results must be interpreted carefully. It is possible that these four variables are highly correlated with other factors which were also of considerable importance. For example, it is likely that mechanization is highly cor- related with economies of size. The limited evidence subsequent to 1964 indicates that if there were economies of size, then they were about ex- hausted by that time. 97 Policy Implications At the beginning of this project, the writer was of the opinion that American agriculture was curseduddia cornucopia from which flowed ever more and better technology with productivity increasing at an expo- nential rate. The analyses emphatically negate this hypothesis. It in- dicates that, during this century, there have been, so far, four major technologies which appear to be responsible for the bulk of productivity increase. The role of the minor technologies is puzzling. Common sense and theory would indicate an enhancement of productivity, however, the statistical analyses indicate only a marginal contribution. The major technologies began their spectacular contribution to pro- ductivity in the mid 1930's. As productivity increased, surpluses ap- peared, the terms of trade turned against the farming sector and a major readjustment was required in order for farmers to maintain a satisfactory level of living as compared to their urban counterpart. In the marginal agricultural regions the adjustment came by way of a reduction in inputs. In the Northeast region the adjustment was less painful than in other regions; as the real cost of commuting declined, many farms in this area came within easy access of industrial jobs. Since the late 1940's, there has been a dramatic reduction in inputs in the Northeast and productivity increased at a greater rate than the na- tional average. In the Delta region, the adjustment was somewhat more painful with the sharecroppers being forced out of agriculture, many of them moving long distances to the industrial cities of the north. However, the ad- justment benefited those remaining in agriculture and productivity in- creased at a greater rate than the national average. 0f the marginal farming regions, the Appalachian and Southeast re- gions have had the most difficulty adjusting inputs downward. Part of the difficulty in these regions has been the lack of mechanization of tobacco production. Restrictive government programs have compounded the problem by discouraging the consolidation which could facilitate mechani- zation. The adoption of the major technologies occurred at different rates in different regions. Perhaps this is the reason productivity in the 98 Pacific region leveled out about 1954 compared to 1964 for the nation. Conjecture Although it is possible that new major technologies are "just around the corner" they are not yet visible. We can only speculate as to when and if such technology will surface. In the meantime, it might be possible to increase output to some extent by terminating or revising governmental programs. Also, it will be possible to increase output by increasing inputs. This avenue is limited to stage II of the production function, however, there is a vast amount of output which would be economically feasible providing the price were high enough. If population continues to expand, eventually it will be necessary for the terms to trade to turn in the farmer's favor unless new major technology appears or unless the population expansion is muted to such an extent that minor technologies can supply the additional productivity needed. Suggestionsfgg Further Study In an expanding economy, there is a tendency for all the variables to expand together giving rise to spurious correlations. This multicol- linearity might explain the close correlations between productivity and the major technologies for the U.S., and the Qorn Belt and Lake States regions. However, in the Northern Plains region the irregular pattern of productivity change is such as to rule out multicollinearity as an explanation for the high correlation with the major technologies. A more extensive analysis of this region might reveal more about the im- pact of scale on productivity. Another hypothesis which might be tested on this region is that of feedback. Can it be that productivity in year t-l affects input use in year t? It might be possible that only certain inputs are affected, such as the four major technologies. The Northern Plains region would also be a good prospect for studying weather effects since rainfall is seldom in surplus. Similar analyses of the remaining seven farm production regions might give a clue to the reason for the high correlations between pro- ductivity and the major technologies. There was a long plateau in 99 productivity in the Southern Plains region. Was there an equally high correlation with the major technologies? In the Delta region will Y predicted follow the cyclical productivity pattern? The Northeast is another region which promises to shed further light on this correlation. Did the drastic reduction in inputs upset the correlation with the major technologies? The four major technologies do not account for changes in the effi- ciency of livestock production. Further study might reveal the role of changing feed conversion rates, artificial insemination, bulk milk han- dling, and other animal related technologies. BIBLIOGRAPHY BIBLIOGRAPHY Articles Adelman, Irma, and Griliches, Zvi. "On an Index of Quality Change," Journal of the American Statistical Association, Vol. 56, No. 295, Sept., 1961, pp. 535-548. Barton, Glen T., and Cooper, Martin R. "Relation of Agricultural Produc- tion to Inputs," The Review of Economics and Statistics, Vol. XXX, No.2, May, 1948, pp. 117-126. Barton, Glen T., and Durost, Donald D. "The New USDA Index of Inputs," Journal of Farm Economics, Vol. XLII, No. 5, December, 1960, pp. 1398-1410. Black, J. "The Technical Progress Function and the Production Function," Economica, Vol. XXIX, No. 114, May, 1962, pp. 166-170. Buttrick, John. "A Note on Professor Solow's Growth Medel," anrterly Journal of Economics, Vol LXXII, Nov. 4, Nov., 1958, pp. 633-636. Domar, Evsey D. "On Total Productivity and All That," Journal of Polit- ical Economy, Vol. LXX, Dec., 1962, pp. 597-608. "On the Measurement of Technological Change," Economic Journal, Vol. 71, No. 284, Dec., 1961, pp. 709-729. "An Index Number Tournament," Quarterly Journal of Economics, Vol. LXXXI, No. 2, May, 1967, pp. 169-188. Evans, W. Duane and Seigel, Irving H. "The Meaning of Productivity In- dices," Journal of the American Statistical Association, Vol. XXXVII, No. 217, March, 1942, pp. 103-111. Fabricant, Solomon. "Of Productivity Statistics: An Admonition," Review of Economics and Statistics, Vol. XXXI, No. 4, Nov., 1949, pp. 309-311. Farrell, M. J. "The Measurement of Productive Efficiency," Journal of the Royal Statistical Society, Vol. CXX, Part 3, 1957, pp. 253-281. Gardner, B. D. "Discussion: MEasuring Input Changes in Agriculture," Journal of Farm Economics, Vol. XLII, No. 5, Dec., 1960, pp. 1430- 1433. 100 101 Gilbert, Milton. "Quality Changes and Index Numbers," Economic Develop- ment and Cultural Changg, Vol. IX, No. 3, April, 1961, pp. 287-294. Griliches, Zvi. "Measuring Inputs in Agriculture: A Critical Survey," Journal of Farm Economics, Vol. XLII, No. 5, Dec., 1960, pp. 1411- 1427. "Notes on the Measurement of Price and Quality Changes," Studies in In- come and Wealth, Vol. 28, pp. 381-404, Princeton University Press, Princeton, 1964. "Research Expenditures, Education, and the Aggregate Agricultural Pro- duction Function," American Economic Review, Vol. 54, No. 6, Dec., 1964, pp. 961-074. "The Sources of Measured Productivity Growth, United States Dept. of Agriculture., 1940-1960," The Journal of Political Economy, Vol. LXXI, No. 4, Aug., 1963, pp. 331-346. Hall, Margaret and Winsten, Christopher. "The Ambiguous Notion of Effi- ciency," Economic Journal, Vol. 69, No. 273, Mar., 1959, pp. 71-86. Heady, Earl 0. "Output in Relation to Input for the Agricultural Indus- try," Journal of Farm Economics, Vol. XL, No. 2, May, 1958, pp.393- 405. Hogan, Warren P. "Technical Progress and Production Functions," Review of Economics and Statistics, Vol. 40, No. 4, Nov., 1958, pp. 407-411 Kendrick, John W. "National Productivity and Its Long-Term Projection," Studies in Income and Wealth, Vol. 16, pp. 67-104, Princeton Univ. Press, Princeton, 1954. "Productivity Trends: Capital and Labor," Review of Economics and Sta- tistics, Vol. XXXVIII, No. 3, Aug., 1956, pp. 248-257. Lave, Lester B. "Empirical Estimates of Technological Change in United States Agriculture, 1850-1958," Journal of Farm Economics, Vol. XLIV, No. 4, Nov., 1962, pp. 941-952. "Technological Change in U.S. Agriculture: The Aggregation Problem," Journal of Farm Economics, Vol. 46, No. 1, Feb., 1964, pp. 200-217. Loomis, Ralph. "Effect of Weight Period Selection on Measurement of Agricultural Production Inputs," _gricultural_§conomics Research, Oct., 1957, pp. 129-135. Massell, Benton F. "Capital Formation and Technological Change in United States Manufacturing," Review of Economics and Statistics, Vol. 42, No. 2, May, 1960, pp. 182-188. bhsucci, Robert H. "Discussion: Measuring Input Changes in Agriculture? Journal of Farm Economics, Vol. XLII, No. 5, Dec., 1960, pp. 1427- 1430. 102 Nevel, Robert O. "Technological Change in Agriculture," Agricultural Economics Research, Vol. 21, No. 1, Jan., 1969, pp. 13-18. Pasinetti, Luigi L. "On Concepts and Measures of Changes in Productiv- ity," Review E; Economics and Statistics, Vol. 41, No. 3, Aug., 1959, pp. 270-282. Ruttan, V. W. and Callahan, J. C. "Resource Inputs and Output Growth: Comparisons between Agriculture and Forestry," Forest Science, Vol. 8, No. 1, March, 1962, pp. 68-82. Ruttan, Vernon W. "Agricultural and Nonagricultural Growth in Output per Unit of Input," Journal g£_Farm Economics, Vol. XXXIX, No. 5, DEC. , 1957, pp. 1566-76. . "Research on the Economics of Technological Change in American Agriculture," Journal 2; Farm Economics, Vol. XLII, No. 4, Nov., 1960, pp. 735-754. . "The Contribution of Technological Progress to Farm Output: 1950-75," Review 2f Economics and Statistics, V01. XXXVIII, Feb., 1956, pp. 61- Ruttan, Vernon W. and Stout, Thomas T. "Regional Differences in Factor Shares in American Agriculture, 1925-1957," Journal gf Farm Economics, Vol. XLII, No. 1, Feb., 1960, pp. 52-68. Solow, Robert M. "A Contribution to the Theory of Economic Growth," anrterlerour. g§_Econ., Vol. 70, No. 1, Feb., 1956, pp. 65-94. . "Technical Change and the Aggregate Production Function," Review gbeconomics and Statistics, Vol. 39, Aug., 1957. PP. 312-320. Steiner, Peter O. "The Productivity Ratio: SOme Analytical Limitations on its use," Review of Economics and Statistics, Vol, 32, No. 4, Nov., 1950, pp. 321-328. Stout, Thomas T. and Ruttan, Vernon W. "Regional Patterns of Technologi- cal Change in American Agriculture," Journal 2; Farm Economics, Vol. XL, No. 2, May, 1958, pp. 196-207. Reports Fabricant, Solomon. Basic Facts on Productivity Change. Occasional Paper 63, National Bureau of Economic Research, 1959. Meiburg, Charles 0. "Nonfarm Inputs as a Source of Agricultural Productivity," Stanford University, Food Research Institute Studies, Vol. III, No. 3, Nov., 1962, pp. 217-221. 103 Mills, Frederick C. Productivity and Economic Progress, Occasional Paper No. 38, National Bureau of Economic Research, 1952. Bulletins Boyne, David H. Changes in the Real Wealth Position of Farm Operators, 1940-1960. Michigan State University Agricultural Experiment Sta- tion Technical Bulletin 294, 1964. Loomis, Ralph A. and Barton, Glen T. Productivity of Agriculture, United States, 1870-1958. 0.8. Dept. of Agriculture Technical Bulletin No. 1238, April, 1961. Loomis, Ralph A. Working in Two Worlds--Farm and Factory. Michigan State University Agricultural Experiment Station Research Report No. 32, 1965. Books Brown, Murray. On the Theory and Measurement of Technological Change. Cambridge University Press, Cambridge, 1966. Kendrick, John W. Productivity Trends in the United States. Princeton University Press, Princeton, 1961. . et 31, "Output, Input and Productivity Measurement," Studies in Income and Wealth, Vol. 25. Princeton University Press, Prince- ton, 1961. Lave, Lester B. Technological Change: Its Conception and Measurement. Prentice-Hall, Inc., Englewood Cliffs, 1966. Mudgett, Bruce D. Index Numbers. John Wiley & Sons, Inc., New York, 1951. Salter, W. E. G. Productivity and Technical Change. Cambridge Univer- sity Press, Cambridge, 1966. APPENDIX TO THES IS 104 INPUT INDEX CATALOGUE FARM LABOR Hired labor (including perquisites). Operator labor. Unpaid family labor. FARM REAL ESTATE Interest on equity in land and service buildings. Interest on real estate mortgages. Land services leased to Government on "voluntary" basis. Depreciation on service buildings and other structures. Accidental damage to service buildings and machinery. Repairs on service buildings and other structures and land improvements. Grazing fees, State forests. Grazing fees, National forests. Grazing fees, Public Domain. Grazing fees, Military lands. Grazing fees, Indian Reservations. MECHANICAL POWER AND MACHINERY Automobile depreciation, farm share. Interest on automobile inventory, farm share. Automobile repairs, parts, and tires, farm share. Automobile licenses, farm share. Automobile insurance, farm share. Tractor depreciation. Interest on tractor inventory. Tractor repairs, parts, and tires. Truck depreciation. Interest on truck inventory. Truck repairs, parts, and tires. Truck licenses. Truck insurance. Depreciation on other farm machinery. Interest on inventory of other farm machines. 105 MECHANICAL POWER.AND MACHINERY Continued: Other farm machines: Repairs, parts, and tires. Fuel and oil. Electricity, farm share. Blacksmithing and hardware. Harness and saddling. Small hand tools. Custom work. FERTILIZER.AND LIME Fertilizer mixing cost. Nitrogen fertilizer. Superphosphate fertilizer. Rock phosphate fertilizer. Potassium fertilizer. Limestone. FEED, SEED, AND LIVESTOCK PURCHASES (Non-Farm Value Added) Seed: Seed: Seed: Seed: Seed: Seed: Seed: Seed: Feed: Feed: Feed; Feed: Corn, hybrid. Corn, open pollinated. Oats, spring. Wheat, spring. Soybeans. Barley, Spring. Cottonseed. Minor grains plus all seeds other than grains. (Includes greenhouse and nursery supplies.) Grain handling by local elevators on grains fed as such. By-product processing cost. Marketing cost on by-products fed as such. Formula feed formulating and marketing cost. Hired trucking, freight and express. Milk hauling. Livestock marketing. Baby chickens purchased, broiler type. Baby chickens purchased, layer type. 106 FEED, SEED, AND LIVESTOCK PURCHASES Continued: Baby turkeys purchased, heavy breeds. Baby turkeys purchased, light breeds. TAXES AND INTEREST Taxes: Real estate. Taxes: Personal property. Interest on inventory of all cattle and calves. Interest on inventory of hogs and pigs. Interest on inventory of all sheep and lambs. Interest on inventory of all chickens. Interest on inventory of all turkeys. Interest on corn inventory. Interest on oats inventory. Interest on barley inventory. Interest on grain sorghum inventory. Interest on wheat inventory. Interest on soybean inventory. Interest on hay inventory. Interest on forage inventory. Interest on operating capital. Interest added by non-real estate debt. MISCELLANEOUS INPUTS Insurance: fire, and wind. (net). Insurance: crop-hail (net). Insurance: Federal crop (net). Containers. Binding materials. Dairy supplies. Pesticides. Irrigation operating and maintenance charges. Veterinary. Telephone (farm share). Ginning charges. Interest on horse and mule inventory. 107 IABOR INPUT Labor Quantity Of the three Federal sources of independent estimates available on the quantity of labor used in agriculture, none are entirely suited to our purpose: 8. The Bureau of Labor Statistics of the Department of Labor reports monthly the hours worked by people who work the majority of their hours in agriculture; i.e., workers are classified fish or foul, depending on where they work the most hours. Thus, many hours of off-farm work are included as agricultural labor since it is worked by part time farmers who work more hours on their farm than in off-farm employment. Also, many hours of farm labor is excluded since it is worked by people who work off the farm more hours than on the farm. The BLS man-hours series began (on a U.S. basis) in 1947. Since 1956 the series has been refined by splitting the hours into the three classes: hired, operator, and unpaid family labor. The Economic Research Service, USDA, publishes a series on man-hours used in agriculture. This series is based on an engineering study of labor requirements and is thus a man-hours used rather than a man-hours available figure. These figures are published, on a regional basis, back to 1919; however, they are not broken down by class of worker, i.e., hired, operator, and unpaid family labor. The Statistical Reporting Service, USDA, reports em- ployment of farm workers by the following categories: Farm employment represents number of family and hired labor working during the survey week. Family labor includes farm operators working on farms one hour or more plus other family members working 15 hours or more without receiving cash wages during 108 survey week. Hired workers include all persons working one hour or more for cash wages during the survey week. The SRS has also collected data for several years on the average length of workday; however, these figures cannot be accu- rately expanded to a man-hours total since we have no way of knowing how many days per week the reapondent had in mind when he reported the length of workday. Beginning in 1965, the SRS collected data on the number of hours worked per week for one week in each month. Thus, it is possible to multiply the average hours worked per week by the average number of workers on farms to ob- tain an aggregate measure of hours worked. It is possible from published data to break this figure down into regions and into hired and total family labor. The SRS has available, but does not publish, figures on the proportion of family labor which is operator labor. By spe- cial request, the SRS released confidential data, since 1965, on the number of unpaid family members working. Since these data are compiled by states, it is possible to compute a man-hours used broken down both by regions and by the three classes of labor, hired, operator and un- paid family members: 1. From the data on the number of operators working per farm and the number of unpaid family mem- bers working per farm, take the ratio of operators working to all family members working. 2. Multiply this ratio by the num- ber of all family members working 109 (from Farm Labor) to get the number of operators working. Multiply the number of operators working by the average hours which they worked (Farm Labor) to get aggregate operator hours by regions. Do the same for other family labor and hired labor. This gives aggregate hours broken down both by regions and by operator, hired and other family labor. Also, beginning in 1965, the SRS in the June Enumerative Survey, collected data di- rectly on the number of hours worked for one week in May broken down by: 1. Farm operators. 2. Hired workers. 3. Unpaid family members (other than operator). a. Working 15 or more hours during the week. b. Working less than 15 hours per week. However, these data have not been published. For hired labor, it is possible to esti- mate hours worked by dividing the wage bill by the hired farm wage. By making a seasonality adjustment, it is possible to blow up the SRS June Enumera- tive Survey data into an annual figure. Below is a comparison of hired labor hours for 1965 computed by the various methods: Hours June Enumerative Survey (seasonally adjusted) 2,175,834,000 (SRS average hours/week) (52) (SRS employment) 2,580,100,000 Wage bill -¥- composite wage rate 2,689,754,000 3,092,000,000 110 A comparison of total labor hours for 1965 follows: Hours ERS (labor used approach) 7,904,000,000 BLS (labor force approach) 10,515,000,000 SRS (employment approach) 11,319,705,600 The procedure used to compile the labor input follows: a. From the 1965 SRS data, compute regional breakdown between hired, operator, and other family members on a percentage basis. Compute percentage deviations from U.S. average. Assume the regional deviations from U.S. averages can be projected back- ward through time. b. Apply SRS regional deviations from U.S. average to BLS breakdown percentages back in time to 1956. This gives regional breakdown percentages. c. Apply these regional breakdown percentages to the ERS regional man-hours to get ERS man-hours broken down both by regions and by class of worker. The composite hired wage rate was adjusted for the value of perquisites by taking the percentage difference between F.I.B. figures on labor expenditures with and without perquisites. This is the wage used for family labor. For hired labor, it was determined that six per- cent should be added to the quantity to account for "stand-by" time. To facilitate computation, this six percent was added to the wage rather then to the quantity. REAL ESTATE INPUT Land Classification A problem arose in measuring the quantity of land in that the Cen- sus classifications have changed slightly during the period involved. There has been a general trend to carry through the sub-classifications 111 with additional refinements and with changes in the headings and group- ings of some classes. The change in land classification did not have a serious effect on the valuation procedure except, in 1939, other pasture was grouped together with all other land. An arbitrary separation was made by assuming that the acreage of all other land was constant from 1939 until 1944. The way in which the Census classifications were group- ed into value classes appears on the following pages. In order to develop the constant dollar value of the stock of land, it was necessary to develop land classifications which were consistent with the per unit values, and for the 17 Western states, also consistent with the rent/value ratios which were used to convert the stock to an annual flow of input. The only annual source of land valuations is a series by the Statis- tical Reporting Service based on a random sample of crop reporting dis- tricts. The Farm Real Estate section, FPED, ERS, has compiled a series of per acre values from these data. Their breakdown by land quality follows: I. For the humid states: 1. Plowland 2. Pasture 3. Other land II. For the mountain and Pacific regions: 1. Irrigated land 2. Non-irrigated land 3. Grazing land 4. Other land The crop report questionnaire asked for values typical for the locality as differentiated from land owned by the respondent. There is a discrep- ancy of approximately 15 percent between the values reported on the SRS survey and the Census report. The Census asked the question as follows: About how much would the land and the buildings sell for? a. Land and buildingsowned by you? b. Land and buildings rented from others? c. Land and buildings managed for others? 112 ------------.3000» 03 0030030 000 x000 303 3000 003 300003300 300 000 0003 .0 H a a H 000A 00:00 ------- ------- -------------------------.0000000 00030003 .m 0 -------------------.30000 u0080>ou083 H303 now 0300 0030 00030000 .N .00000000 000000000 00003003 0330 0030003m003030 0309 -- -------- -.30000 0008 no 000 now 0303 0000 00: 00303 000 .0300: 0030 w03000~0 03030 o\s 030m0 0030H0 0n 0H000 000 0030H0 0000 00: 00303 0w00u0< .0 O--- ........ UO---'U'---‘-UOUUUUOUUUUUOO'UUUUUOIUIOOUUImum0% “oumfi :flwaumozwg 000 00000003 00000 w03souw an 00000000 000 00um0>u00 000 0w0000< .0 000330H0 A - --------------- -.0w0uuo:3 00003 no 300300 30H 00 0300000 00030>000 000 0w0000< .n .000 33000303 .0000: 303300.30 -- --------- - ------ ----- ------- -umo 0300000 00H300 30000 00303 00 0w0000< .0 --- ------ ---------------------------300000300 000 000 00000>000 000 00030000 00000 .H r -!-- -------- ----------------------------------300000300 000 000 00030>u00 000 000H0ouo .u an--- ---------- ----.>ufiu0u0e w030000u 0uow0n 00u0uw no 00000000 .wmo 00wwo£ 00ouo .N --- ----------- -------uu---------------.00050>ou083 H000303000 0003300 0\3 30000 000 0030 000a 0>0c 00000 0000 w0330uw 00.0300 000 0000 0000 .H -u-- ---------- -----nu-unuu-u-----------30000000 00m maco 0000 000H0ouu .m - ---------------- ---------------.00300000000 000 303003000 330000003> 0000000 0000000 03 000 no 00000>000 00:0003v 30005000 .0030uw gamem .N 00033030 -- ---------------------- ---------------------------------.000 0333 000 000 0000300H .H -- ------------------- --- ------- -----------------300030>000 00030000 .0 mzm 0000um 00800 .onH¢UHOHmm¢Ao mamzmu mmma 302m 0233 113 0003 00000 ................ uuunuuuuuc uuuuuuuuuuuuuuu on--uu-u--uuuuau-nuuusnun.000300303 000 30000 .3000030 .30000 .30003 .000038000 .300303300 00000 0003 "0003 00000 33< .m ----- ......... -u---n-uu-u-uuuu--u.03000 00 30003 00 0000030 00 .0003000 .000003003 0000300 .000003 .0003330000 .00833 0000 000 0000 0000300 0 "0000300 00>00053 .3 nan ............. -uuuannunnuunnuu- ...... uuuauuuunua--uuuuau-nun-u-----uu-30000300 00000 .0 30303333333330 33333 333330300 «330303 33330 «30303030033330 333230 3333 114 3333 03000 3 000003003u0oz .8 A 000003003 r -unnuuucunuuuuu .30000 00 0000030 0003 0000 3303 30003003 .00000300 000 00030003 -uuuuuucnnuau--nu---uuuuuuuuu-auunun-uuunuuuuuu--uu-u--nuuuuuu-.0000300 00030003 --u----unnuunnnnuuuuununnuuu-n.30000 00080300083 3303 000 0300 0030 00030000 .m --u----n-unuunnuuuuuuunuuu-nuuuuuunuuu-uuuanunnnan-.303300 008803 0000330300 .N -anuuuuuunuuunuu-nu.30000000 000000000 30003003 0330 00300030333030 3309 .3000» 0008 00 000 000 0303 0000 000 00303 000 .03000 0030 00300030 03030 0\3 030w0 003030 00 03000 000 003030 0000 000 00303 0w0000< .0 uunuu-nuuuunuuauuun-u-unuuanuuuuunuuuuuuuuunuuuuu.3000» 00003 03 0303000 000 00000003 30000 0033000 00 00300000 000 00030>000 000 000000< .0 -u------uu.0w000003 00003 00 300300 303 00 0300000 000303000 000 0000000 .0 I---an---uuuannuuuuuuuu-uuuuuuuuuuu-ununun.000 .3000303 .30003m .0000000 "00 0300000 003300 30000 00303 00 0000004 .0 unuunuunu-uuuuu-u--uuuuuuunuun"00000300 000 000 000303000 000 00030000 00000 .3 au-a-------a--un---a-------u-un-un-uuuau"00000300 000 000 000303000 000 00030000 .. ..... nunuuuu.00300008 00300000 000000 003000 00 00000300 .000 00ww00 30000 .N ..... uu---unuuunuuuuuu-u-unuuunusunnunuuuuuuuuuuunuuu.00080300083 3000303000 0\3 30000 000 0030 0000 0300 03000 0000 0030000 00.0300 000 0030 0000 .3 ...... nun-unnuuaaunuuunun-nunuununuuununnuunusu-u"0000300 000 0300 0030 00030000 .......... uuuu--nu-anu-uunuunuuuuuunnnu.30300000000 000 303003000 .300000033 3003003 0000000 03 000 00 000303000 00000030 30000000 .303000 33080 .N -u--- ..... uuuuu-u---u---u----uun-uu-unuuuuuuuun-nn.000 0333 000 000 30003003 .3 a ...... annuuuunnuuuuuuununuuuunnuuuunuuunnnu uuuuu uuuuuuuuuuauu00030>000 00030000 .0 .m .< mzm¢fi zH QZ 333303 3000333 03 .20303003333300 333230 3333, 115 nu:nununuuunnuauun-uuuuuunununuuunuuuu:nunuuununuunuuuu-aan-nun .000300303 000 30000 0003 00000 .3000030 .30000 .30003 .000008003 .300303300 00000 0003 H0003 00000 330 u-- ....... u----unu-un-uu--uu- .03000 00 30003 30 0000030 00 .0003000 .000003003 0030000 .000003 .0003330003 .00833 0000 000 0000 0000300 < "0000300 00300083 .3 'IIIUIUII--I-'-|-'I.I-III'-I-'-I.--I.l.l---'I-'--'-I.IIIllaIIl-I-I---'-I--I.I "wusuwwa ”~50 .: .0 33003300033300 3303> 330330 8030333 00 .z0303003333300 303230 3330 116 d. Land and buildings rented to others? It is hypothesized that the discrepancy is due partially to: 1. The Census asked for total value of the farm in dollars, thus the respondent may have given a book valuation, rather than a present market valuation. The wording of the Census question might lead the reapondent to report a value no higher than the assessed valuation. He might believe that reporting a higher value could conceivably result in an increase in taxation. The SRS question asked for the valuation of other peoples land, thus there would be no incen- tive to under report. It is possible that the SRS report covered the more commercial- ized farms compared to the Census Report. Complete data on land classified by qualities was available only from the 5 year Cen- sus benchmark. There are annual reports which give partial data as follows: a. The Statistical Reporting Service reports the acre- age harvested of S9 cr0ps. These data are by states. b. The Natural Resource Economics Division, ERS, has data on acreage of cropland used for crops by regions. It is hypothesized that the values reported to the SRS survey did not change abruptly with changes in distribution of land between crop- land and pasture. Thus a precise annual break-down between cropland, pasture land and other land did not seem necessary. Therefore, a deci- sion was made to base the land quantity entirely on Census benchmarks. Alternatives considered for interpolating between Census years follows: 1. 2. 3. A straight line interpolation between Census years. A 6th degree polynomial equation which would give a smooth curve passing through each Census point. A polynomial curve of less than 6 degrees. Although the curvilinear regression has certain theoretical advan- tages, it suffers in that subsequent up-dating of the series might change the regression line. For this reason a decision was made to use a straight line interpolation. 117 T_h_e_ Egg 9_f_ Buildings In order to build up total land values by classes, it was necessary to use weight period per acre values which excluded buildings. The op- erator's dwelling was not considered to be a production item, thus it was desired that it be excluded. However, the remainder of the farm buildings are production items and should be included as an input. Serv- ice building values were available which could be deflated to obtain a constant dollar value. However, annual regional data were also available on the value of buildings as a proportion of total real estate value, also the value of dwellings as a proportion of total building value. These ratios permitted the calculation of a constant dollar value for land plus service buildings. An illustration of the equation used for this calculation appears on page 118. Data Sources for rent/value ratios 1. The Census reports the ratios of cash rent/value. These figures were rejected for this study since in the 17 Western States public grazing land is included; the rental rate in many cases is less than the market rate on this land. 2. The Crop Reporting Service has, for several years, collected data on cash rents, together with estimated values for the farms so rented. These figures were considered to be superior for this study even though there were a limited number of observations in some states 0 In order to estimate net rents, it was necessary to estimate Land- lord expenses. Data for these costs came from: 1. The 1955 Survey of Farmer's Expenditures, compiled jointly by the Census Bureau and the USDA, published as USDA, ANS-354 December, 1959. 2. The 1956 Survey of Landlord's Expenses, an unpublished small sample survey of expenditures by landlords on tenant-operated farms. Building maintenance was estimated from both sources and averaged. 118 Payment for insurance premiums was estimated from the 1956 survey, and insurance claims collected from the 1955 survey. Accidental damage and maintenance of land and water improvements was estimated from the 1955 survey. Building depreciation and obsolenscence was assumed to be 1% percent of building value, and management was assumed at 2 percent of the gross rent. Building depreciation and management costs were assumed at the same rates for the 48 states. Other expenses were estimated separately for each of the ten farm production regions. The net-rent-value ratios calculated are tabulated on page 119 along with a 1959 comparison to the mortgage rate of interest. Procedure Used for Land Compilations The weight period per acre values, from the farm real estate section, are for bare land. These values were applied to the interpolated census quantity figures to obtain the PQ for the stock of bare land. Annual figures on the value of dwellings as a proportion of buildings and build- ings as a proportion of land plus buildings were obtained from the Farm Real Estate section. The following equation was derived to obtain the value of land plus service buildings: Procedure for convertingiland value to land + service building value (Using 1939 Northeast region figures as an example) (in $1,000) Land as a proportion of land + buildings - l - buildings as a pr0portion of land + buildings - l - .555 = .445 Then land - value of land + buildings - 2,745,463 - 6,169,580 .445 .445 and (.555) (6,169,580) I value of buildings - 3,424,117 (.531) (3,424,117) = value of dwellings - 1,818,206 subtract to get value of service buildings 1,605,911 value of land + service buildings - 4,351,374 (2,745,463) (.555) (.531) - value of residence (l - .555) (2,745,463) - value of land + buildings (1 - .555) x L r A, fi 2 745 463 2,745,463 (.555) (.531) - value of land + service 1 - .555 - .555 buildings 4,351,374 d d e 119 Letting: L - Value of bare land, d - buildings as a proportion of land plus buildings, e 8 dwellings as a proportion of buildings, x - value of land plus service buildings, (fir-Ms) de M H Comparison: Net rent/value gs, mortgage rate g£_interest (cash rented farms) Ratio: Net rent/value Mertgage Interest 24 Region 1947-49 1957 1958 1959 Rate, 1959 Percent Percent Percent Percent Percent Northeast 3.2 3.5 2.7 3.1 5.40 Lake States 3.1 3.9 3.8 3.9 5.10 Corn Belt 2.9 4.0 3.8 3.7 5.17 Northern Plains 3.9 4.0 3.9 3.7 5.03 Appalachian 6.8 7.0 6.7 6.0 5.59 Southeast 7.9 6.8 6.6 7.0 5.96 Delta States 9.7 9.0 8.6 7.8 5.63 Southern Plains 5.5 4.9 4.6 4.5 5.38 MOuntain 5.8 5.2 5.6 4.7 5.36 Irrigated 8.6 7.6 6.8 5.9 - Nonirrigated 12.9 6.8 6.8 6.0 - Grazing 3.1 3.3 4.7 3.7 - Pacific 7.3 4.9 5.1 4.7 5 65 Irrigated 8.2 5.2 5.2 4.8 - Nonirrigated 7.4 4.6 6.0 5.7 - Grazing 5.7 4.5 4.7 4.2 - 0.8. Average 5.3 5.0 5.0 4.7 5.35 24Eitel, Van E., "Farm Mortgages Recorded in 1959," USDA, ERS-61, April 1962. 120 DEPRECIATION Current dollar estimates of depreciation were obtained from the Farm Income Branch. These estimates are compiled as follows:25 "Depreciation.--Depreciation covers. . . service buildings and other farm structures, motor vehicles, and farm machinery and equipment, plus accidental damage to farm buildings. It is the estimated outlay, in current prices, which would be required if farmers were to replace the plant and equipment used up during the year. The estimates are based on a "declining balance method" in which a constant percentage represent- ing the annual rate of depreciation of each type of capital is applied to its estimated value at the beginning of each year. Depreciation rates are applied separately to the various forms of capital equipment on farms, but only 40 percent of automobile depreciation and 78 percent of farm truck depreciation is charged as a production expense. The as- sumed rates of depreciation result in a total chargeoff of approximately 95 percent of total value over the period of years estimated as the use- ful life of the capital item, with the remaining 5 percent representing scrap value. For each of the capital items, the appropriate rate is applied to the sum of its total value in constant dollars at the beginning of the year and half the value in constant dollars of purchases during the current year, because purchases are assumed to be distributed evenly over the year. Constant dollar values are employed for these calcula- tions to measure the real consumption of capital. An index of prices paid by farmers is used to convert from a current to a constant dollar basis or from a constant to a current dollar basis. Estimates of accidental damage to farm buildings from fire, wind, and hail are based chiefly on loss reports to insurance companies. State estimates of depreciation for each category are derived by allocating the 0.8. totals among the States on the basis of the best in- formation available. For example, depreciation on buildings is distri- buted in proportion to the estimated value of buildings on farms in each State. Depreciation of motor vehicles and other farm machinery is 25FromAgriculture Handbook No. 365, Vol. 3, "Gross and Net Farm Income," pg. 10f, U.S. Department of Agriculture, 1969. 121 allocated in proportion to numbers of each type of vehicle on farms, or other relevant information such as State totals of the cost of repairs to farm machinery." MAINTENANCE OF SERVICE BUILDINGS AND MACHINERY Current dollar estimates were obtained from the Farm Income Branch, ESAD. These estimates were compiled as follows:26 "Repairs and operation.--This covers repairs and maintenance of farm buildings, repairs and other operating expenses for motor vehicles and farm machinery, and expenses for petroleum, fuel, and oil used in the farm business (net of refunds or tax credits). EXpenditures for building repairs and maintenance are estimated sep- arately for dwellings and other farm buildings. In general, U.S. totals are derived by applying to total expenditures for construction, estimated ratios of repairs to total construction outlays. State totals are de- rived by allocating the U.S. totals among the States in proportion to the value of buildings on farms. U.S. expenditures for petroleum, fuel and oil used in the farm busi- ness are from the quinquennial Census of Agriculture. Estimates for other years are projected from the census year on the basis of changes in prices of petroleum products, numbers of motor vehicles on farms, and average consumption of fuel and oil for each type of motor vehicle. Repairs and other operating costs of motor vehicles and machinery are estimated separately for farm autos, trucks, tractors and other farm machinery. They include such items as the cost of repairs, tires and tubes, registration fees, and insurance. Only 40 percent of the total cost of auto repairs and operations is charged to farm production and 78 percent for trucks. State totals of repair and other operating costs of motor vehicles and farm machinery and equipment are derived in general by distributing the U.S. total expenses for each type of vehicle among the States in proportion to the number of vehicles on farms, the number of tractors on farms, and other measures such as the estimated retail value of farm machinery parts shipped and an index representing changes in labor costs. Changes in prices for such items as tires, and services 26From Agriculture Handbook No. 365, Vol. 3, op. cit., pg. A-l9. 122 such as insurance premiums, are utilized in constructing this account." INTEREST ADDED BY NON-REAL ESTATE DEBT Data were obtained from the Farm Income Branch for current dollar expenditures for interest on chattel mortgages. Their accounts are made up of the following components: a) Interest paid to all commercial banks, obtained from Carson Evans, Agricultural Finance Branch, adjusted for interest paid on non-farm credit. b) Interest paid to all federal agencies except the CCC, also obtained from the Agricultural Finance Branch. c) Interest paid to the CCC, obtained from the A808. d) Interest paid to dealers and others. Since most short term debt is incurred to finance inventory, and since we are charging interest on inventory at the mortgage rate, it would be double counting to include all of the interest paid on short term debt. Thus, the weight period difference between the short term rate and the mortgage interest rate was calculated reasoning that this is the added cost to the farmer of borrowing short term credit rather than using his own capital. Thus, the short term principal was multiplied by the weight period difference between the short-term and mortgage rates and the resulting interest added figure was deflated with the SRS index of Production Costs. Intiutively this may seem like erroneous double deflation but consider the example: Assume a farmer borrows the money to buy a grain drill in the weight period. The drill cost $400. The interest rate was 6%. Assume the loan was for one year. The loan cost was $24. Suppose in 1970 the farmer buys an identical grain drill but the price is now $600 and the interest rate is 8%. The loan cost is $48. We wish to deflate in such a manner that the two loans have equal constant dollar cost. To do this we have to both de- flate the price of the machine and use a weight period inter- est rate. 123 This input was entered in the computer program as follows: Let t - the current year t-l . the current year minus one, in other words, last year P - amount of short term money borrowed S - current short term interest rate So S (t-l) - interest rate on short term credit outstanding, assuming the current rate lagged one year approximates the rate on loans outstanding Sw = the weight period interest rate on short term credit out- standing L - the weight period long term interest rate (use the mort- gage rate of interest as a proxy) I 8 the index of production costs Want P/I(Sw-L) knowns: SOP, Sc, Sw, L, I. Sw SOP sop-g- - SwP and L — = LP 0 0 SW sop sop-g; - L 3: - PSw - PL = P(Sw-L) divide by I to deflate: (1) S P sw L (so?) (1) 3,? L) sop (3,, - L) I 0 so sO I 50 SW ISO 8 p s P 8 ISO 2 I sw -L 8°(sw -L PETROLEUM, FUEL AND OIL It was necessary to construct price indices for this input since none existed on a regional basis. The indices constructed were based on the tank truck price of regular gasoline. These price figures were available by regions since 1959. The Agricultural Estimates Division supplied confidential data for the years 1957 and 1958, although these figures were subsequently published in the December 30, 1966 Agricultural Prices. 124 For the years prior to 1957, regional data were not available for tank truck gasoline. Data for these years were manufactured by the fol- lowing procedures: A. For the years 1948 through 1956, it was assumed that the national percentage variation between tank truck prices and filling station prices could be applied to each re- gion. Thus, filling station prices were computed by re- gions and decreased by this percentage difference to derive tank truck prices. No expenditure measurements were made between the years 1939 and 1947. Thus the F.I.B. expenditure figures are a straight line interpolation for this period. To de- rive a tank truck price for this period, the regional tank truck price was computed by procedure A for the years 1939 and 1947. Then the regional variation from the United States average price was computed for these two years and this variation was interpolated for the other years in the interval. These interpolated vari- ations were then applied to the annual United States tank truck prices to obtain regional tank truck prices for the period. CUSTOM WORK The Farm Income Branch keeps accounts on cotton ginning but for no other custom work. A decision was made to handle ginning separately and assume that it is all done by non-farmers. The 1955 survey of farmers expenditures has entries for both custom work expenditures and income from custom work. This would supposedly enable a calculation of value added by non-farmers which came out: Proportion of Custom Work Done by Non-farmers EEEESEE. North 70.8 South 71.9 West 61.0 125 Since these figures were obviously too high, Paul Strickler, Produc- tion Resources Branch, FPED, was asked for an educated estimate which was 35 percent for the Pacific Region and the Plains States and 20 per- cent for the remaining regions. A subsequent conference with Grady Crowe, Production Adjustments Branch, FPED, indicated that this estimate was low for cotton regions since aircraft are customarily used for insecticide application. Thus a decision was made to raise the Delta and Southeast regions to 35 percent. The first census measure on custom work was in 1949. The wording on the 1949 and subsequent census questionnaires appears on a following page. At first it was believed that ginning could be subtracted from the census figure to obtain a figure for custom work other than ginning. The results indicate that if the Farm Income Branch's figure on ginning is subtracted from the census expenditure for custom work the rate per acre comes out negative for the Delta states in 1949 and only 15 cents in 1954. Conferences with Rudie Slaughter and Herbert Brown, Production Ad- justments Branch, FPED, indicated that most farmers consider ginning to be a processing cost rather than custom work. The changes in the word- ing of the census question and comparison with the 1955 figure from the survey of farmers expenditures indicates that the 1959 census figure probably included most of the ginning with custom work. In order to estimate a 1954 figure comparable with the non-cotton regions, the percentage change from 1954 to 1955 for the non-cotton re- gions was projected from 1955 back to 1954 for the cotton regions. (Much of this change was likely sample bias rather than a change in per acre rates.) For the cotton regions; the Delta states, the Southeast, Southern Plains, Appalachian, and Mountain, these 1954 rates were used for all years from 1939 to 1954. For the non-cotton regions, the 1949 rates per acre were used for all years from 1939 to 1949. The 1964 census excluded ginning but included contract work. Since much of the Bracero labor was contracted, an effort was made to separate this out of the 1964 census figure by: a. Calculating the change is custom work expenditures from 126 1959 to 1964 for the United States, excluding the states of: California, Texas, Florida, and Michigan27 since these states use most of the Bracero labor (ERS Ag. Econ. Report #77). b. Applying the rate of change calculated in (a) for these four states to the 1959 census expenditure. This assumes that the rate of change 1959-64 for the remainder of the nation can be applied to these four states. To move between census benchmarks, the custom rate per acre of cropland harvested was interpolated. This figure was applied to the an- nual cropland harvested acreages to obtain the P0 in current dollars. The 1964 custom rate/acre was used for subsequent years. The deflater used was an average of the indices of prices paid for: a. Motor supplies. b. Farm machinery. c. Wage rates. Wording of census questions regarding expenditures for machine hire and custom work: The question was first asked in 1949: How much was spent last year for machine gigs? (Include tractor hire, threshing, combining, silo filling, baling, plowing, and spraying.) 1954: How much was or will be spent this year for machine gigs} (Include custom work such as tractor hire, threshing, combining, silo filling, baling, ginning, plowing, and spraying.) 1959: (Question 276) How much was or will be spent this year for machine higgfl (Include custom work such as tractor hire, threshing, combining, cotton pick- ing, cotton ginning, silo filling, corn picking, baling, plowing, fruit picking, spraying and dust- ing.) (Question 277) How much was or will be spent this year for hired labor? (Do not include housework, 27Since this procedure increased rather than decreased the 1964 Michigan Expenditures for custom work, a decision was made to use the 1964 census figure for Michigan without adjustment for contract labor. 127 custom work, or contract construction work. In- clude cash payments only.) 1964: (Question 345) How much was or will be spent for machine hire, custom work, and contract work in 1964? (Include expenses for the hire of farm machinery and equipment; and custom work such as grinding and mixing feed, plowing, combining, corn picking, silo filling, sprayifig, dusting, and contract work such as fruit picking, berry picking, fruit harvesting, etc. performed by a contractor, crew leader, a cooperative, etc. Do not include expenses for cotton ginning.) (Question 346) How much was or will be paid in cash for hired labor in 1964? (Include payments to members of the family and payments made or to be made for Social Security taxes. Do not in- clude payments for housework, custom work, or contract work.) VEHICLE LICENSES The cost of licensing farm vehicles can be considered a farm input in so far as the expenditure represents a payment for road construction and maintenance. Data on the current dollar expenditures for the pro- duction share of this item were obtained from the Farm Income Branch, ESAD. The SRS index of prices paid by farmers, including interest, taxes, and wage rates was used as a deflator. VEHICLE INSURANCE The production share of the cost of insuring motor vehicles was de- rived similar to vehicle licenses. Data on current dollar expenditures, net of loss adjustments, were obtained from the Farm Income Branch. The index of motor vehicle prices was used as a deflator. 128 BLACKSMITHING, HARDWARE, AND SMALL HAND TOOLS Current dollar expenditures data for these items were obtained from the Farm Income Branch, and deflated with the SRS index of prices paid for farm supplies. ELECTRICITY Regional data for expenditures on electricity were not available prior to 1949. An assumption was made that there was no change in the regional distribution for the period 1939-1948. Data for this period and regional data for subsequent years were obtained from the Farm In- come Branch. These current dollar expenditures were deflated with an index based on the price per KWH paid by farmers for electricity. HARNESS AND SADDLING Current dollar expenditure data for these items were obtained from the Farm Income Branch. These figures were deflated with a price index constructed from the price of horse collars reported in Agricultural prices. FERTILIZER The fertilizer input was compiled as follows: The composite fertilizer grade was calculated for each region, then a grade as nearly the same as possible was found for which a price was available. Mixing cost was then assumed to be the difference between what farmers paid for this grade and what they would have paid if they had purchased the equivalent quantity of nutrients at straight material prices. Quantities and prices for the early period (1939-54) were based on oxides. That is, P205 for phosphorus and K20 for potash. For the late period (1955 to 1970) prices and quantities were on an elemental basis. For the late period, anhydrous ammonia was priced separately from other forms of nitrogen. For both periods rock phosphate was priced separately 129 from other forms of phosphate. For everything except rock phosphate, prices were derived on a nutrient basis. Rock phosphate was priced as a straight material. Non-farm use was then estimated and subtracted from the total. The procedure used to estimate non-farm use follows: Procedure for determining non-farm use 9: fertilizer There are three sources of data on non-farm use of fertilizer: 1. 3. Article by Scholl and Wallace in V01. 5, #2 of Agricul- tural Chemicals, NAL #381 Ag8, June 1947-June 1948 Data. USDA Bulletin 216, Bottom of Table 2, 1947, 1950, 1224, data. USDA Bulletin 348, 1252, data. Only the sources underlined were used because: 3. There was no significant difference between 1 and 2 for 1947 and source 2 lacked data for the Lake States and the Corn Belt regions. The 1950 data from source 2 was judged to be very rough estimates by Don Ibach, FPED, ERS, and did a very poor job of fitting the trend. The three sources underlined were used as follows: 1. For the years prior to 1948 the 1948 ratio of non-farm/ total was applied to the total figures. 2. For the years 1948-54 and 1954-59 a straight line inter- polation was used. 3. For the years subsequent to 1959, the rate of increase from 1954 to 1959 was cut in half and continued. In regions where non-farm use decreased from 1954-59, the 1959 rate was continued. Limestone I. Quantity: The 1954 and 1959 census tonnage figures were adjusted for under- enumeration using the adjustment figures for acres of cropland II. 130 harvested.2 Then: A. For the Mountain Region, census data were interpolated between census years from 1939 to 1959. Lime Institute data were not used since there were no reports for several mountain states during this period. For the Mountain Region, the census did not report lime tonnages in 1964. Thus for 1964 and subsequent years, for the Mountain Region, the Lime Institue figures were used. For the period 1959 to 1964, an interpolation was made between the 1959 census tonnage and the 1964 Lime Institue tonnage. B. For the other regions: 1. For the period 1939 to 1954: a. For the Pacific Region a straight line inter- polation was made between census benchmarks. b. For the humid states, the lime association and/ or institute figures were wedged to fit the cen- sus data. 2. For the period 1954 to 1964, the Lime Institute figures were wedged to fit census data for all re- gions except the Mountain Region. 3. For the period 1964 to date, the 1964 wedge factors were applied to the Lime Institute figures for all regions except the Mountain Region. Price: A. For the 1957-59 weight period, SRS spring and fall prices were averaged. B. For 1947 and 1948, spring SRS prices were used. For 1949, spring and fall SRS prices were averaged. THE SEED INPUT Two basic approaches are available to estimate constant dollar 28Underenumeration figures were not available by states from the Gen- sus bureau as of 1 December 1968. The figures on page 59 of ERS Agricul- tural Economics Report #149 imply a correction of only 1% percent. Thus a decision was made to use the 1964 census data without an adjustment for underenumeration. 131 expenditures for seed. 1. The quantity planted for some of the major grains is published in Agricultural Statistics. For some of the remaining grains, the quantity planted can be estimated by multiplying acres planted by the seed- ing rates from Agricultural Statistics. These quan- tities have been computed, for most seeds for which prices are available, by the Farm Income Branch. These quantities can then be multiplied by the pro- portion purchased29 and the weight period value added to obtain the value added in constant dollars. 2. Expenditures in current dollars, available from the Farm Income Branch, can be deflated with the index of seed prices. This gives expenditures in constant dol- lars which can then be multiplied by the weight period marketing margin to obtain value added in constant dollars. The first approach is more accurate for those crops for which price data are available, however, prices are not available for many of the minor crops. Thus, the first approach was used for the major grains and the second approach for the remainder of the seed input. The grains treated as major grains were: Corn hybrid Soybeans Corn, open pollinated Barley, spring Oats, spring Cottonseed Wheat, spring This is the way F.I.B. classifies the grains (plus a misc. class). It was necessary to use their classification system since prices were not available for all of the minor grains, thus prohibiting them from being 29An unsuccessful attempt was made to run down the source of F.I.B.'s figures on the fraction of seed purchased. Neither George Tucker nor Jim Kendall, SRS, could recall making these estimates and had no knowl- edge of published figures. An estimate could be derived from the series "Field and Seed Crops" (Statistical Bulletin 386, and its predecessors, NAL library number 1 Ag 84 St) by subtracting the seed used on farms where grown from the seed planted. The difference represents the quan- tity purchased, however a portion of this comes from neighboring farms and does not move through commercial channels. 132 treated physically. The second approach, i.e., deflating current expenditures, was used for all seeds except the major grains. Since F.I.B. expenditures are adjusted to the Census after being totaled, the data had to be manipulate! as outlined in Procedure a (which follows below) in order to remove the major grains from expenditures. Considerable difficulty was experienced in locating data for estimating the weight period marketing margin to apply to the deflated expenditures. To estimate this marketing margin the following state or regional data are needed: 1. The price paid by farmers. 2. The price received by farmers. 3. The quantities planted by states if both 1 and 2 are not quoted by Farm Production Regions. This is to enable the state prices to be weighted into regional averages. After removing the major grains from expenditures, the more important seeds remaining are: Alfalfa, red clover, and potatoes in the Northern regions; winter wheat and alfalfa in the Plains states (plus grain sor- ghums in the late period); and lespedeza in the South. It was not pos- sible to compute the weight period marketing margin for alfalfa seed since the prices paid are quoted by common and certified, and data on the proportion certified were not available. (On a quantity planted basis, i.e., we know where the certified seed was produced but not where it was "consumed" for seed.) For potatoes, the price received by farm- ers for potato seed was not available. Thus, a decision was made to base the marketing margin (for the de- flated portion of the seed input) on the three seeds: Winter wheat, red clover, and lespedeza. However, even these data were not "clean". The price paid for lespedeza seed is quoted only by four varieties. The quantities produced of each variety were used to weight the prices into an average for all lespedeza seed. There were no data available on the quantity of red clover seed planted. The state prices paid were weighted to regions with the acres harvested of clover and timothy hay. Of the wheat seed purchased, a large portion is bought from neighbors but data is very skimpy on this. Except for the grain sorghum regions, fragmen- tary data was used as a basis to estimate 55 percent purchased from 133 neighbors. For the grain sorghum regions, for the 1957-59 weight period, this proportion was dropped to 45 percent to reflect the increasing use of hybrid sorghums which were assumed to be purchased through commercial channels. Procedure a Unadjusted expenditures for all grain seeds were subtracted from the unadjusted expenditures for all seeds. Then, unad- justed expenditures for the minor grains was added to this figure to obtain total unadjusted expenditures for all seeds except the major grains. This total was adjusted to Census using the F.I.B.'s adjustment factors and the result was deflated with the index of seed prices from the Farm Cost Situation. Procedure b (which follows below) was used to account for both the neighbors and the processors margins as a proportion of expenditures. An arbitrary assumption was made that farmers paid 10 percent over the grain price for open pollinated corn and wheat seed which they purchased from neighbors. These procedures resulted in marketing margins for: wheat in all regions, red clover and/or lespedeza in all regions. Based on 1955 data from the F.I.B., expenditures for seeds other than major grains were divided into two classes: Highly processed: grass seeds, vegetables seeds, green- house and nursery supplies. Lowly processed: minor field seeds and potato seed (including winter wheat). A weighted overall marketing margin was then calculated by weighting the red clover marketing margin with the proportion of highly processed seeds and weighting the wheat marketing margin with the proportion of lowly processed seeds. This overall marketing margin for all seeds other than major grains was then applied to the deflated expenditures to obtain the constant dollar value added by non-farmers. Procedure Let a - 134 b - Procedure for determining seed processing margin commodity grain price bu. of seed purchased fraction of purchased seed bought from neighbors fraction of purchased seed bought from processors price paid to farmer for seed price farmer pays processor for seed 135 mousuapcoaxo .m_.H...ar mo co«uuoaoua m an cuppa o=~m> I n\w I oawwmw H MMMMHWW Maw H I muomnoooua one muons—mac: .3 women 03.3 I HAS $.73 + 336M— CC 9 x m encamoooua an. .aosawame Agavxxc Asa-uae + Acevxxc As.ovx.e Illlll(\llll\ IIIIII(\IIII\ IIIIII(\III|\ Illlll(lllll\ uonnwuuc ou 03 up an eaves asam> Hammooono pommooouo uoanwfioa f Scum .uon d 3 cameos .an weak I Eouw Jon a campus :5 Hawk 4 < muonswaoa ou esxuma w .muOmmoooua ou maxums w H I meadowpcoaxo .m.H..m I wovranmaém— Cc ceaumsoc eaHHAHQESm a x a u x e .3332: a: . Tao: xi + Assoc as; .Iul..I\||I\ muOmmoooua uonnwfioa uoonwwoa I O O Q Eoum qu c rsown won 0 Ou v scram1 muoocwuoa Ou muuauuvcoaxm muommoooua cu mouauaccooxm "cone .oofiua u0uo>oao Eoum enzyme ucoouoa 0H muow uonnwwo: maamm< Avoaawucoov n ousvmooum 136 Details §2£,Maig£_crain Seeds The details concerning the tabulation of major grain seeds are dis- cussed in procedures f to m which follow below. Except for hybrid corn, there were no data available on the price paid to farmers for grain seeds. An arbitrary 10 percent was added to the grain price for open pollinated corn and wheat. The commodity price was used for the remaining grain seeds. Procedure f: Hybrid corn seed Purdue Research Progress Report 272 indicates that the retail price figures in Mr. Punk's letter30 are list rather than sale prices. To adjust for discounts,find Punk's weighted average price and adjust it down to the U.S. average price shown in the June 1965 Agricultural Prices. (The Agricultural Prices figure is for regular flats, whether for double cross or all types was not stipulated on the questionnaire, per Mr. Stok- sted, SRS.) Est. Bu. Planted, U.S. Single X 2,553,200 @$22.00 I $ 56,170,400 Double X 8,297,900 @313.70 I 113,681,230 Triple X 1,914,000 @$15.50 I 29,667,000 12,765,100 $199,518,630 Weighted average I 199,518,630 I $15.63 12,765,100 Thus, the average discount I 15.63 - 12.20 I 3.43 or as a percent 3.43 I 21.9451 (loo-21.945 I 78.055) 15.63 Thus Punk's sale price becomes: Single cross ($22.00) (78.055%) I $17.172 Double cross ($13.70) (78.0552) I $10.694 Triple cross ($15.50) (78.0551) I $12.099 and Punk's value added per bushel becomes: Single cross $17.172 - $3.47 I $13.702 Double cross $10.694 - $1.41 I $ 9.284 Triple cross $12.099 - $1.43 I $10.669 30Letter from Thomas F. Funk, Graduate Research Assistant, Depart- ment of Agricultural Economics, Purdue university, dated May 25, 1967. 137 Procedure f (continued) These values added weighted by the quantities planted: Bushel Single x 2,553,200(? $13.702 I $ 34,983,946 Double x 8,297,900 @ 9.284 I 77,037,704 Triple x 1,9144000 @ 10.669 I 20,420,466 12,765,100 $132,442,116 £23762410018u 2 $10,375 I average value added per bushel. Value added as a proportion of the sale price I §10.375 I .85041 I852 12.20 Therefore use 85% of sale price as value added by non-farmers. The average weight period sale prices were derived from F.I.B. records by dividing expenditures by quantity purchased to ob- tain a price per 1,000 pounds. This average sale price was multiplied by 85% to obtain the value added per 1,000 pounds. This value added per 1,000 pounds was then applied to the Farm In- come Branch's quantity purchased to obtain the expenditure in con- stant dollars. Procedure g: Open pollinated corn seed F.I.B. assumes that 5 of the open pollinated corn planted is purchased. To obtain the Q of open pollinated planted they multiply the acreage planted by the seeding rate and subtract the Q of hybrid corn planted. F.I.B. expenditures were divided by the Q purchased to ob- tain a weight period price paid for open pollinated corn seed. The price received by farmers for open pollinated seed was assumed to be 10 percent above the commodity prices which were obtained from Don Durost, FPED, ERS. (Statistical Bulletin 273.) The difference between the price paid and the price received was calculated and this weight period marketing mar- gin was multiplied by the quantity purchased to obtain the value added in constant dollars. Procedure h: Spring oats seed The F.I.B. computes quantities planted by multiplying seeding rates (from Ag statistics) by acres planted (from 138 Procedure h (continued) the crop report). They then, multiply by the fraction pur- chased and prices paid for oat seed from the June Ag. prices to Obtain expenditures. To obtain weight period average prices, F.I.B. expenditures were divided by the quantity purchased. To obtain the processors marketing margin, the grain price was subtracted from the price paid by farmers for oat seed. This weight period value added per bushel was then multiplied by the bushels of spring oats purchased to obtain the expenditure in constant dollars. Procedure 1: Spring wheat seed The quantity of spring wheat seed purchased was obtained from the F.I.B. Their expenditures were divided by the quan- tity purchased to obtain the weight period prices paid. The price received was assumed to be 10% above the commodity price obtained from Durost. The difference between the price paid and received was calculated and this weight period processing margin was ap- plied to the quantities purchased to obtain the expenditure in constant dollars. Procedure 1: Soybean seed F.I.B. expenditures were divided by the quantity of soy- bean seed purchased to obtain the weight period price paid by farmers. Durost's commodity prices (Statistical Bulletin 273) were assumed to be the prices received by farmers. The dif- ference is the weight period value added processors. These factors were applied to F.I.B.'s quantity purchased to obtain PQ's in constant dollars. Procedure k: Cottonseed for seed F.I.B. expenditures were divided by F.I.B.'s quantity pur- chased to obtain weight period average prices paid by farmers. The commodity price received by farmers was assumed to be the price received for cottonseed for seed. The weight period value added per 1,000T was computed, then applied to the Farm Income Branch's quantity purchased 139 Procedure k (continued) to obtain expenditures in constant dollars. Procedure m: Barley seed The bushels of seed planted and the proportion purchased were obtained from the F.I.B. They obtain bushels planted by multiplying acres planted by the seeding rates from1Agricul— tural Statistics. Weight period prices paid were taken from the June issue of Agricultural Prices and weighted into regional averages. Durost's commodity prices (Bulletin 273) were assumed to be the prices received by farmers for barley seed. THE FEED INPUT Feed Grains, prOportion goigg through commercial channels An abortive attempt was made to estimate these proportions by start- ing with the following procedure: Let q I Feed grains fed (from Production Resources Branch, FPED, Statistical Bulletin 337 and unpublished worksheets). w I Feed grains fed on farms where grown (from SRS, Statistical Bulletins 115, 208, 311 and 404). Then q - w I Feed grains purchased from all sources. However, this procedure produced negative purchases for the 1957-59 per- iod for two regions, the Lake States and the Northern Plains. In an effort to pinpoint the source of the discrepancy, a tabulation by states was made for the 1964 crop year. This resulted in negative purchases for the states of Michigan and North Dakota and positive pur- chases which were obviously too small for many remaining states. A con- ference with the Feed Grains Branch, SRS, indicated that the figures for "Feed grains fed on farms where grown" were obtained by subtracting what farmers report as sold from what they report as produced. These data were obtained from questionnaire C.E. 2-308, a copy of which appears on the page following. Labeling this residual as "Feed fed on farms where grown" is suspect for the following reasons: ‘ PLEASE MAIL BY _ nuts I, was FARM REPORT U. S. DIPAIIMIN' OP AGRICULIUI Statistical Reporting Service Please "a necessary sometime in nuns and address it's about that time... when the busy Sprin aessonlreeps us jumping. Your report will help us give you Ip-to-rhte in- formation which will in turn help you in planning and marketing your crops. This service is possible only with your valu- able help, a thalta a lot. Please reuse-ber to: 1. Read the special instructions. 7. Mail your report pmaptly in the enclosed envelope which needs In steep. Respectfully, flfl. G. D. Simpson Chairman. Crop Reporting Bond P. S. Individual reports are kept confidential. "Fans With Facts’ ’ a SPECIAL INSTRUCTIONS e I"utter dash (—) for the questions that do not ap- ply to you locality. On questions relating to your operations, enter 0 when zero or none is the answer. 140 C.E. 2-308 N.J.,alle,lnd.,|l|.,Minn.,lewe., N.Oah.,S.Deh.,Kens.,V.Ye. In... A..." n... Questions For The W Feral er Ranch You Operehe h" ----------------- ---------------- ................ ............... .,..FlELD CROPS? ..... CORN produced on this farm or ranch last year (1967 crop) - 70 lb. ear or 56 lb. shelled BUSNELS CORN of 1967 crop sold and to be sold - 70 lb. ear or 56 lb. shelled BUSNELS OATS produced on this farm or ranch last year (1967 crop) - 32 pound BUSNELS OATS of 1967 crop sold and to be sold — 32 pound BUSNELS BARLEY produced on this farm or ranch last year (1%7 crop) — 48 pound BUSNELS BARLEY of 1967 crop sold and to be sold- 48 pound BUSNELS MN produced on this farm or ranch last year (1967 crop)- 56 pound BUSNELS soasuuu GRAIN of 1967 crop sold ""- ' ' and to be sold — 56 pound EUSNELS HAY of all kinds harvested on this farm or ranch last year (1%? cropl- -TWS - TONS HAY of 1967 crop sold and to be sold CONS MILKED on this In or reach yesterday - NUMBER ALL MILK COWS on this farm or ranch yesterday (both gland in milk) - MILK PRODUCED on this In or r-ch yesterday (Report only one day's { POUNDS 1 production) Report in either GALLONS EGGS produced by your flock yesterday - NUMBER HENS and PULLEIS of laying age in your flock yestir: ? CNICKENS EATEN during the NUMBER past month that were produced TOTAL POUNDS on your farm or reach liveweight EGGS EATEN and used for cooking during the last 7 days that were produced on your farm or reach - NUMBER lNENS and PULLETS of laying age that died during the past month due to disease, accident, exposure, etc. - NUMBER 141 I. No account is taken of change in stocks. II. There is an implicit assumption that farmers report grains under CCC loan or purchase agreement as "sold or to be sold“. The discrepancy suggests that this assumption is unwarrented. Legally, liquidation of grains under CCC loan is a foreclosure of chattel mortgage rather than a sale. Purchase agreements also lack one of the two re- quirements of a legal sale contract: The CCC offers to buy but the farmer does not accept the offer at the time " I... ..l' s ——-. F" the purchase agreement is signed. Thus the purchase agreement resembles an option to sell rather than a sales contract. III. If a cash grain farmer is asked "How much grain was sold or will be sold", this is tantamount to asking him "How much will your income be?" If this question is asked by the USDA, a Federal agency, how is the farmer to know that the Internal Revenue Service, another Federal agency, will not review his report? Thus it is prudent to expect some grain farmers to make a conservative report, especial- 1y as to their anticipated future sales. Statistical Bulletin 268, "Grain Transportation Statistics for the North Central Region" estimated the sales of feed grains by country elevators to farmers for the calendar year 1958. Thus this year was chosen to attempt an adjustment that would arrive at a plausable estimate of feed grains used for feed on farms where grown. (Sheets 91.) Unfortunately, this study was regionalized in such a manner that data on complete states were available only for the states of: Iowa, North Dakota, South Dakota, Ne- braska, and Kansas. Following the bottom half of sheets 91(on the pages following) q I total grains fed t I feed grain purchases from local elevators by farmers q-t I u I grains fed on farms where grown plus purchases from neighbors. 142 :3 Rain on: 8...: use... to. a as: unle- an: as so. 3...: .2. as a: a one. one no..- an an: Re. Re. aa.aa no... I8 RS 4" p I n I a s u u I v a I an a a r lasso ooo—lll' .ssoog snagged sus-amass Loco-u was .ufiea Isaac...- sEu sandman-n an: . On an no: as: quasar—.5 axons use... I»: on. ‘8 an 3 on. .as_. .— .!olsu us 58 a .3. aausuu vs: no '8 ends... we: loo-bone use: noun seesaw—an :95 anal—h decor QWN saueansn aauuuaauauu an: .10.th .aoabl anus—=0 guns: on» use accuses-n sausages“ can Iowan-$0.: sousea 3‘ 034a — — h «each 2 .8. a: \\\\\\\\\\\\ N I\\\\\ S on- R- «3 oz; .8. and: RS 83 a3 .1362 \\\\\ \\\ \\\\ I\\\ 1 RS «3.2 5 3“ Stan 813 as: «3 R2 Ra Ra cum; 3" a2; :3 a: N \ R R i semi 08.2 of. finance R3 at.” ¢~o 30...: or? 31? 33 $93 \\\\ «.3? Rod... ~oo.3~ oanS no.5 loo RS H llllla if??? n u as a quIeIo-s n n a a 0 u e s «\v a u a assoOOOu lad-ness Good nae-ans 000‘ noun-u node 000 a» venous arou- uuebaon. van->«Aev In: .uuu :vuaeensu: cash-.— nso~ uuu sues.- lluan hangs-pea alus~ no aband- caonu shuns awn: eases-enua on gene»: aunts-enu- ease" have: .0 uses a« unease neon-2:: :om nude—June so vs: any use: 25s.; use.— eaanouam use also.— usenuuzn canons lush nausea-n hon veep veal—:4 n4 enasno 1 I .31" :3 Nua— .»a: .p‘ In ion-La .3933 Jan .3: Juan n o n 1 . a o a s L n a a o q a s .— u n o a u a n eaouuu lush .asum .anouu venue Be: sunse- as: Eu suav- Huflkaba< nou ushssaaev lua~ .uuu :culeevehs vaensh naoa nuu shuns alhsu huonevsna eluau no o sflau¢e use... ease... alas“ sun-assau- au van->3 suns-sense aesoa have: as case n« ume-cu Anounueia 1." m nude-June no we: a8 use: 3:...- uesn :33: new ease.— a-saua.‘ nausea an: seas-aa- 53 use: 1332.4 an euaaau usage—Goa.— .nda son Nom— .hal .hsfl flu .nunlun nausaufln Jan .12.: Jean no «a . a a use: use a a an ego scan neon-n ans- ..aum .aoouu seaweeds: ceases scan £55.. who: 33- Eksaa (Fan guano an no so}... 144 n8- 38 3.. 3.3 3cm 3 cue. no. no: 5.!- on: 3A us. can; no: :0 one: o QNO. nhN hens E: snag can So. 3.12 an... Ian R: a I n I a a s a I v u s e.- u u v Alla-cg coaul ._.sOOOq Inouogn loaned-a ~3qu an: in.» tonne: no: couscous... so: On a» ham.— :5. .23.: 3 $8 a :5 loan sea-nouns en‘su- no... uses. :2.) luau as ueu Ina-kl was; on. .03 an 3 on. anon-buds «soon so: cassava... ensue OwN lauesdan «luauauoaum wum< .nueaun dose-a aa.aaa-u nuns: ens sou asuanossn sus-«asum antennas-nan“ sauna acumen—Souk sense- :8 ABM — — .— d‘OF 3 .3. in; \\\\\\\\\\\\\\\\\\ “3;. an. NR 33 2.13 3.2 .9.- id: 13 I a \ . . . h :2 So to no \I\\\ \\\\\\\\\\\ t\o= 2 En can ‘8.— n~o R an. flu a... was one. a no. «2 :~ on 3n 2n 3...“ Sné no... Ellen RS a \\ t 33 Re. R: 2% 2: SE \\\\ mu :3 ~23 .36.. 2.13“ .2. I8 RS .— I .- . x q a a e a . I u} a a I 3332 .~.-- coon 2......» eooal ._euuau nous 000 3 venous Insaaoe sen-L3: seen .00 easeaeh: sauna.— asoa uou nasal-E: ask: as o sulfide axon. eases alusu suns anus on vanes: uses-snu- ansoa Lesa: aw sue-Au unsung-q 3 m message so as: can can! sea-Ln as: easnunan Iev sass sassy—an ssaoue seen ”-13-! veal—ave :4 aha-do Irina-ire .5n. .8 32 in: is. .a: .23-.5 .351.- .:n .33 2.3- a a o a u a n e s o s u a a sane: Isa.- .usua .180 seen a was: euuaoa Isl <§a ISO a scan- Ehgac «Hz :30 a! no so... 145 92 8% one 2? 1.8 0 sun. now no: 51- one. 2 2o Men... 3: «3 RS 3 30 ~o~.~ no.3 nun—lea RS s3 30 5o$~ ac..- Ioo R: u a s n I a a a a I v a s I a .- v lasso Oooal' .aaOOOa IeoaOsA Eon-nas- uouu: one.— in.» sue-anus. so: Ewe-vase as: . on o» thA no... casinos—S sion. one.) sane.— Qn. .usn an e» on. do... a .au-Esu no use a .vsu saut- vees as v: en‘s: wash ensue-sue uses.— eouu nus-nukes sus sail as»: ofi Salaam 13:21; 3.: £3...- .uoaul :Luuen 3.3: on» no. a-uusoa-fl sun-sass.» nozsuhunsaeak aa.aaa nouns-5o...“ unless a»! «can —I _ 4 1.8 2. .8 So... \\\\\\\\\\\\\\\\\\ 3“ an 1. s: 2.»; 2a.. .2. 5: as l. \ . . «R as at a \\\\ \\\\\\\\\\\\ .2 at u? n8 5 a 8n 3 .2. as a: 3. 3.. 3a.: 8: 23.. 2% R \\\ t a: 33 33 as... 235 .2. .333. RS on: .3. $3: :3 23.8 as 23 \\\t z :8 :1: SEE 8.13. .5 l8 R: "H. lautwldllr , n a II a .uouIeIuIa x n a a D u e - N\v v u a eneuOOOu sus-ass 000‘ eue-asa ouna Leauau anus 80 3 essays are...- Iaebaou sea-51' luau .uQu Ives-even: zen: used 000 seen: elm-u hashes-nu «an: so a shad—a one: shoes ash-u anus-sauna 3 cans»: sun-assua- saaeu menu: no case ea rue-no ”nausea: 3 m asap—Suns as teau Lou vase aaqauu use; seasons.— Iev also.— easaouam venous lash oeuaee‘ he. was: .3331 an euasau season‘s: .3! son ~02 .3: £8 .lu .Aout- uuuenuln .1: 2:: Idea! nuns .ssassn essay-none ease eases—sass .usuw lfluee-nseA-E sea-asses: quali- 33- ulna-.2: :3 lie on! :e .8: 146 II?- III.. Io III I393 Icon 23::- 193 I9: «330.23 :95 QQN qua-Aaan uloau-auluw nun—L . 1 01¢»- gl hand :nw 509.— 38 N: (No. 35.4 no... 5: 30A 3 So. uooJ .2. 5° 23 NR 30. nnOIOd .9: E50. shad $~ o3. :3; .8. I8 .32 H I I n u I s I u u v a I I.- I h ' Ill-o» coaulll" .3508» 1.00084 III-nu:- Iaaunu unfiu £th loneuII I9: :32... In: . On an khan Eu volcnohsn n19: IIIAI Inn: on. .08 an on on. out». a .ILIIL: 5. .30 A .1: .Ioautm ”quIIu Auuo: Ina he ugh-o... Inaunaqum - .595 Ioauuacoha I922; III.— NOtu — — F Adana. R~ .3. Sn; \\\\\\\\\\\\\\ «3 E «3 - o3 - 8o; “2.: .2. 5.3: a: n . . n v . . on o3 mo 2 \\\ \x \ \\\\ M V \ cu 8o 33 RR man ~ 3.. 3 2:9 .20 32 Rm 30. 30.2 8% 3? Ron 4: \\\ x 2; 3c 82 Info 23.52 no: .333- RS 3. ”3. 23:: a: 2.3 an 3.. \\\x 5 3n :2 $3.3 81* .2. I3 :3 I57??? a I II I auunonqu n n a a U u I I «\e c I I 3882 Alta-._.... oooa .«..-=. coca .3»qu In: uuu a» vzouI aloha quIIou IILIIJII In: .03 141:. II: nun, 9:5. III: huoquEI III: In I chi—«I 5.9; In...) an: Inn-loouul nu Iago»: 2:: ~35. no nI-I In Inc-cu 5.15:5... 3 m u-«vsavuo no :5 I3 v.3. Ina-nu so... can»: .Iv 3.3 I223: vIIIuI In: P: In I3 :5 ‘33:: nu Iuazo 1033.15 .33 aon $3 .»I: :8 Eu £23... .3115 .13 .3: .quu m u n < . n I u I .— n I I o a u I u I a o I a I a can»: In: .aIun .Iaouu 1IIm .- _ZIE Inn-5. lul— Iulun Exkfl—ZQ «had “was an: no u...- 147 We can logically expect a substantial proportion of purchased grains to come from neighboring farms in regions that have grain surpluses be- cause: 1. The feeder can avoid the local elevator's handling charge of about 10¢ per bushel. 2. The neighboring grain farmer can avoid the cost of deliv- ering the grain to the local elevator. The feeder can be expected to purchase a larger proportion from the ele- vator: a. In grain deficit areas, b. Shortly before harvest since the neighboring grain farmer would not normally have grain on hand at this time. Typically the feeder can be expected to line up his anticipated needs at harvest-time and purchase as much as possible from neighboring grain farmers. TWO factors limiting this activity are likely to be: 1. The feeder will be conservative in estimating his needs in order not to purchase an excess. 2. The feeder and the neighboring grain farmer may not have sufficient storage to handle inventory for the entire feed year. Thus we can expect substantial purchases from neighboring grain farmers in the post harvest season and purchases from the elevator in the prehar- vest season. We would expect the preharvest or evening-up purchases to run fairly constant in relation to total feed fed. For some of the North Central States, we can obtain this relationship from the two publications indi- cating the sales of feed grains by country elevators to farmers: 1. "Grain Transportation Statistics for the North Central States", by Kenneth R. Farrell, June 1958, Data for the calendar year, 1954. NAL #280.3S9 F24. 2. "Grain Transportation Statistics for the North Central Region", USDA Statistical Bulletin 268, August 1960, Data for thecalendar year 1958. NAL #lAg84St These purchases from country elevators as a percentage of total feed grains fed are: 148 State Percent purchased from elevator 1954 1958 Minnesota 9 Ohio 6 Indiana 4 Illinois 9 Iowa 11 13 Missouri 13 North Dakota 4 11 South Dakota ll 19 Nebraska 24 19 Kansas 29 32 Based on the foregoing logic, (following worksheets 91) the residual "feed grains fed on farms where grown" was estimated for 1958 as follows: 1. The crop report questionnaire asked farmers to report only on the previous year's crop. Thus, the change in inventory from one year to the next is not accounted for. If data were available it would be desired to adjust for change in inventory of non-government stocks on farms. Since data for 1958 were not available to make a separation, an as- sumption was made that one-half the change in inventory was "free" stocks. An adjustment was made for feed grains used for seed on farms where grown. Concerning CCC activities, since the crop report was made on March 1st, logic indicates that the corn and sorghum crops should be treated differently from oats and barley. At the time the crop report survey is taken, the previous year's oats and barley crops have been mostly disposed of except for CCC reseal. Much of the sorghum crop is either sold or under CCC loan. Much of the corn crop is still uncommitted and can either be sold, fed, or put under CCC loan. To obtain a reasonable residual for "feed grains fed on farms where grown" trial and error indicated that CCC activities should be handled as follows: For corn and sorghums: Deduct the farm stored portion under loan or purchase agreement. For oats and barley: Deduct only that rescaled under CCC loan. 149 Referring to sheets 91, by subtracting p from u, we obtain the quantity of feed grains purchased from neighbors. Comparing states, the residual p for 1958 appears plausable for the states of Iowa and Nebraska and low for the Dakotas and Kansas. The causes of this divergence is believed to lie in: l. the questionable change in inventory figureS, 2. differences in time periods being compared: a. 1 October 1957 to 1 October 1958 feed year b. l) 1957 corn and sorghum crops 2) 1958 oats and barley craps c. change in inventory: 1) for corn and sorghums - 1 October 57 - 1 October 58 2) for oats and barley - 1 July 58 - 1 July 59 d. purchases of grain from country elevators, 1 January 58 - 31 December 58. Except for change in inventory, the same adjustments were applied to the 1964 figures for all states. Data for CCC stocks on farms were available for 1964 allowing an accurate estimate of change in inventory of free stocks. Although some of the state estimates were obviously inaccurate, the regional averages fell in the range whid: would be expected. Eggg Grains: ggigg_charged for local handling Costs of storing and handling grains by local elevators was reported in ERS-288, "Costs of Storing and Handling Grain in Commercial Elevators, 1964-65", by Allen Schienbein. These figures do not allow for profit, however, they include more storage than is required for inventory. Thus the two factors were believed to counterbalance. Although the study is partially regionalized, the charges vary little from 10¢ per bushel and a decision was made to use this figure for all regions. Conferences with Schienbein and with Frances Yeager (Farmers CoOperative Service) indi- cated that this figure is applicable to both 1947-49 and 1957-59. The following counterbalancing effects are believed to have been operating: labor costs have increased, mechanical efficiency has increased, compe- tition has increased, with a net effect of no change in the charge to farmers. 150 By-Product Egggg_ Before proceeding further some definitions are in order: Supplement: Materials used to supplement grains to im- prove the balance of the ration. Processor: Buys oilseeds, fish, copra, etc., through the wholesale market, processes these into the various by-products such as millfeeds, tankage, etc. These by-products are sold to either the formulator or through a job- ber for eventual retailing to the farmer for feeding as such. Formulator: Buys various ingredients on the wholesale market and mixes them into a livestock feed. These can be further divideded into: Formulated supplements: These feeds may contain grains but are intended to be fed as supplements to grains rather than as complete feeds. Most state laws require the formulator to tag these feeds indicating the analysis. Complete feeds: A mixture of grains and supplements which forms a complete balanced ration. Reasonably good data on the quantity of supplements fed were availa- ble on a national basis. The remainder of the data required varying degrees of "manufacturing" before they could be utilized. The available data indicated that the feeds should be divided for analysis as indicated in Figure 22. Fragmentary data on the value added by formulators was available from the Census of Manufacturers. However, these data required considerable processing before they could be used. Oakley Ray's31 discussion of double counting in the census figures ap- pears on the following page. The 1961 relationship between tons produced and tons of ingredients used was utilized to make regional adjustments for this double counting. Referring to Figure 22, the Census of Manufacturing supplies the value 3IOakley M. Ray, Director of Market Research, American Feed Manu- facturers Association. WHAT ABOUT THE The Census Bureau recently published the resultso fa survey of 1961 formula food product ion. The table indicates the it by states. Total production re- ported was about 50 million tons. All of the larger feed manufacturers were included in the survey. A sample of the dealeromixers and smaller manufac- turers was contacted. Estimates for the mallet operations were developed from the sam . The get-sus Bureau statistics cannot be compared with AFMA estimates of formula feed production. because the Census estimates include the tonnage of dealers who mix supplements and concen- trates into complete feeds. AF MA statis- tics inchide only the output of those who manufacture feed from basic ingredients or with the help of a highly concentrated premix. llrnltotlonsofDoto The tonnages listed in the table include an undetermined amount of double count- ing because the supplement and concen- trate tonnage of the feed manufacturer is . as well as estimates of the com- me feed tonnage of the dealer- mixer. double counting would likely be test in the Corn Belt where an esti- mated 60 to 65% of the tonnage sold by feed manufacturers is high protein con- untrates and supplements. The Census Bureau survey did not in- clude the tonnage of concentrates and Implements produced by small manufac- turers. This has been added to the ques- thnnlire for 1962 feed production. The data listed in the table include 0.5 million tons of custom milling reported by large manufacturers and 8.4 million tons produced by the smaller operators. The Eton indicated that the coverage of mo- mills is believed to be low. The questionnaires returned indicated the production of 2.9 million tons of com- lete feeds “for own feeding operations" It addition to the tonnage listed in the faith. The 2.9 million tons was not in. cluded because it appeared that many of themillsproducingfeed fortheirown FEED MARKETING SEMINAR Pick Congress Hotel 5m. 25-26. 1963 mm seen-ours (American Food Manufacturers Association, Vol. 1, No. 4) 151 CENSUS REPORT ON FEED TONNAGE? OAKLEY M. RAY Director of Market Research animals were not contacted on this sur- vey. The discussion above indicates some of the limitations of the data. This is the first feed census which has attempted to in- clude feed produwd by small companies where feed is only a sideline such as ele- vators. hatcheries. machinery dealers. etc., lnduuryCansusVlllllolrnprovsd m Census Bureau recognizes the weaknesses in this survey and expects to eliminate some of them in future years ' The questionnaire for the 1962 survey has been substantially revised. The size. di- versity. and the rapid changes which are occurring in the formula feed industry make it impossible to have, a completely accurate census at a reasonable cost. A number of feed manufacturers corn- plained to AFMA about'the time required to complete the 1961 questionnaire. A substantial portion of the total time was needed to answer the questions concern- ing quantities of the various feed ingredi- ents which were used. AFMA discussed the problem with Census Bureau repre- sentatives. They agreed that it is not necessary to collect ingredient data each year. The questionnaires for 1962 produc- tion have been mailed by the Census Bureau. and ingredient questions have been omitted. Complete Report Avail-ble The Census Bureau report also includes the following estimates for 1961 food pro- duction- l. Tonnage of complete feeds pro- duced in each state by all manufac- turers and dealer-mixers. 2. Tonnage of total formula feeds pro- duced in each state by large manu- facturers. 3. Tonnage of complete poultry feeds by type produced by large manu- facturers. by state. 4. Tonnage of complete livestock feeds by type reduced by large manufacturers. y state 5. Tonnage of poultry supplements and concentrates by type produced by large manufacturers. by state. 6. Tonnage of livestock supplements and concentrates by type produced by lar rx manufacturers, by state. if you wt one or more copies of the complete eight page report. send 10 cents wr copy to the Bureau of the Census. ashington 25. DC. Ask for Current industrial R .Series: M208(6l)-1. entitled Poultry and Livestock Feed Pro- duction.l961. Formula Peed Produced in 1961 by Large Manufacturers Plus Complete Peed Produced by Smafl Manufacturers md Dealer- Mixers. As Estimated by the State Thousand Tons State Thousand Tons Texas 4.269 Kentucky 760 Iowa 3.513 Florida 726 lllinois 3.198 New Jersey 635 California 3.050 Delaware 579 New York 2.877 Maryland 441 Ohio 2.387 441' Tennessee 2.261 U 437 Missouri 2.199 Colorado 349 lndiana 1.980 South Dakota 339 Minnesota 1.675 Louisiana 309 Georgia 1.644 Massachusetts 210 Pennsylvania 1.533 Idaho 158 North Carolina 1.426 Montana 152 Wisconsin 1.244 New Hampshire 135 Nebraska 1.108 North Dakota 118 Virginia 1:083 South Carolina 118‘ Mississippi 973 Maine 1 16' Alabama 966 Connecticut 85' Oklahoma 918 Arizona 80' Kansas 916 Wyoming 32 Washington - 893 West Virginia 32‘ Arkansas 845 Nevada 5 Michipn 820 New Mexico 4' Vermont 767 Rhode island 2' U.S. TOTAL 49.819 "l'his tonnage includes only production by large manufacturers. Production by small manu- facturers and dealer-mixers was not publ 'l'his tonnage includes only production by small manufacturers and dealer-mixers. Production by large manufacturers was not publ' “IV I”! 152 added for the preportion: (c)(a + b) i.e., the feeds fed as formula feeds. The quantity figures for total supplements can be approximated from protein supplement requirements. These are not published but can be developed from unpublished data as outlined in procedure d which follows: Procedure d: Supplement requirements The Production Resources Branch takes Mr. Clough's32 national total supplement fed figure and breaks it down by classes of livestock. They also have unpublished figures on animal numbers by states. Thus the total supplements required can be obtained by multiplying the U. 8. average supplement requirement by the number of animals fed and summing over all classes of livestock as illustrated belowz Let h - Year: 19(59), 60, . . ., 68. i - class of livestock: 0110 for milk cows, 0201 for heifers, etc. k state: 1 for Maine, 2 for New Hampshire, etc. A - number of livestock in class i in thousand head. Ex. Ahik = A(S9)(0110)(2) = number of milk cows in New Hampshire in 1959. Let B 8 Annual U. 8. Protein supplement requirement in thousand tons. Ex. Bhi - B(59)(0110) - 1000 tons of protein supple- ment fed to all milk cows in the U. S. in 1959. then: 11 :EBhi - national total protein supplement fed to farm 1.1 livestock, year h. ZAhik - national total number of livestock, class 1, year h. Bhi I national total protein supplement fed to class of livestock 1, year h. 32Malcolm Clough, Head, Feed Section, Economic and Statistical Analysis Division, Economic Research Service. 153 Bhi ipr-- = tons of protein supplement fed per head, k hik class of livestock 1, year h. Bhi Ahik°-7f_7' = tons of protein supplement fed to class a; hlk of livestock 1, state k, year h. 11 Bhi IE ‘Ahik'-7f__' = tons of protein supplement fed to all I1 I: hik farm livestock, state k, year h. k 11 ll Bhi I: :: Ahik.iizhi; = tons of protein supplement fed to all farm livestock, Northeast region, year h. k-l i-l k This method would be too time consuming if it were necessary to start from scratch. However, the Production Resources Branch had the animal numbers on data cards for the period 1 October 1959-60 through 1 October 1967-68. 154 Subdivision of the Livestock Feed Input T F”_ __"ll“ “mm“ ' ”“”"‘Q d H 2(31..;~-;u:~a :3 j +2.6 H C2 ._.—._....3‘ L. CQ I o» E Q) 3 CL 3 Q- :1 J: fed in formula feeds 3 -<—————————-aA-————————fi-1< b 4444:: Figure 22 155 Method g£_Compilation An outline of the general approach used follows: I. Estimation 2£.U- S, Totals: A. Total supplements fed: Mr. Clough made calendar year esti- mates for the Farm Income Branch for the period 1949 to date. For the period 1939-48 it was necessary to use feed year estimates from Statistical Bulletin 159, Revised June 1962. Supplements formulated: Mr. Clough also made calendar year estimates for supplements formulated for the Farm Income Branch for the period 1949-date. For the period 1939-48, the ratio: supplements fed in formula feeds as a proportion of total formula feeds fed, was interpolated through the benchmark years, 1939, 1947, and 1949. Formula feeds formulated: For the period 1949-date, Mr. Clough made estimates of grains formulated which added to B above sums to total feedsformulated. For the period 1939-1948, see procedure c on page 157 following. Supplements fed as such: This is equal to.A minus B. Since the marketing margin on supplements fed as such is computed separately for each ingredient, it was necessary to have the tonnage of each ingredient fed as such. Mr. Clough estimated these figures for the Farm Income Branch for the period 1949- date. For the period 1939-1948, the ingredient tonnages were computed by subtracting the quantity formulated from the total quantity fed. The quantity formulated was estimated by interpolating the composite formula feed formula through the benchmark years: 1939, 1947, 1949. II. Regional Distribution: F. Feed grains purchased: The general approach is discussed previously, pages 139 to 149. The tonnages of feed grains were multiplied by the preportion going through commercial channels to obtain the tons of feed grains fed going through commercial channels. Formula feeds fed: a. For the period 1939-1958: 156 Regional percentage distributions were interpolated through the benchmark years, 1939, 1941, 1942, 1950, 1958, 1959. 1) 2) For the period 1959-1968 the percentage dif- ference between the regional distribution of the quantity produced and the quantity fed was interpolated. These adjusters were applied to the distribution of production for the inter- vening years to obtain the regional percentage distribution of the quantity fed. The percentages derived from 1) above were applied to the U. S. formula feed fed figures to obtain the regional tonnages of formula feeds fed. Grains fed in formula feeds: Percentage distributions were interpolated through the benchmark years, 1939, 1950, 1968. These distributions were applied to the U. 8. quantity of grains formulated to obtain regional tonnages of grains fed in formula feeds. Total supplements fed: For 1) 2) For 1) the period 1960-68: A regional percentage distribution was made from the computer print out of supplements required. (See procedure d on page 152 for a discussion of the computer procedure.) The percentage distributions derived in 1) above were applied to the U. S. tonnages of supplements fed to obtain regional tonnages. the period 1939-1959: A calculation was made of the regional percentage difference between the regional percentage distri- butions of: (a) supplements fed (b) grain consuming animal units fed for the benchmark years: 1942, 1944, 1950, 1960. Interpolations were made through these 157 benchmark years to obtain the regional percentage difference for the intervening years. 2) These regional percentage differences were applied to the regional percentage distribution of grain consuming aninal units to obtain the regional per- centage distribution of supplements fed. 3) These regional percentage differences from 2) above were applied to the U. S. tonnages of supple- ments fed to obtain the regional tonnages of sup- plements fed. J. The tonnage of supplements fed as such was calculated by sub- tracting H from I above. Procedure c: Procedure for determining the quantity of feeds formulated, 1939-48. A conference with Mr. Clough indicated that unpublished data were not available. Various sources of data and the estimates are listed on sheet 8 on the page following. The estimates underlined were con- sidered to be the most reliable and were used as benchmarks. An abor- tive attempt was made to move the intervening years with trade estimates from the publication "The 3.5 Billion Formula Feed Industry", NAL # 389.7F325. This publication indicates an increase in formula feed production from 20 million tons in 1942 to 29 million tons in 1943 and a reduction in production from 30 million tons in 1945 to 25.5 million tons in 1946. These industry statistics are believed to include inter- plant transfers and, at least in the early years, some by-products fed as such. However, there are no supporting data and little theory to justify this drastic increase and subsequent fall in production. There are no data supporting a similar movement in total supplements fed. Thus, for formula feeds to rise and subsequently fall, there must have been a rise and fall in the proportion of supplements fed in formula feeds or else a rise and fall in the quantity of grains fed in formula feeds, or both. Attempts at distributing the residual, "Supplements fed as such" re- sulted in negative figures for one or more regions during this period. Thus a decision was made to reject the industry data for this period, 158 and formula feed production was moved with total supplements fed. To obtain the ingredients formulated, for 1939, the 1939 quantity formulated was distributed with the 1940 average formulas from Cir. 836. This resulted in negative quantities fed as such for some ingredients. This is believed to be caused by changes in composition. Thus, the neg- ative quantities were "borrowed" from other ingredients to leave zero or positive quantities fed as such. By-Product Processigg.Margins Using soybeans as an example, suppose a farmer sells soybeans to a processing plant and purchases soybean meal from his local feed dealer. There is non-farm value added not only by the feed dealer but also by the processor. To estimate the value added by processing the wholesale value of the meal as a proportion of the value of both oil and meal was applied to the price received by the farmer for soybeans to approximate the farm value of meal. The difference between the farm.value and the wholesale value represents the value added by the processor. This processing cost is incurred regardless of whether the feed is mixed and sold as a formula feed or sold as an ingredient to be fed as such. Thus the processing cost was applied to the total quantity of ingredients fed and the formulating and marketing costs were picked up at the wholesale pricing point. Feed Marketing Margins Supplements Fed gg_8uch It was possible to estimate accurately the marketing margin on sup- plements fed as such by ingredient on a national basis since both the wholesale price and the price paid by farmers were available. However, there were no data available to distribute the individual ingredients to regions. Thus, the national total ingredient tonnage was multiplied by the ingredient marketing margin to obtain a national total marketing cost for all ingredients fed as such. This national total expenditure was then distributed to regions with the tonnages of supplements fed as such in each region. Formula Feeds Although prices paid by farmers for some formula feeds were available 159 .353- 32. 05 on op “.3343 2a nob-mum 35.1093 05. .auomeu 25:00 Engage < a “.8350 mass: no 33 usage» 93.8 . 53885.. was up 335° 33: x :25 no no.“ 32605» .3 can noou moo 8.0a unsaved has 058.3. nit. A. Sauna—5.5 953.395" 2.3 398.9% H33. 1. .8m.- 08$ 08.8 8a.? one :36 an: .m a... gown om «manJH 3230355: no cannon Sm." . Anmfifioa .n .5 .2133 . . . 2.00m «.99: Hague-=00 Mo v5 08 aw 80 8 08 ma Sefiafiwfi woos 383on no 833.2%3 a do; one oomfiw 20am 8.85 52:: m Eon ._RNJN m s. H 83.33 a o noon 3% 82. .3 .mm a a a a x g a a a a « ban—63 80m 8m 8 8m mw 08 3 08 mm ooo on 8m mm 08 mm 80 8 08 3 08 E 88 5V 33:8 a 533m m.n 2:. $3 33 SS 33 33 $2” 33 ~43 33 0:3 mmma 3.26m m £36 2.3 08H mamaumnma 303%380 a3 53833 6:: “Soon «Ha-non 160 there were no data for effecting a breakdown of the tonnages to match the price categories. Neither are wholesale prices quoted for formula feeds. However, they can be estimated with procedure e (on the page following). There are little published regional data on retailing margins for livestock feeds. 1939 estimates for farmer's supply stores are published in Standard Ratios for Retailing, by Dun and Bradstreet, Inc. NAL # 386.2 M69. Brensike and Vosloh33 analyzed the Illinois price spreads for poultry feeds for the period 1953-58. Askew and Brensike34 discussed the 1947 value added by manufacturers. However, their conclusions are suspect because they did not adjust for double counting in the production tonnages. Rickey35 made an accounting report based on a survey of several co- operative feed mills. Phillips36 made a similar study of the 1955 costs of operation of selected Iowa feed manufacturing companies. The latter two studies excluded profits. A conference with Carl Voslow indicated that this retailing margin is somewhere in the neighborhood of $10 to 14 dollars in the Delmarvia area. The formula feed retailing margin should be less than the market- ing margin on supplements fed as such since this figure includes trans- portation from the wholesale pricing point to the point of purchase by farmers. Application of the Dun and Bradstreet ratios resulted in mar- gins approaching the fed as such margins. However, livestock feeds are more competitive and move faster than most farm supply items, thus, the Dun and Bradstreet regional ratios were cut by one-half to bring them in line with the levels indicated by the studies by Brensike, Vosloh, Askew, Rickey, and Phillips (see footnotes 33 thru 36). 33V. John Brensike and Carl J. Vosloh, Jr., "Price Spreads for For- mulated Poultry Feeds in Illinois," USDA Marketing Research Report No. 378, December, 1960. 34William R. Askew and V. John Brensike, "The Mixed-Feeds Industry," USDA, Marketing Research Report No. 38, May, 1959. NAL # l Ag84Mr. 35Lacey F. Rickey, "Operating Costs of Selected Cooperative Feed Mills and Distributors," USDA Farm Credit Administration Bulletin No. 56, November, 1949, NAL # 166.2 B87. 36Richard Phillips, "Costs of Procuring, Manufacturing, and Distri- buting Mixed Feeds in the Midwest," USDA Marketing Research Report No. 388, April, 1960, NAL # lAg 84Mr. 161 A charge for bags, tags and twine was also added since the value added by manufacturing does not include the cost of these materials added. Procedure e: The wholesale price of formula feeds For 1961, the Census of Manufacturers current industrial reports in- dicate the tonnages of production of formula feeds and the tonnages of ingredients used by states. These data provide a means for removing the double counting in the production tonnages: ingredients1961 production1961 production1958 - ingredient31958 This procedure, of course, assumes a constant degree of double counting. To find the wholesale price, adjust the value of products shipped for double counting and divide this by the tons of ingredients used in the manufacture of formula feeds. HIRED TRUCKING, FREIGHT.AND EXPRESS The Farm Income Branch had a series of expenditures for hauling milk but did not have series on hauling other items. It was assumed that the 1955 relationship between hired transportation and the value of farm output was a constant that could be applied to all years. This constant was derived in the following manner: The value of farm output (obtained from Don Durost, FPED, ERS) for 1955 (in 1957-59 dollars) was divided between livestock and crops and deflated back to 1955 current dollars: Billion_gollars a. Productive livestock 8,562.0 Index of livestock and livestock product values (May 1962 Ag prices) .90 7,705.8 b. Crops 17,966.2 1,664.1 19,630.3 Less feed for horses and mules 228.4 l9,401.9 Index of crop values (May 62 Ag prices) 1.04 20,177.976 162 Thus the 1955 current dollar value of farm output is: Billion Dollars 7,705.8 20,177.976 - 27,883,776,000 From the 1955 survey of farmer expenditures the cost of transporting items other than milk was derived as follows: Hired trucking 375,136,000 Trucking costs not included elsewhere 33,551,000 408,687,000 Less income from trucking and hauling 65,485,000 343,202,000 Plus other transportation 36,199,000 379,401,000 Less milk hauling (by truck) 185,153,000 194,248,000 Thus if we exclude milk hauling, tranSportation as a proportion of the value of farm output was in 1955: 194 248 000 ———-—J-——J-—_- 27,883,776,000 0'006’966’3 This proportion was applied to the constant dollar value of farm output for all years to estimate transportation excluding milk hauling by regions. BABY CHICKENS PURCHASED The number of baby chickens purchased was obtained from the Farm In- come Branch by the two categories: Broiler type and Layer type. The average weight period price paid was obtained by dividing expenditures by the quantity purchased. Confidential data on the price received for hatching eggs were ob- tained from the Ag Price Statistics & Farm Labor Branch of SRS. Since these data were not complete for all states, it was believed to be more accurate to tabulate the difference in price or the premium paid for hatching eggs over table eggs. The states, for which weight period figures were available, were weighted to regional averages with the 163 volume of chicks batched in commercial hatcheries. This premium was then added to Durost's weight period average price of table eggs to ob- tain the hatching egg price. A hatchability adjustment was made and the difference between the price paid for the chick and the value of the hatching eggs required to produce one chick gave the value added by hatcheries. The hatchability figures were obtained from the Poultry and Egg Sec- tion, ERS. These figures were: 1947 68.4% 1948 68.7 1949 69.0 3/206.l 68.7 1957 71.4 1958 71.7 1959 72.0 3 215.1 71.17 Since k is constant for all years at 1.818 m - i-ij(1.818) - i(l-l.818j) For the early period, j was weighted with the 1957 split between light and heavy: . (3313) (73,350) 25.10335) (13,301) . 18341 BABY TURKEYS PURCHASED The Farm Income Branch had quantity figures separated into light and heavy breeds since 1955. Thus a decision was nade to split the account at that year. Weight Period Prices were obtained by dividing expenditures by the number purchased. The only data available on the hatchability and price of turkey hatching eggs was from the Income Section of the Farm Income Branch. They estimate, for both weight periods, that 1.818 eggs were required to hatch one turkey poult. Prices of hatching eggs were available only for the late period. Since these prices were estimated as a proportion of the price of a 164 turkey poult, they provide a means of estimating the early period prices. MILK HAULING Current dollar expenditures were obtained from the Farm Income Branch and deflated with the average of the two indices: Farm Wage Rates (SRS) Truck Repairs, Parts and Tires The latter index is a composite made up of: 0.91 times the BLS index of prices paid for Auto repair and maintenance plus 0.09 times the BLS Truck and Bus tire price index. LIVESTOCK MARKETING The Farm Income Branch keeps an account on livestock marketing ex- penses which they obtain by: a. For livestock going through commission markets: They multiply the average cost per head for commission and yardage (which they obtain from Ken Potter, Packers and Stockyards Division) by the number of animals marketed. b. For livestock going through auction markets: They mul- tiply the value marketed by the average cost of market- ing per dollar value of livestock sold through auction markets. This rate is also obtained from Mr. Potter, Packers and Stockyards Division. These current dollar expenditures were deflated with the SRS index: Production, Interest, Taxes & Wage Rates. REAL ESTATE TAXES Tax figures were obtained from the following data sources: 1939 Statistical Bulletin 189 1940 RET-9 Farm Real Estate Taxes (an ERS publication) 1941-44 Statistical Bulletin 189 1945 RET-9 1946-49 Statistical Bulletin 189 1950-67 Statistical Bulletin 441 1968 RET-9 165 PESTICIDES Dr. Shepard, ASCS, was consulted regarding the construction of a price index. For the period 1939-55 he recommended that lead arsenate, nicotine sulphate and cube be used as a basis. Prices for the early period came from Agricultural Statistics. Wholesale prices were used since farm prices were not available. Weight period quantities were estimated by Dr. Shepard for nicotine sulphate. Quantities for the other two materials were computed by subtracting exports from production. Weight period quantities for lead arsenate were in Agricultural Statis- tics. Quantities for cube Dr. Shepard took from a book which he authored. For the period 1955 to date, copper sulphate, 2,4-D and DDT were used as a basis. Wholesale prices came from a workbook Dr. Shepard furnished in which he had summarized into annual averages (from 1949) the weekly wholesale price quotations from the "Oil, Paint and Drug Re- porter" (NAL #306.8 015). Weight period quantities came from the pesti- cide situation. IRRIGATION OPERATION.AND MAINTENANCE COST The following indices were used as deflators: a. For the period 1939-48: Thg_Engineering Nggg_Record index of construction costs: Page 144 in the March 23, 1950, issue; NAL #290.8 En 34. b. For the period 1949-63: Figure 2 in the unnumbered publi- cation "Irrigation Operation and Maintenance Cost Trends" U. S. Department of Interior, Bureau of Reclamation, 1964. c. For the years 1964 to 1970, the same index as b, figures obtained from the Economic and Statistical Branch, Divi- sion of Irrigation and Land Use, Bureau of Reclamation. CROP INSURANCE The bulk of this insurance is against hail damage. Current dollar expenditures net of loss adjustments were obtained from the Farm Income Branch. Discussions with Edward Reinsel, Agricultural Finance Branch, and Don Durost led to the conclusion that the index of crop values was 166 the best deflator for this expenditure, reasoning that a farmer insures the potential value of the crop. CONTAINERS Current dollar expenditures for containers was obtained from the Farm Income Branch. These were deflated with an index of prices of selected items obtained from.Agricu1tura1 Prices. For the period 1955 to date, the items included were: stave baskets, burlap bags, open mesh bags, fruit box shook, lug box shook, vegetable crate shook, and hampers. BINDING MATERIALS An index constructed from the cost of baler twine as published in Agricultural Prices was used as a deflator for binding materials. VETERINARY AND TELEPHONE Estimates in current dollars for these expenditures were obtained from the Farm Income Branch. The telephone estimates were prorated to represent a production share. Veterinary expenditures were deflated with the SRS index of prices paid for farm supplies. For deflating telephone expenditures, an index was constructed from the price paid for local service as published in Agricultural Prices.