Knowledge that Transforms

Two OBJECTIVES underlie this study: The first is an examination of the relative merits of limited information and least squares procedures in estimation from small samples. The second is the presentation of a quantitative study of the poultry industry, which is an important sector of the farming industry in the United States economy. The selection of this industry for investigation has been influenced by the former aim. It was deemed advisable to choose an industry where currentperiod economic influences on supply might be expected to be strong and where interrelations between demand and supply would consequently be pronounced. In principle, it is in these situations that the limited-information method should be expected to assert its superiority and therefore provide a suitable empirical testing ground for the alternative procedures. The poultry industry of the U. S. A. is a very important contributor to gross farm income, and eggs are the most important component of gross poultry income. Poultry production is widely diffused over the country as a whole, the areas of greatest intensity occurring near large cities, on the Eastern and Western seaboards and in the Mid-West Corn Belt area. Small flocks predominate, a considerable proportion of the eggs and farm-raised chickens coming from flocks containing less than 200 layers per flock. Nevertheless flocks of 10,000 to 20,000 and even 40,000 exist on the Pacific seaboard, and large farms are also to be found near Boston, New York, and Philadelphia. In areas characterized by large laying flocks, the export of eggs to other regions is naturally high, while in the Southern states, where flocks are very small, production is absorbed mainly by the needs of the owner families or by those in the immediate neighbourhood. The industry would seem to fulfil approximately some of the requirements for intensive competition. Producers in general operate on too small a scale to exercise individually any perceptible influence on the marketing conditions for their products, the number of sellers is large, and movement in and out of the industry is not difficult. Relative to domestic production, egg imports (principally dried ones from China) are very small, as also are exports which go mostly to Canada. 1 The writer wishes to acknowledge helpful discussions with Professor J. R. N. Stone and Dr. Burgess Cameron during the progress of the enquiry. For valuable assistance in the execution of the computing programme the writer is indebted to Dr. M. V. Wilkes of the Mathematical Laboratory, Cambridge, who permitted the use of the EDSAC machine, to Mr. J. A. C. Brown and Dr. S. J. Prais who inverted the larger matrices on it, and to Mr. P. R. Fisk, Statistical Research Assistant of the Faculty of Economics, Cambridge, for calculations on pages 57-60. The writer bears entire responsibility for the computations presented. An earlier version of this paper was presented at the European MIeeting of the Econometric Society, Cambridge, England, September, 1952. 37