Normative Estimates of Class I Prices Across U.S. Milk Markets
Pratt, James E.; Bishop, Phillip M.; Erba, Eric M.; Novakovic, Andrew M.; Stephenson, Mark W.
Economists have long considered issues of spatial economic activity, trade, and location values. Among all the various theories presented over the past century, it is safe to say that not one predicts that goods, services, or factors of production must attain the same value at different locations in geographic space. Only under the most extreme conditions, such as zero transportation costs, would it be even conceivable that the same commodity or factor of production be expected to command the same price in two geographically separated markets. With costly transportation, it is possible that two separate markets have nearly the same, or even identical, prices, but there are no theoretically justified reasons to expect such an outcome, a priori. When the question is raised why should the same hundred pounds of milk sell for as much as $3.00 more in some regions of the country than in the Upper Midwest?1 the answer is that I) local supply, 2) local demand, and 3) transportation costs, as well as all those attendant conditions which determine these three factors simultaneously, interact to determine the location value of milk. Indeed, one would not expect that the same hundred pounds of milk have the same value everywhere. The important component of this question is 'how much different' these location values will be. Different prices for identical goods at separated locations is not difficult for the 'man/woman in the street' to acknowledge and should not be a mysterious concept to the dairy industry. Why is fresh seafood more expensive inland, why are grapefruit dearer in Ithaca, New York than Miami, Florida, why are apples higher priced in Miami; supply, demand, and transportation. The U.S. Dairy Sector Simulator (USDSS) was constructed to provide insight into the optimal, efficient geographic flows of milk and dairy products; to provide guidance with respect to efficient plant location and size; and to evaluate the spatial value of milk and milk components across the U.S. Using 240 supply locations, 334 consumption locations, 622 dairy processing plant locations, 5 product groups, 2 milk components, and transportation and distribution costs between all locations, USDSS determines mathematically consistent location values for milk and milk components. The results of our analysis demonstrate that, under conditions which prevailed in May and October of 1995, milk produced in the U.S. has distinct location values at geographically dispersed points of processing. A mathematically derived price surface for milk used in fluid uses indicates that these values have a range of $3.63 from the lowest valued location to the highest in May and $3.99 in October. Locally low levels are found in the Upper Midwest, the Northwest, and the West. Eau Claire, Wisconsin is not the lowest valued location in the U.S. nor do values increase unifonnJy away from Eau Claire or from any of the low valued locations. In fact, the spatial dispersion of values from low to high valued locations is much less than transportation costs alone would indicate. From the low valued locations there is an increasing value gradient generally to the east and, more markedly, to the southeast. Milk used in other dairy products also has locationally distinct values which are much less pronounced than those for fluid milk uses. When compared to USDA's current system of location differentials for class I uses, and holding the total level of class I differential dollars constant, the calculated differential surface is generally flatter than the actual surface, despite the fact that the range from lowest to highest value for the calculated surface is slightly greater. Relative, revenue-adjusted values in the Upper Midwest, much of the Midwest, and in Florida are calculated to be higher than the current differential levels, while for much of the rest of the country, the calculated values are lower than current levels. Some of these federally-regulated areas with low calculated values, such as near Dallas, Texas, are estimated to have location values at or exceeding a dollar lower than the current relative differential. Other state regulated areas, such as Maine, Montana, and Virginia also have differences in excess of one dollar. As with fluid milk uses, the value of milk used in manufacturing varies with location, despite the conventional wisdom that the 'national' character of these product markets somehow means there should be only one national manufacturing milk value. The optimally determined manufacturing values vary much less than the fluid values.
Charles H. Dyson School of Applied Economics and Management, Cornell University