AN ECONOMIC ANALYSIS OF CLASS I MILK PRICiNG 1N LOWER MiCHIGAN final: hr the Mm of Ph. D. MICI‘flGAN STA?! UNIVERSH’Y Robert Lee Beck 1963‘ IBESIS IHINIHIUHHUHUHIIHTIHIIHIHIIIHI!HIITITII I 93 10631 8680 This is to certify that the thesis entitled AN ECONOMIC ANALYSIS OF CLASS I MILK PRICING IN LOWER MICHIGAN presented by Robert Lee Beck has been accepted towards fulfillment of the requirements for _Eh..D_degree inJgniquLural Economics {it vLm/Lfl)? CIR)” K—(fiQ) L L" Major professor Date—MMQFIEL— 0-169 LIBRARY Michigan State University lV153l_l RETURNING MATERIALS: P1ace in book drop to LIBRARIES remove this checkout from .—_—- your record. FINES win he charged if book is returned after the date stamped below. 3!)! 0 8 $987 \ I‘ ,l‘ 2“- b .\ ABSTRACT AN ECONOMIC ANALYSIS OF CLASS I MILK PRICING IN LOWER MICHIGAN by Robert Lee Beck Pricing formulas are used in Federal milk order markets in establishing a minimum Class I price. The two types of formulas com- monly used are the manufacturing milk formula and the economic for- mula. A manufacturing milk formula bases the Class I price on the value of milk used for manufacturing dairy products while an economic formula bases the price on a series of economic indicators. The two formulas differ in the underlying assumptions and the rationale upon which each is based. The use of a manufacturing for- mula assumes: (1) the primary market for milk is for the manufactur- ing of dairy products and (2) resources used in the production of milk have few alternative uses. The economic formula assumes: (1) the primary market for milk is for fluid uses and (2) the resources used in milk production are faced with many alternatives, both farm and nonfarm uses. 9 A manufacturing milk formula is used as a basis for estab- lishing minimum Class I prices in the Federal order markets in Rich- igan. Trends in urbanization and industrialization and substantial declines in production of milk of manufacturing quality have led to the suggestion that the use of an economic formula, as a basis for Class I pricing, would be more applicable to existing economic con- ditions in.Iower‘Michigan. This study was designed to examine these conditions in light of the differing assumptions upon which the use of Robert Lee Beck the two types of formulas is based. First, a thorough examination of the dairy industry in Michigan was made with the view of determining the relative importance of the fluid milk sector and the manufacturing milk sector and thus determine the primary market for milk. Secondly, an analysis of both the farm and nonfarm alternatives for resources used in milk production was made to determine the degree to which milk production competes for factors of production. From the study, the following general conclusions were drawn: 1. The declining production of manufacturing quality milk, the increasing production of milk eligible for fluid consumption, and the resultant dependency of the manufacturing market on Class II milk as a source of supply leads to the conclusion that the primary market for milk in Michigan is the fluid sector. 2. Milk production competes with nonfarm industries for labor; urban development and other nonagricultural uses compete for land; and other farm enterprises compete for all factors of production -- land, labor, capital, and management. Thus, the conclusion that resources used in the production of milk have alternatives and that the dairy industry must offer comparable returns to these factors in order to hold them, or attract other resources, seems Justified. 3. The conclusions that milk is produced primarily for the fluid market and that the factors of production have alternatives agree with the assumptions upon which the use of an economic formula is based. Thus, the feasibility of the use of an economic formula as a basis for establishing Class I prices in the Lower Michigan markets was suggested. Robert Lee Beck 4. This study does not support the hypothesis that existing economic conditions in Michigan justify the continued use of a manu- facturing milk formula as a basis for establishing Class I prices. It does support an implied alternative hypothesis that changed con- ditions now suggest the use of an economic formula. Based upon these conclusions, an economic type formula was developed as a basis for establishing minimum Class I prices in Lower Michigan. The dependent variable chosen was the effective Class I price in the Southern Michigan Federal order market. To reflect. changes in general economic conditions, supply, and demand, the fol- lowing independent variables were selected: (1) index of U. S. whole- sale prices, (2) index of prices paid for manufacturing milk, (3) in- dex of prices received by Michigan farmers for all farm products ex- cept dairy products, and (4) index of the percentage of total receipts used as Class I. Each independent variable was assigned a weight by the single equation regression model: Y3 =ok+ flixfj +flzxzj+flaxsj+~f4x4j+ 113- Using annual indexes for the period 1985~6l, the following coeffi- cients were obtained: if : - .68 + °49X1 + .4ox2 + .ovxs + .04x4. With this model, 98 percent of the variation in the Class I price, on an annual basis, was explained or associated with the changes in the four independent variables selected. AN ECONOMIC ANALYSIS OF CLASS I MILK PRICING’IN'LOWERflMICHIGAN By Robert Lee Back A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Economics 1963 ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to all who made this study possible. Special recognition is due Dr. Glynn McBride, Chairman of the author's Guidance Committee, for his patience, liberality of time, and helpful suggestions throughout the development and com- pletion of this study. Sincere thanks is expressed to Drs. Dorian Feldman, Lester V} Henderscheid, and Boris P. Pesek for their suggestions and criti- cisms of earlier drafts. These added immeasurably to the final manuscript. Appreciation is expressed to Dr. L. L. Boger and the Depart- ment of Agricultural Economics for financial assistance provided during the author's graduate program. The author is indebted to the Board of Directors and the general manager of Michigan Milk Producers' Association and to Mr. George Irvine, Market Administrator for the Southern Michigan marketing area for arranging the financing of this study under a special grant to Michigan State University. Recognition is due my family for their patience, understand- ing, and encouragement. A special note of thanks must go to my wife, Betty, for the many hours spent in typing the various drafts of this thesis, including the final copy. ii TABIEOFCON‘IENTS Page Acmowmbcamur 11 LIsrcrrABms......................... vi HSTOFIILUSTRATIONS.....................viii HSTOFAPPENDIGES..............'......... ix Chapter I O MODUCTI ON 0 O O O O O O O O O O O O O 0 O O O O O O O 1 Objectives 0 O O O O O O O O O O O O O O O O O O O O l Hypoth681s O O O O O O O O O O O O O O O O O O O O O 2 Organization of Thesis . . . . . . . . . . . . . . . 2 II. ‘A REVIEW OF THE HISTORY AND THEORY OF FLUID MILK PRICING. O O O O O I O O O O O C O O I O O O O O O O O 4 Theoretical Models Used in Fluid.Milk Pricing. . . . . 4 Location of production . . . . . . . . . . . . . . . 4 Price determination in a fluid milk market . . . . . 6 The Evolution of Fluid Milk Pricing. . . . . . . . . . l4 FormulaPricing.................... 17 Types of pricing formulas. . . . . . . . . . . . . . 18 Manufacturing milk formula . . . . . . . . . . . . 18 Economic formula . . . . . . . . . . . . . . . . . 21 III. CHANGES AND MNDS IN MICHIGAN'S DAIRY INDUSTRY AND IMPLICATIONS FOR THE TYPE OF PRICING-FORMULA . . . . . 24 Fluid Milk Sector. . . . . . . . . . . . . . . . . . . 24 Manufacturing Milk Sector. . . . . . . . . . . . . . . 29 Negotiated Class 1 Prices. . . . . . . . . . . . . . . 31 Implications . . . . . . . . . . . . . . . . . . . . . 32 iii Chapter IV. RESOURCE ALTERNATIVES AND THE IHPLICATIONS TIE TYPE OF PRICING FORK-111:3; o o o o o 0 Population Growth and the Urban movement Sociological aspects . . . . . . . . . Property tax . . . . . . . . . . . . . Farm wage rates. . . . . . . . . . . Resource Alternatives in Nonagricultural Employment of the labor force. . . . . Sources of income. . . . . . . . . . . FOR Resource Alternatives in Agricultural Uses . . . . Part‘tme farming. o o o o o o o o o 0 Alternative farm enterprises . . . . . mplications O O O O O O O O O O O O O . RSCOM'IE‘IIDED CLASS I PRICING PLAN FOR LOWER Ii: CIR GAN o o o o o o o o o o e o o e e 0 Introduction . . . . . . . . . . . . . . Factors Associated with the Price of iilk. . . . . Construction of an Economic Formula. Time period. . . . . . . . . . . . . . Time unit. . . . . . . . . . . . . . . Base period for indexes. . . . . . . . iethod of weighting variables. . . . . Variables included in the economic formula . United States wholesale prices . . Percentage of Class I sales. . . . . Index of prices paid for manufacturing milk. . Index of prices received by Michigan for all farm products except dairy MOdBl o o o o o o o o o o o o o o o o 0 Interpretation . . . . . . . . . . . Application. . . . . . . . . . . . . Evaluation of the Economic Formula . . . Price certainty. . . . . . . . . . . . Seasonal price behavior. . . . . . . . Milk prices in surrounding markets . . Comparison of historical Class I price the economic formula Class I price . iv farmers products. 36 38 41 43 43 45 46 46 47 49 010101 room 03 O C) O 62 65 65 66 66 67 68 Chapter Adjustment Factors . . Supply-demand adjuster Seasonal adjustment. . Need for Re-evaluation . Class II Price . . . . VI. SUMMARY AND CONCLUSIONS. . Summary. . . . . . Conclusions. . . . . . . B IBLI OGMPM O O O O O O O O O O O APPENDIX.A. . . . . . . . . . . . APPENDIX B. . . . . . . . . . . . Page 70 7O 71 75 75 76 76 79 81 87 90 Table II. III. VII. VIII. XII. XIII. LIST OF TABLES Structural Changes in the Organizational Units of Milk Production, Michigan, 1950-1959 . . . . . . . . . Distribution of Farms Reporting Milk Cows by the Number of Cows per Farm, Michigan, 1950-59 . . . . . Changes in the Number, Size, and Product Line of Fluid Milk Plants, Michigan, 1950-1960 . . . . . . . . Change in the Number and Size of Manufacturing Milk Plants, Michigan, 1950-1960 . . . . . . . . . . . . . Pepulation of Michigan, 1920-1960 . . . . . . . . . . . Farmers"Realized Net Farm Income and Taxes Levied on Farm.Property, Michigan, 1949-1961 . . . . . . . . . . Index of Nonfarm and Farm Wage Rates, Michigan, 1950-1960. Michigan Labor Force and Employment by Major Industries, Selected Years . . . . . . . . . . . . . . Industrial Sources of Civilian Income Received by Parsons Participating in Production, Selected States, Selected Years . . . . . . . . . . . . . . . . Number of Farm Operators‘Working Off Farms, Michigan, Census Years 1929-1959 . . . . . . . . . . . Average Mbnthly Receipts of Producer.Milk During The 5-Year Periods, Southern Michigan Federal Order Market 0 I O O O O O O O I O O O O O O I 0 O O 0 Average Uniform Blend Price per Hundredweight During Two 5-Year Periods, Southern.Michigan Federal order M8 rbt O O O O O O O O O O O O O O O O O O O O 0 Economic Formula Class I Price, Southern Michigan Federal Order Market and the Minimum Class I Price, Tbledo, Northeastern Ohio, and Chicago Federal Order Markets, 1952-1961 . . . . . . . . . . . . . . . vi Page 26 27 28 SO 35 4O 42 44 45 48 91 92 93 Table Page XIV. Minimum Class I Price, Economic Formula Class I Price, and Negotiated Class I Price, Southern Michigan Federal Order Market, 1955-1961 . . . . . . . 94 IV. Data Used in Multiple Regression of Minimum Class I Price for Milk, Southern Michigan Federal Order market, 1935-196]. 0 e o o o o o o o o o o o o o o o o 95 vii LIST OF ILLUSTRATIONS" Figure Page 1. The Relation of the Price 0 Milk to Distance from Market in Different Product Zones . . . . . . . . . . . 5 2. A Hypothetical Model of a Bilateral Mbnopoly and Discriminatory Pricing by a Seller in Two markets . . . 8 3. A Comparison of the Economic Formula Class I Price for Southern Michigan with the Minimum Class I Price for Toledo, Northeastern Ohio, and Chicago Federal Order Markets, 1952-1961. . . . . . . . . . . . 69 4. Comparison of the Economic Formula Class I Price with the Minimum Class I Price and the Negotiated Class I Price, Southern Michigan Federal Order Market, 1935-1961 . . . . . . . . . . . . . . . . . . . . . . . 69 5. Percent of Total Milk Received Each Month Based on Average Monthly Receipts for Two 5-Year Periods, Southern Michigan Federal Order Market. . . . . . . . . 74 6. Monthly Average Blend Price and Producer Receipts of Milk Based on a Percent of the 5-Year Average for the Periods 1952-1956 and 1957-1961, Southern Michigan Federal Order Market. . . . . . . . . . . . . . . . . . 74 viii LIST OF APPENDICES Appendix Page A. Sources of Data for Computing 3 Formula Class I Price by the Regression Method . . . . . . . . . . . . 87 B. Tables 0 O O 0 O O O O O 0 O O O I O O O O O I O O O O O 90 ix CHAPTER I INTRODUCTION The type of formula used for establishing Class I milk prices should be a function of the economic conditions which exist within the dairy industry and the economic environment of the society within which the industry Operates. As these conditions change, it is necessary to adjust the pricing formula accordingly. Basically, there are two types of formulas commonly used in pricing fluid milk. The manufacturing milk formula bases the Class I price on the value of milk used for manufacturing purposes. The economic formula uses indexes of economic factors as a basis for establishing the Class I price. The two formulas differ sig- nificantly with reference to the assumptions regarding the primary market for milk and the alternative uses for factors of production. A manufacturing milk formula is used as a basis for estab- lishing Class I prices in the Federal order markets of Lower Michigan. However, recent discussion among dairy leaders in the state regarding formula types has led to the suggestion that economic indexes as a basis for Class I pricing might be more applicable to exiSting eco- nomic conditions. For that reason, the author has attempted an economic analysis of Class I milk pricing in_lower'Michigan. Objectives There are three objectives of this study. The first is to l analyze the changes and trends in Michigan's dairy industry in light of the underlying assumptions of the manufacturing type formula and those of the economic type formula concerning the primary market for milk. A second objective is to examine the alternatives for re- sources used in producing milk. This analysis is directed at an examination of the alternatives as they relate to the differing assumptions of the two types of formulas. The third objective is to suggest a Class I pricing fermula for Lower Michngan which will reflect the findings of this study. Hypothesis The basic hypothesis of this study can be stated as follows: The existing structure of the dairy industry in Michigan and the economic environment within which the industry functions justify the continued use of a manufacturing.milk formula as a basis for establishing Class I price. The basis for rejecting or not rejecting the hypothesis will be a thorough analysis of the existing conditions as they relate to the economic rationale and the assumptions of the different types of formulas. Qgganization of Thesis The procedure followed in this study is that of first re- viewing the history of fluid.milk pricing and presenting theoretical models which might be helpful in explaining price determination in the dairy industry. In addition, the assumptions, components, and rationale of the two types of pricing formulas will be discussed in detail. Chapter III is devoted to an examination of the changes and trends in.Michigan's dairy industry and the implications for the type of pricing formula. The analysis attempts to determine the relative importance of the fluid and manufacturing milk sectors in determining which sector represents the primary market for milk. Chapter IV'is an examination of the alternatives for resources used in milk production. The basis for Chapter IV is the differing assumptions of the formulas concerning whether there are alternative uses for the resources used in milk production. A recommended Class I pricing plan is presented in Chapter‘v. The plan uses the type of pricing formula which seemed applicable based upon the re- sults found in Chapters III and IV. The summary and conclusions are set forth in Chapter VI. CHAPTER II A REVIEW OF THE HISTORY AND THEORY or FLUID MILK PRICING]- Theoretical Models Used in Fluid.Milk Pricing In an analysis of fluid milk pricing the use of theoretical models facilitates an understanding of price determination. This section is devoted to: (l) a discussion of the location of milk production and (2) price determination in a market in which a pro- ducer bargaining association controls a large portion of the sup- ply of fluid milk and in which Federal order provisions provide for price discrimination through a system of classified pricing. location of Production The location of agricultural production, as set forth by Vbn Thunen in the early part of the 19th century is particularly relevant to the location of milk production and to the form in which milk is marketed. VOn Thunen's isolated state, in explaining the location of agricultural production with respect to the consuming IThroughout this study, numerous references will be made to fluid milk, manufacturing milk, Class I, and Class II milk. Fluid milk is defined as that milk which is produced under sanitary stand- ards which make it eligible for fluid consumption. manufacturing milk refers to milk produced under less rigorous sanitary standards and, as a result, is only eligible for use in manufacturing dairy products. Class I and Class II designates the use classifications of fluid milk. Milk and cream consumed in fluid form make up Class I. Class II mdlk is the amount above Class I use and is used in manu- facturing dairy products. center, states that the production of goods will be determined by their value in relation to the costs of transportation and by their form. As one moves away from the consuming center, land will be used less intensively and devoted increasingly to goods which are relatively less perishable and which are valuable enough to bear the cost of transportation. When this principle is applied to the location of milk production and the form in which milk is marketed, the zones from which the different products are shipped will tend to take the form of concentric circles around the consuming center. The boundary lines between adjacent product zones are defined by the formula P1 - T1R = P2 - T2R. P1 and P2 equals the price of one hundred pounds equivalents of milk made into products 1 and 2, re- spectively, while T1 and T2 are the transportation rates fior products 1 and 2. Solving the equation for'R establishes the boundary between the two zones. Based on the Vbn Thunen type analysis, the relation of price to distance from the market in establishing product zones may be represented graphically as in Figure 1. Price \\ 1000 " 56 _ 1 o 150 a 260 Distance Figure l.--The relation of the price of milk to distance from market in different product zones. 6 The product zones change when changes occur in transportation rates or in the relative prices of the different products. Price Determination in a Fluid Milk Market The forces of supply and demand generate market prices within the limits of an organizational framework. At one limit of the range of possible organizational patterns is the purely competitive market. At the other limit is the monopoly. Between these two ex- tremes there exists an array of market structures. As the market structure approaches that approximating monopoly conditions, the market price becomes a function of the relative bargaining power of the buyer and seller. Recent developments in the organization of supply in some fluid milksheds have brought about a market composed of one major producer bargaining association which sells a high percentage of the milk furnished for that market to a few buyers. This has introduced a market structure similar to that of a bilateral monopoly and is often referred to as monopolistic oligOpsony. Because of the similar- ity, in terms of the market structure, a bilateral monopoly model is often used to cast light on price determination in the fluid milk in- dustry. It should be pointed out that while it is possible to fix definite limits within which price may be agreed upon, the exact price level within these limits is indeterminate, from.a strictly economic point of view. The actual price will depend upon the relative bar- gaining strength of the buyer and seller. The form of bargaining which prevails in some fluid milk markets is that in which the cooperative and the distributor negotiate the price of milk for some future period during which the supply may 7 change. There is no attempt to fix the total quantity of milk to be supplied nor the total amount of money to be paid. The limits of the bargaining range for a Class I price are shown in Figure 2. Given the assumption of profit maximization, the seller would seek to establish price Oj, at which he would furnish 0A quantity of milk. At this point, the seller would maximize pro- fits. The buyer would be equally desirous of establishing a price 0d. At this price the buyer could maximize profits by purchasing 0G quantity of milk. The conflict of objectives involves the buyer's efforts to force the price down to 0d, where the seller will produce not less than quantity 0G, while the seller is striving to force the price up to Oj, where the buyer will purchase quantity 0A. Since the price becomes, within comparatively broad limits, a question of relative bargaining power, the outcome will be dependent upon factors affecting the bargaining position of each party. The introduction of a third party, namely, a Federal milk marketing order, into the analysis changes both the relative bar- gaining positions and the bargaining range. The basis for referring to a Federal marketing order as a third party in the bargaining process is based on the philosophy and objectives of Federal regulation of markets as set forth in the Agricultural marketing Agreement Act of 1957. These objectives include the following: Whenever the Secretary finds, upon the basis of the evidence adduced at the hearing required by section 608b of this title or this section, as the case may be, that the parity prices of such commodities are not reasonable in view of the price of feeds, the available supplies of feeds, and other economic conditions which affect market supply and demand for.milk and its products in the marketing area to which the contemplated agreement, order, or amendment relates, he shall fix such prices Price MC1(buyer) MC (seller) J'mn‘Du \ i \ ‘s\ : AC(seller) “I \ I \ I \\\ I l : \ l \\ I | : ‘MRI(buyer) ' | : DII(aeller) ' I I : MR(seller) : I l l I H G 3 Quantity Figure 2.--A hypothetical model of a bilateral monopoly and discriminatory pricing by a seller in two markets. 9 as he finds will reflect such factors, insure a sufficient quantity of pure and wholesome milk, and be in the public interest.1 The statement "be in the public interest" leaves the con- notation that society is being represented. This point is further exemplified by the philosophy associated with market performance and the role played by society in influencing market performance.2 As society develops, policies and programs evolve which have an objective of influencing the marketing environment. The type of programs and the extent of their use is an expression of the set of values and beliefs held by society at a given time. These values and beliefs are brought to bear on the marketing system through the political system. If we assume that society has accepted some notion of market performance which involves the welfare of the market par- ticipants as well as society then it follows that this notion is ex- pressed in the formulation and implementation of policies which it believes will bring about such performance. - It is in this context that Federal milk marketing orders may be viewed as a third party in the bargaining process. Through the political system, society has expressed its notion of the market per- formance expected from the dairy industry and the procedure it believes will bring about such performance. ‘IUnited States Department of Agriculture, Compilation of Statutes Relating_to Research, Statistics, and Reports Service and Re ulatory work, and Food Distribution, Agriculture Handbook no. 201 lashington: U. 8. Government Printing Office, 1961), p. 40. 3Market performance refers to the economic results that flow from the industry as an aggregate of firms. Society is concerned with how an industry performs in terms of its efficiency, its pro- gressiveness, and its stability. For a discussion of the principal dimensions of market performance see Joe S. Bain, Industrial Organiza- tion (new York: Jehn‘liley and Sons, Inc., 1959), p. 12. 10 If an additional assumption is specified, generalizations may be made relative to the bargaining range when a Federal order mini- mum price is introduced into the bargaining process.1 If it is as- sumed that the cost structures of both the buyer and seller remain unchanged after a Federal marketing order becomes effective, the establishing of a minimum Class I price narrows the bargaining range except in that case in which the Federal order minimum price is 0d or below, or price Oj or above. In such a case, the bargaining range re- mains unchanged and the Federal order minimum price becomes irrelevant in the bargaining process. If the Federal order minimum price is be- low Od, the buyer would bid up the price to 0d to maximize profits. Likewise, the seller would bid down any price above Oj. Thus the bargaining range is narrowed by the amount which the Federal order minimum price exceeds price 0d. Profits are influenced by the level of price. Within limits, the higher the price established for milk, the greater the seller's aggregate profits; the lower the price established, the greater the buyer's profits. These limits are established by the prices and quantities which maximize the total profits of the buyer and seller, respectively. Profits are definable at these two points. If the seller dominates the bargaining process and can dictate the terms of exchange, he would choose a price of Oj and quantity 0A. At this point the seller's profit would be fDKj. If the buyer dominates and is able to establish his desires, he will pay price 0d for quantity OG. AlThis modification of the method of establishing a price via collective bargaining may be referred to as authoritative-negotiated pricing. With the Federal order minimum price as a lower limit to the bargaining range, a Class I premium above the minimum price is negotiated. 11 In this case, the seller's profit is gRTd. In the long—run, however, INC 3 AC and this profit disappears. The buyer gains a profit dTZi plus any profits he may gain from selling to the final consumer. In the same manner, profits can be shown for any given price and quantity. The distributor is not only a monopsonist with respect to the seller but is a monopolist in selling to the consumer. Thus, he will try to maximize total profits by negotiating a price such that the total of monOpsony profits which he is able to extract from the seller plus the monopoly profits which he is able to extract from the con- sumer is greatest. If we assume that monopoly price is defined as that price above which the seller is willing to take and likewise, monopsony price as that price below which the buyer is willing to pay, for a given quantity, then, dTZi becomes monopsony profit to the buyer. If, on the other hand, we assume that monopoly profit is defined as profits over and above the implicit cost elements, pro- vided that average revenue is above average cost, and that monopsony profits accrue to a firm because it is able to pay factor-owners less than the competitive factor-income price,1 the buyer's monopsony profit is not the same as above but, rather, is dependent upon the level of the competitive price. The introduction of a Federal minimum order price influences, to some extent, the distribution of profits between the buyer and seller. The level of the minimum price as well as the definition of monopoly and monopsony price is crucial to the discussion of the distribution of profits. Again, if we define monopoly as that price TPIJohn F. Due, Intermediate Economic Analysis (Homewood, Illinois: Richard D. Irwin, Inc., 1956), pp. 440-441. 12 above which the seller is willing to take and monopsony price as that price below which the buyer is willing to pay, a Federal order minimum price above 0h limits the handler to the monopoly profits which he is able to extract from the consumer. If we assume monopoly and monopsony price to be associated with implicit costs and competitive factor prices, respectively, the handler may still extract monopsony profits from the seller provided the Federal order minimum price is below the competitive market price for factors. Again, monoPsony profit is dependent upon the level of the competitive market price for the given quantity. When the Class I price has been negotiated, given the supply and demand functions for each market, the quantity to be sold in the Class I and Class II markets is determined. The quantity and price for each market are shown in Figure 2. If, for example, the negotiated price is 0e, quantity CH will be sold in the Class I market and quantity EB will be disposed of in the Class II market at a lower price, 0a. This method of selling milk in two separate markets is referred to as classified pricing of milk. This method was developed by producer bargaining associations in the sale of milk to handlers in city markets and represents the application of a practice known as price discrimination. Price discrimination is the term applied to any practice whereby a seller sells a homogeneous commodity at the same time to different buyers at different prices. There are two conditions which are essential to the success— ful application of discriminatory pricing to milk pricing. These are: (1) different elasticities of demand for the commodity in the price categories established and (2) the markets must be kept separated. 13 The seller practicing price discrimination determines the distri- bution of sales among the different markets. For any given volume of sales, the seller will distribute them among the markets in such a manner that the marginal revenue in all markets is equal. Since more milk than the buyer is willing to take at the agreed Class I price is produced, the Class II milk must be disposed of at a lower price. In this case, the cooperative will profit by divert- ing part of the supply to the Class II market in order'to protect the Class I price. The essential condition that makes discriminatory pricing a profitable practice is the existence of demand functions with dif- ferent slopes in the fluid and manufacturing markets. The demand for Class I milk is normally quite inelastic within the range of reasonably anticipated price changes while the demand for Class II milk is relatively elastic in most fluid markets. Demand functions with different slopes may be associated with geographic areas, or with in- come, occupation, and ethnic groupings within the same area. The differing elasticities of demand for.milk and manufactured dairy products, however, are usually attributed to the fact that the market for manufactured dairy products is nationwide while the Class I market is limited by the bulkiness and perishability of fluid milk and by local health regulations. The producer association is limited, to some extent, in practicing price discrimination.l If the seller has some degree of 1For a discussion of certain features which differentiate classified pricing of milk from the more usual application of price discrimination by sellers, see: Edmond S. Harris, Classified Pricing of Milk, Some Theoretical Aspects, USDA Technical Bulletin 1184 (Washington: U. S. Government Printing Office, 1958), pp. 37-39. l4 monOpoly control over prices and distribution he will distribute sales such that marginal revenue in all the markets is equal. Total revenue is greatest when the quantity is divided between markets in this manner. However, the producer association is un- able to do this except as is implicit in the negotiated price. If the bargaining association is strong enough to dominate the bar- gaining, prices may be set which would equate the marginal revenue in the Class I and Class II markets. If the producer association occupies a relatively weak bargaining position, it is conceivable that the price might be one which would not equate the marginal revenue in both markets. Since the seller faces a relatively elastic demand curve for Class II milk, this market is used primarily to salvage as much revenue as possible from the Class II milk. This section may be concluded by stating that these models can prove useful in shedding light on price determination in the dairy in- dustry by providing a conceptual framework for an evaluation of Class I milk pricing. The Evolution of Fluid Milk Pricing The methods of determining prices to be paid producers for fluid milk have passed through at least four phases during the past century. The first phase was characterized by numerous unorganized producers and many buyers, none of which could affect prices by their actions. Under these conditions, the price of fluid milk was a function of the impersonal movements of supply and demand. The intersection of the supply schedule and the demand schedule indi- cated the equilibrium. Price changes were brought about by the inter- action of adjustments in the demand for and the supply of milk. 15 The second phase in the evolution of price determination oc- curred simultaneously with the development of large scale Operations in the fluid.milk industry. The trend toward a few large distributors controlling a major portion of the fluid market was followed by the formation of producer associations fer the purpose of bargaining collectively with the handlers. This type of price determination under conditions of imperfect competition is referred to as ad- ministered pricing. This introduced a market situation closely related to a bilateral monopoly and described by Nicholle as monop- olistic oligopsony.l The producer bargaining associations became more active in their role as bargaining agents for the producer. To overcome the problems associated with a flat-rate price far all milk, the associations developed a classified pricing system wherein milk was priced according to the form in which it was used by the dis- tributor. By practicing price discrimination in the Class I and Class II markets, the seller was able to increase total returns. Producers shared equally the lower prices received in the Class II market as well as the higher prices received in the Class I market. Distributors also received equal treatment in that each was given the same competitive advantage regardless of varying class utilization. The third phase of this evolutionary process is called authoritative pricing and consists of price determination by public action through government regulation. In general, this has been limited to the determination and legal enforcement of minhmum prices by both Federal and state agencies. In the early thirties, milk lwiiiian H. Nicholle, Imperfect Competition within Aggi- cultural Industries (Ames: Iowa State College Press, 1941), p. 14. 16 prices declined rapidly and to very low levels in relation to other agricultural products. Producer associations, which for a short period had prevented a decline in price, were unable to cope with the problem. jMeanwhile the Federal government was taking steps to prevent further declines and even restore the pre-depression price levels for farm commodities. A series of Acts, starting with the Agricultural Adjustment Act of 1933, was passed by Congress. The Acts were specifically designed to increase prices of agricultural products. The Agricultural Adjustment Act authorized the Secretary of Agriculture to enter into marketing agreements between producer associations and distributors of milk. Marketing agreements estab- lished minimum producer prices but they were unsuccessful in estab- lishing minimum resale prices for milk. In 1934, the Department of Agriculture terminated all former agreements on minimum resale price and directed its activities toward the determination and maintenance of producer prices only. The terminated marketing agree- ments were replaced by licenses which reflected the new policy of administering only producer prices. Following a series of amendments and constitutionality cases involving the Agricultural Adjustment Act, the order-agreement provisions were included in the Agricultural Market Agreement Act of 1937. The Act, as amended, has successfully withstood some of the tests of constitutionality wherein the order- agreement provisions of the Agricultural Adjustment Act failed. It is on the basis of the provisions of the Agricultural Market Agreement Act of 1937 that Federal milk marketing orders Operate. Such orders establish minimum class prices. Authorita- tive pricing of fluid milk came into use during a period in which all 17 agriculture was in a depressed condition and needed help. It has re- mained as an aid in the marketing of a perishable product. A fourth phase may be referred to as authoritative-negotiated pricing. Through collective bargaining, a Class I price is estab- lished above the Federal order minimum price. This method of pric- ing is used in several Federal order markets, including those in Michigan. The Federal order minimum price becomes a floor from which to negotiate. The level of the negotiated price is determined by the relative bargaining position of the producer associations and the handlers. The effect of the Federal order minimum price on the bargaining range and on the relative bargaining positions of the buyer and seller was discussed in the previous section. Formula Pricing Formula pricing of milk dates back to World War I. At that time milk prices in most major markets were under government regu- lation. The formula technique was an attempt to introduce auto- matic elements into an administered fluid milk pricing system. As classified pricing developed, formulas were used for pricing Class II milk in markets where the classified pricing plan was in operation. The price of Class II milk was based upon the value of milk used in manufacturing dairy products. Aside from that, there was very little emphasis on formula pricing until the mid 1930's. A formula system for pricing Class I milk was adopted in the Chicago market in 1935, almost four years prior to the estab- lishment of a Federal order for that market. Class I premiums above the minimum Federal evaporated milk code price were estab- lished. The evaporated milk code price was determined by a formula 18 based on the prices of butter and cheese. Prior to the passage of the Agricultural Marketing Agreement Act of 1937, Class I prices in markets regulated by the Federal government were fixed prices. They were established or changed by a public hearing. However, the Act established a more rigorous procedure for issuing and amending orders which greatly increased the time required to make changes in Class I prices. Because of the time required to establish prices under these conditions, formula pricing was adopted in many markets as a means of making Class I prices more responsive to changes in economic conditions. This trend was interrupted by World War II conditions which neces- sitated the use of price controls. very few changes in the pricing structure in fluid milk markets occurred during this period. Hew- ever, the anticipated rise in price with the removal of price ceil- ings gave further impetus to the adoption of formula pricing in Federal order markets. Fermula pricing is now the accepted method of establishing Class I price in all Federal order markets. Types of Pricing Formulas i There are basically two types of formulas used in pricing Class I mdlk. The type most commonly used is the manufacturing milk formula which bases the Class I price upon the value of milk used for manufacturing dairy products. The second type, usually referred to as an economic formula, bases the price of Class I milk on a series of economic indicators. Manufacturing milk formula An analysis of a fluid milk market will suggest many factors 19 which may influence Class I price. It will also indicate certain relationships or correlation between the historical Class I price in the market and other prices. The manufacturing milk formula does not attempt to isolate and incorporate the relationship between Class I price and all the factors influencing the price of fluid milk. Instead, the formula is based upon the assumption of a close and reasonably constant relationship between fluid milk prices and the price of milk used in manufacturing dairy products. The for- mula attempts to maintain what appears to be a normal relationship between the Class I price and the price of one or more dairy products in the market. This is based on an analysis Of past relationships and a knowledge of the difference in costs of producing milk for different uses. Prior to the development of health regulations, milk was considered a homogeneous product and could be used for fluid con- sumption or manufacturing purposes without restriction. Under those conditions there was a direct causal relationship between the price of manufactured dairy products and fluid milk. Health regulations did not destroy this relationship but the relationship became less automatic because of the increased costs associated with the produc- tion of milk for fluid consumption. The assumption of a close and reasonably constant relationship between fluid milk prices and prices received for manufactured dairy products may be questioned on the basis of differing effects on fluid milk and variOus manufactured dairy products of the same economic forces. An example is the effect of changes in income on the con- sumption of fluid milk and manufactured dairy products. Estimates of income elasticities for fluid milk and cream and manufactured dairy 20 products for the post World War II period of 1947-54 are given by Rojko as follows:1 Fluid milk and cream 0.27 Butter 0.36 American Cheese -0.99 Other dairy products 3.06 This is only one example of the differing effects of a demand shifter on the various milk products. Perhaps other shifters of supply or demand act similarly. Because of the assumption of a close correlation between fluid milk prices and the price of manufactured milk products in the past, the manufacturing milk formula has been extensively em- ployed in markets where the causal relationships were considered evident. These markets are located primarily in the middlewest and the far western section of the United States. ,Manufacturing milk formulas vary in composition. Hewever, the raison d'etre of pricing Class I milk on the basis of the value of milk used for manufacturing purposes remains unchanged. To accomplish this objective, some manufacturing milk formulas use the prices paid for whole milk used for manufacturing dairy products. The midwest condensery price has long been used as one of the principal measures of the value of manufacturing milk. HOwever, because of the declining number Of plants used in this price series, the Minnesota-Wisconsin price series has been developed as an alternative measure. The Minnesota-Wisconsin price series provides a much broader base since it represents the prices paid for more than 50 percent of the lAnthony S. Rojko, The Demand and Price Structure for Daigy Products, USDA Technical Bulletin 1168 (washington: U. S. Government Printing Office, 1957), p. 105. 21 manufacturing milk produced in the United States. Other manufacturing milk formulas use the market price of various dairy products such as butter, cheese, and non-fat dry milk powder to derive a measure of the value of manufacturing milk. This value is than used as a basis for determining Class I price. Re- gardless of the technique used to derive a measure of the value of manufacturing milk, the basis for the Class I milk price remains unchanged. The assumptions and the philosophy associated with the types of formulas are crucial to this analysis. The reasoning used for basing the Class I price on the value of milk used for manufacturing purposes is the assumption that the manufacture of dairy products represents the primary market for milk. It is further assumed that the factors of production have limited alternative uses other than in the production of milk. The latter assumption implies that factor alternatives differ among markets and that formulas differ with respect to the extent to which they reflect opportunity costs in the Class I price of milk. The use of a manufacturing milk formula or an economic formula in any given market assumes that knowledge regarding the alternative uses for resources used in milk production in that market have been taken into account in considering the economic feasibility of the formula. Economic formula Failure of the manufacturing milk formula to make prompt price adjustments for major changes in economic conditions prompted a search for a new type of Class I pricing formula following WOrld 22 War II. In an effort to devise a formula which would be acceptable and at the same time make these adjustments promptly, the Boston Milkshed Price Committee was appointed to study the pricing problem in the Boston market. The committee's major objective was that of determining an alternative method of establishing future Class I prices in the Boston market. The resulting economic formula recom- mended by the committee differed from the manufacturing milk for- mula in that it abandoned the attempt to correlate the Class I price of milk to the value of manufacturing milk. Instead, Class I price was associated with what was considered to be several of the major factors with which movements of Class I prices were associated. These factors then became the price movers in this new type Class I pricing formula. Price movers for economic formulas may vary because of the individual market conditions but basically they fall into the follow- ing interrelated categories: (1) factors which indicate a change in the general price level, (2) factors which reflect a change in supply, and (3) those which reflect changes in demand. Various combinations and modifications of price movers may be used, depending upon the market. The primary function of a factor which measures the general price level is to keep the Class I price of milk in a reasonable relationship with the general level of economic conditions. A secondary function is to stabilize short-run price movements which may be associated with irregular or episodic time elements. Fluctu- ations in the general price level are usually less violent than those brought about by other shifters of the supply and demand function for milk. 23 Factors which reflect changes in the local supply of and the demand for milk are usually confined to the local market. To a large extent, fluid milk markets are still locally oriented and thus price should reflect the changes in supply and demand which occur in each market. The assumptions and the economic rationale upon which the economic formula are based differ from those of the manufacturing milk formula. The use of an economic formula assumes that the pri- mary market for milk is the fluid market. The manufacturing market is considered the residual market for the Class II milk from the fluid market. The use of an economic formula also assumes that the resources used in fluid milk production are not faced with limited alternative uses, as is assumed by the use of a manufacturing milk formula, but that there are alternative uses such as other farm enter- prises and nonfarm uses. If resources used in producing fluid milk have alternative uses each factor must, in the long run, earn a re- turn which will prevent its shifting to other uses. In order to take alternative opportunities for resource use into account, a formula must include variables which will result in milk prices which reflect the earning power of the resources in competing alter- natives. CHAPTIR III CHANG'SS AND 'I'RDNDS IN I-iICHIGAN 'S DAIRY INDUSTRY AND IMPLICATIONS FOR THE TYPE OF PRICING FORMULA The dairy industry in Michigan has experienced tremendous changes during the past decade. The trend toward specialization in milk production has left this segment of the induStry in a position to be greatly affected by decisions made outside agri- culture. An example is the dependence on credit for greater in- vestments, the rates of which are determined in the money market. With the trend toward specialization, there has been a trend toward increased production of milk for the fluid market. The total re— ceipts of fluid milk in the Southern Michigan Federal order market increased 43 percent during the period 1952-1959. At the same time, total sales of whole milk increased 19.5 percent. This indicates that an increasing portion of manufactured dairy products are being made from milk of fluid quality. The changes and trends in the dairy industry have implications for the type of Class I pricing formula. One objective of this study was to examine these changes as they relate to Class I pricing. Fluid Milk Sector Production of milk for the fluid market has increased although the number of producers has decreased. This may be attributed to 24 increased production per cow and increased herd size. The average number of producers delivering milk to regulated handlers in the Southern Michigan order market decreased 1.5 percent per year for the period 1952-59. The average daily delivery of milk per pro- ducer in this market was 60.5 percent greater in 1959 than in 195.2.1 A comparison of these changes with the structural changes which occurred in the production units of all milk producers pre- sents a more meaningful view of the relative importance of the fluid milk segment. As shown in Table I, the number of dairy farms de- creased 46 percent while the number of cows oer dairy farm increased 72.7 percent. Table II indicates a decline in the number of farms reporting less than 20 milk cows while the number with 20 or more cows increased during the period 1950 to 1959. This is significant in view of the fact that most manufacturing milk producers fall into the less-than- 20-cows-per—farm category. Also, the rate of decrease in the number of fluid milk producers has been less than the rate of decrease for manufacturing milk producers. In.Michigan, 96 percent of the milk sold by farmers in 1961 was in the form of whole milk. This represented a 15 percent increase over the amount sold as whole milk in 1947. A large portion of this milk is produced under sanitary conditions which make it eligible for the fluid market. In 1961, forty-one percent of the total milk pro- duced was used for manufactured dairy products. Since this included Class II milk from fluid markets, a much higher percentage than the 59 percent indicated was Of fluid milk quality. IThis type of comparison is valid despite the fact that the marketing areas for the two years are different. 26 TABLE I.-Structural changes in the organizational units of milk production, Michigan, 1950-19591 Percent Item 1950 1959 change Farms reporting milk cows 105,990 50,587 52.3 Average number of cows per farm reporting milk cows 7 12 71.4 Number of dairy farms2 45,729 24,673 46.0 Number of cows per dairy farm ll 19 72.7 Tbtal number of milk cows on farms 794,341 628,044 21.4 Average production per cow (pounds) 6,200 7,830 26.2 Farms reporting sales of milk and cream 91,426 43,110 52.8 1United States Bureau of the Census, U. 8. Census of Agri- culture: 1950, V61. 1, Pt. 6 (Washington: Office, 1952). U. S. Government Printing United States Bureau of the Census, U. 8. Census of Agri— culture: 1959, V01. 1, Pt. 6 (washington: Office, 1962). 2A8 defined by the census of agriculture. U. S. Government Printing 27 inane II.-Distribution of farms reporting milk cows by the number of cows per farm, Michigan, 1950-59l Less than 5-9 10-19 20-49 More than 5 cows cows cows cows 50 cows (number of farms) 1950 40,600 34,787 25,945 4,856 136 1959 15,996 10,912 15,957 10,244 685 10. R. HOglund, Michigan Dairy Farming, Agricultural Experi- ,ment Station Special Bulletin 440 (Fast Lansing: Michigan.State University, 1962), p. 8. Major developments and trends in the processing of fluid milk are evident from the data shown in Table III. The most noticeable changes were: (1) a decline in the number of fluid milk processing plants, (2) an increase in the size of plant, and (5) shifts in the product line. The number of processing plants decreased 52 percent during the period 1950-1960. The increase in average plant size of fluid milk plants, in terms of pounds of milk received, was even more noticeable. This average increased from 4.9 million pounds of milk in 1950 to 12.2 million pounds in 1960, or an increase of 149 percent. Other data in the same table indicate changes which occurred in the product lines of fluid milk plants. The fluid milk segment Of Michigan's dairy industry, even though changed in structure, has become an increasingly important part of the industry. The trend in production is toward fewer and larger producers. Fluid milk sales have increased, even though per capita consumption has decreased. Fewer, larger, and more 28 TABLE III.-Changes in the number, size, and product line of fluid milk plants, Michigan, 1950-19601 Percent Item 1950 1960 change Number of plants 562 269 52 Amount of milk received (million pounds) 2,770.4 5,275.2 18 Percent used as Class 12 68.0 64.0 n.a.3 Average size of plant (million pounds) 4.9 12.2 149 Whole milk sold for direct consumption (million pounds) 1,719.4 2,218.4 129 Number of plants processing various milk products Fluid cream for direct consumption 512 255 50 Butter 53 19 64 Creamed cottage cheese 138 67 51 Ice cream 143 83 42 Ice cream mix 90 76 16 Percent of total milk products processed by fluid milk plants Fluid cream for direct consumption 100 100 n.a. Butter ll 11 " Creamed cottage cheese 45 51 " Ice cream 25 35 " Ice cream mix 30 45 " lGlynnllchride, Structural Changes in Michigan's Dairy Indus- try and Their Implications, Department of Agricultural Economics Preliminary Report (East Lansing: Michigan State University, 1962). 2Southern'Michigan Federal order market area. 5N0t applicable. 29 specialized plants are now handling the increased production and sales. These changes have implications which are relevant in con- sidering formula types which might be used in pricing Class I milk. manufacturing Milk Sector Changes and trends similar to those observed for the fluid sector are also evident in the manufacturing milk segment, but on a somewhat larger scale. One major difference is the fact that the production of milk for manufacturing purposes is decreasing at an extremely fast rate as compared to the trend in fluid milk produc- tion. A recent study of changes which occurred in the manufacturing milk sector between 1955 and 1957 indicated a 27.6 percent decrease in receipts of milk (milk equivalents) for manufacturing purposes for the plants in the sample.1 A more recent study indicated that the trends in the 1955-57 study were continuing in 1960.2 The decreasing relative importance of manufacturing milk is apparently due to the tendency to shift to fluid milk production and a failure on the part of remaining producers to increase the herd size. . Data are not available to determine the portion of the total milk production which is produced for manufacturing uses. Some in- sight is provided by a comparison of the data in Tables III and IV. During the period 1950-1960, the amount of milk marketed as whole milk in the state increased 18 percent. During the same period, the 1Glynn McBride and Willard H. Blanchard, Changes in Michigan's Manufacturing Milk Industry, Agricultural Experiment Station Special Bulletin 427 (East Lansing: Michigan State University, 1959), p. 19. 2Glynn McBride and William B. Hellegas, "Fewer Producers of Manufacturing Milk,".Michigan Farm Economics, No. 216, (January, 1961), pp. 1-2. 30 amount of milk received by fluid milk plants increased 18 percent while the amount received by manufacturing plants declined 7 percent. The composition of the change in the amount received by manufacturing plants is more significant than the actual change. The manufactur- ing milk sector is the market for a considerable portion of the Class II milk of the fluid sector. The increase in fluid milk re- ceipts and a relatively stable Class I use means that a higher portion of the milk received by manufacturing milk plants was pro- duced under conditions which made it eligible for the fluid market. When the Class II milk used for manufacturing purposes is accounted fer, the amount of milk marketed as manufacturing quality actually declined more than the 7 percent indicated. TABLE IV.-Change in the number and size of manufacturing milk plants, Michigan, 1950-1960l Percent Item 1950 1960 change Number of plants 201 153 24 Amount of milk received (million pounds) 3,090.6 2,885.6 7 Average size of plant (thousand pounds) 15,376 18,860 23 lGlynnMcBride, Structural Changes in Michigan's Daigyglndus- try and Their Implications, Department of Agricultural Economics Preliminary Report (East Lansing: Michigan State University, 1962). In conjunction with the decline in production of manufactur- ing quality milk, manufacturing plants decreased in number but in- creased in size. The number of manufacturing plants declined 24 31 percent during the period 1950-1960. At the same time the average capacity per plant increased 23 percent. Thus, the trend is toward a continuing decline in importance of the manufacturing milk sector of Michigan's dairy industry. Be- cause of an apparent shift of resources to fluid milk production and to nonfarm alternatives, the sector devoted to the production of milk for manufacturing purposes is declining, not only in absolute terms, but also relative to the fluid milk sector. There is also a trend toward the use of higher quality milk for manufacturing purposes. Class II milk from the fluid markets continues to increase in importance as a source of supply for manu- facturing plants. The increased use of this milk for manufacturing purposes indicates the changing relative importance of the two sectors Of the industry. If the present trends continue in both sectors, it is con- ceivable that the manufacturing milk sector may become entirely de- pendent on Class II milk from the fluid markets as a source of supply. This point is fundamental to the type of Class I pricing formula used since the formulas differ in the assumption concerning the primary market for milk. NegOtiated Class I Prices Federal milk order regulations establish minimum class prices. Producers may bargain for higher prices if they wish. Negotiated Class I premiums have been paid to fluid milk producers in Michigan since 1956. Through 1961, negotiated premiums increased the Class I price in the Southern Michigan Federal order market an average of 59 cents per hundredweight. This means that the effective Class I price 32 has been a negotiated price rather than the minimum price established by the Federal orders. Negotiated premiums are often pointed to as evidence of the shortcomings of the present formula for establishing Class I prices in Michigan. While there are other factors involved in negotiating Class I premiums, it is still an indication of the need for an evaluation of the formula in use. The fact that the effective Class I price has been a negotiated price rather than the formula price is one basis for questioning the appropriateness of the manufacturing milk formula. It is on this basis that negotiated premiums are here regarded. Implications This chapter has been devoted to an analysis of the changes and trends in.Michigan's dairy industry during the past decade. The objective was to relate these trends to the type of pricing formula which might be considered as being applicable to existing market con- ditions. Changes have occurred in the relative positions of the fluid and the manufacturing milk sectors of the industry indicating the declining importance of the manufacturing sector and the increasing emphasis on production of milk for the fluid market. A second change has been the use of negotiated prices rather than order prices for ‘Class I milk. Since the type of formula used should reflect existing eco- nomic circumstances and since it should be reasonably consistent with the rationale underlying its use, these findings appear significant in considering changes in the type of formula now being used. CHAPTER IV RESOURCE ALTERNATIVES AND THE IMPLICATIONS FOR THE TYPE OF PRICING FORMULA In a free enterprise economy, resources are allocated among different uses in such way as to increase the efficiency of the economy.1 Resource prices furnish the mechanism for reallocation of resources. ,A resource will shift when its VMP (value of marginal product) in one use exceeds its VMP in another use. The transfer will continue until its VMP is equalized in all its uses. There are a number of forces which prevent the realloca- tion of resources. Even though the price system is free to operate as an allocator, there are other impediments to the movement of re- sources. Lack of knowledge on the part of resource owners may pre- vent their movement to alternative uses. This is particularly true in the case of labor where sociological and psychological factors may impede movements. Ties to particular communities, friends, or family may restrict mobility. Virtues of a particular occupation, community, or way of life may restrict mobility. Spatial location is extremely important when considering land resource alternatives. The location of agricultural production, as set forth by VOn Thunen in the early part of the 19th century seems ‘lFor a detailed discussion of resource allocation refer to R. H. Leftwich, The Price System and Resource Allocation (Revised edition; New York: Holt, Rinehart and Winston, 1960), Chapter xv, pp. 320-337. 33 34 particularly relevant in relating the location of fluid milk pro- duction to the alternative uses for resources used in milk production. Von Thunen's isolated state model of explaining the location of agricultural production may be applied quite meaningly to the dairy industry in.Michigan in explaining the location of milk pro- duction, location of plants, and the form in which milk is marketed.1 This in turn, should be enlightening in the following analysis of the alternatives for resources used in milk production. Pepulation Growth and the Urban Movement During the period 1950-60, the total population of Michigan increased 22.8 percent.2 The most significant change was the trend toward centralization. In 1960, based upon residence, 73.4 percent of the state's population was classified as urban and 26.6 percent as rural. The percentages for 1950 were 70.7 urban and 29.3 rural. The numerical and percentage distributions by residence, since 1920, are shown in Table V. During the past decade, there has been a growing concentration of population in the southern part of the state. The ten Standard MetrOpolitan Statistical Areas, comprising only 14 counties, account for 87 percent of the urban population and 36 percent of the rural ISupra, p. 4. 2Allan Beegle et 81., Michigan Population, 1960, Agricultural Experiment Station Special Bulletin 438 (East Lansing: Michigan State University, 1962), p. 4. 35 population of the state.1 This fact becomes more significant when it is noted that the greatest concentration of fluid milk producers is found in the same general area. This situation leads to competi- tion for the land between urban and agricultural uses. As the de- mand for land for nonagricultural uses becomes greater, the resource will move from agricultural to nonagricultural uses and agricultural production will shift farther away from the population centers. TABLE V.-Population of Michigan, 1920-1960l Population ' Percent Census year Total Urban Rural Urban Rural 1920 3,668,412 2,241,560 1,426,852 61.1 38.9 1930 4,842,325 3,302,075 1,540,250 68.2 31.8 1940 5,256,106 5,454,867 1,801,259 65.7 54.5 1950 6,371,766 4,505,0842 1,868,682 70.7 29.5 1960 7,823,194 5,739,132 2,084,062 73.4 26.6 1Bureau of Business and Economic Research, Michigan Statisti- cal Abstract, 4th edition, compiled under the direction of David I. Verway (Graduate School of Business Administration, Michigan State University, 1962), p. 4. 2The new urban definition includes unincorporated urban pkmes. 1A Standard Metropolitan Statistical Area (SMSA) is a county or group of contiguous counties which contains at least one city of 50,000 or more or "twin cities" with a combined population of at least 50,000. Contiguous counties are included in the SMSA if they are metropolitan in character and are socially and economically integrated with the central city. Ibid., p. 10. Data are not available to indicate the quality of the acre- ages which go into nonfarm uses. There is little doubt, however, that much of the land now in urban and related uses was the better grades of agricultural land. Cities are usually located in places most accessible from the standpoint of transportation. This is usually the lowest, most level, land in the area. Much of the urban resi- dential expansion takes place in the areas most easily developed. Railroads and highways are usually located on the most easily trav- eled routes. An exception is that parks and recreation areas tend to take land which is not particularly suited for agriculture. Thus, the urban movement of Michigan's population has provided an alterna- tive use for land which is not available in states which are more agriculturally oriented. There are other factors associated with the urban movement which deserve mention in this discussion of resource alternatives. These include: (1) the sociological aspects of the changes in values of the rural population, (2) the effect of urbanization on the farm prOperty tax structure, and (3) the impact on the farm wage rate structure. These will be discussed as they relate to the analysis of resource alternatives. Sociological Aspects The impact of the urban movement on the rural population has been great. The declining isolation of farmers and the growth of larger trade center communities have contributed to the changes in rural society. The farmer's relative isolation from others in our society is largely disappearing because of modern transportation and communication systems. Many of the former rural institutions such as schools and churches have been transferred to the larger community. 57 The extension of industrial technology, the growth of urban markets, the increase of transportation facilities, and the general rise in the standard of living have exerted tremendous pressure upon the farmer to become a more integral part of the vast and in- finitely complex economic system. Values and ideals, once held exclusively by the rural segment of the population, are giving way to those normally held by urban residents. values and attitudes regarding hard work and leisure and the concept of comfort no longer differentiate rural and urban society. With the adOpted urban attitudes toward work, leisure, and comfort, rural people are taking on the status values of urban so- ciety. The concept of farming as a way of life, different from other occupations, is gradually changing. The way of life of the farmer and those in other occupations are becoming similar. Farmers, be- cause of increased contacts and the change in the nature of the business of farming, are accepting the values and ideals which were once only associated with the urban society. The location of milk production near urban centers puts the dairy farmer in close contact with these urban ideals. A recent study of the membership of Michigan Milk Producers' Association suggested a change in the attitude toward complete farmer inde- pendence.l Farmers recognize that they are a part of a larger so- ciety and are willing to accept broader roles. Off-farm work has brought farmers into contact with organized labor unions and their process of bargaining. The willingness of the membership to back lGlynn McBride and Glen L. Taggart, "Michigan Milk Producers' Association: An Analysis of MMPA-Member Relations, Attitudes, and Characteristics,” Quarter1y_Bulletin, (February, 1957), 301-306. 38 their association in bargaining ventures indicates a change in atti- tudes toward bargaining and a recognition of their ability to in- fluence their own well—being by such ventures. Through closer contact with urban society, sociological and psychological barriers to resource reallocation are being removed. The rural population is becoming aware of the opportunities available and is willing to take advantage of them. Property Tax The upward trend in property tax has given rise to several important problems, not the least of which is the pressure on farmers located near urban areas. Increased property tax often results in the eventual shift of farm land to nonagricultural uses. Rising taxes have added to the pressure on many farmers in peripheral areas to sell their lands for residential and other ur- banized uses. The burden of the increased taxes has not been equally distributed. Most farmers have felt the squeeze of rising taxes against reduced farm incomes. A recent study of property tax trends in Michigan, and their effect on farmers, provides evidence that an unequal tax burden has been placed upon farmers located near an urban area.1 The high tax rates applied in the urbanized areas often constitute a real burden to those remaining farmers who depend upon agriculture for their livelihood. Property taxes have risen in urbanized areas as a result of IWilliam H. Heneberry and Raleigh Barlowe, Property Tax Trends AffectinggMichigan Farmers, Agricultural‘Experiment Station Special Bulletin 421 (East Lansing: Michigan State University, 1959). 39 higher valuations and higher tax rates.1 The farmers in these areas often find themselves paying more than their share of the tax load for two reasons. First, the delay in getting newly constructed properties on the tax rolls often shifts a major share of the ex- penses in the early years of suburban expansion to the farmers in the area. Second, there is a tendency in partially suburbanized areas to assess farm properties at higher levels relative to current market value. The assessed valuation of farm land is often based on its potential value as subdivided property rather than its pro- ductive capacity as farm land. One measure of the impact of increased property taxes on farmers is the relationship of taxes to income. The property tax is levied without respect to the property owner's current income. Such taxes, however, represent a production expense to the farmer. Table VI shows the relationship of taxes on farm property to‘net farm income for the period 1949-1961. During that period, net income was quite variable while taxes on farm property increased steadily. In 1961, the amount of prOperty taxes paid was almost triple the amount paid in 1949 while net income was considerably less. This combination of a rising level of taxes and a low average level of farm income has given rise to serious problems for some farmers. In Michigan, these problems are especially acute in areas surrounding urban expansion and in areas where farm lands are of relatively low productive potential. What, then, are the implications of the changing property tax structure? First, the total tax burden is not likely to be lessened 1Ibid., p. 25. 40 TABLE VI.-Farmers' realized net farm income and taxes levied on farm property, Michigan, 1949-1961l Taxes as a percentage of net income Farmers' realized net income Taxes on Before payment After payment farm Before After Year I of property tax of property tax property taxes taxes --------- million dollars - - — - - - - percent - - 1949 316.5 302.0 14.5 4.58 4.80 1950 309.9 294.9 15.0 4.84 5.09 1951 360.9 344.9 16.0 4.43 4.64 1952 349.8 332.8 17.0 4.86 5.11 1953 330.4 312.9 17.5 5.30 5.59 1954 292.5 272.8 19.7 6.74 7.22 1955 266.1 242.2 23.9 8.98 9.87 1956 275.7 250.0 25.7 9.32 10.28 1957 294.4 265.0 29.4 9.99 11.09 1958 270.3 238.2 32.1 11.88 13.48 1959 233.4 199.7 33.7 14.44 16.88 1960 264 4 226.1 38.3 14.49 16.94 1961 278.9 238.4 40.4 14.49 16.95 lUnited States Department of Agriculture, Economic Research Service, Farm Income, State Estimates, 1949-61, Supplement to the Farm Income Situation FIS-187 (Washington: Office, 1962), pp. 18, 41-65. U. S. Government Printing 41 so long as the property tax remains an important part of the tax structure. This is because of the continued shift of the popula- tion to suburban areas and the increasing costs of services required in these areas. As the tax increases because of the higher value assessments associated with the ripening of the land for suburban uses, the cost of using land for agricultural purposes will become prohibitive and the land will shift to nonagricultural uses. A second inference is that the location of milk production in relation to the population centers is such that the impact on dairy farmers is probably proportionately greater. Much of the land which shifts to nonfarm uses may shift away from use in milk production. Farm Wage Rates Farm wage rates in Michigan have increased during the past dec- ade but at a slower rate than nonfarm wage rates, as shown in Table VII. Bbwever, this may not be representative of the effects of wage increases on the production of milk. Dairy farming requires a spe- cialized type of hired labor and therefore must pay a higher wage. The index does not properly reflect the impact on milk production, be- cause of the amount of lower paid, less specialized, farm labor used in Michigan. The location of milk production, with respect to the in- dustrial centers, is such that it competes for labor and therefore must pay wages accordingly. This means a higher cost of production. Consequently, centralization of the population has influenced the farm wage rate structure. The dairy industry has felt undue pressure because of the high percentage of total hired farm labor which is used on dairy farms. 42 TABLE VII.-Index of nonfarm and farm wage rates, Michigan, 1950-19601 (1947-49 = 100) Gross weekly earnings of production workers in Year manufacturing industries Farm.wage rates 1950 114 99 1951 123 114 1952 136 120 1953 143 125 1954 145 125 1955 157 128 1956 157 133 1957 162 136 1958 164 134 1959 179 138 1960 185 139 Board, Farm Labor (February, 1961), p. 7. Michigan Department of Labor, Vols. XVIII-XXIX. 1United States Department of Agriculture, Crop Reporting iichigan Labor and Industry, 43 Resource Alternatives in NOnagricultural Uses Alternatives outside of agriculture have played a large role in the shift of resources out of agriculture in recent years. This is especially true for labor. Expansion in the nonfarm sector of our economy has facilitated the movement of labor to nonfarm Jobs. Dur- ing the past decade, resources have shifted to nonfarm uses in Michi- gan. The most noticeable shifts to nonagricultural uses have been observed for labor and land. An analysis of the changes in the em- ployment of the labor force and the sources of income gives some indication of the nonfarm alternatives for labor. The movement of land to nonagricultural uses was discussed earlier in this chapter. Employment of the Labor Force Farm labor has been able to move quite readily into nonfarm employment in recent years. Some evidence of this is shown in Table VIII. The total labor force increased 6.5 percent since 1952. At the same time the nonfarm labor force increased 7.6 percent and the farm labor force decreased 10.7 percent, indicating a movement of labor from the farm to nonfarm labor force. During the same period, there was a slight increase in nonfarm employment. Sources of Income Another indication of the relative importance of agricultural and nonagricultural uses of labor is the sources of income. Table IX is an attempt to show the relative position of Michigan, Wisconsin, and Massachusetts, with respect to sources of income. Wisconsin is considered an important dairy state with few alternatives for re- sources, while Massachusetts is considered as being highly 44 TABLE VIII.-Michigan labor force and employment by major industries, selected years Percent change, Item 1952 1961 1952 to 1961 (thousands) Total labor force 2,790.0 2,971.7 6.5 Nonfarm labor force 2,623.0 2,822.5 7.6 Farm labor force 167.0 149.2 10.7 Tbtal nonfarm employment 2,499.0 2,514.3 0.6 Wage and salary workers 2,275.0 2,221.2 2.4 Manufacturing industries 1,104.0 870.3 21.2 Durable good industries 924.0 682.3 26.2 Hen-durable good industries 180.0 188.0 4.4 Nonrmanufacturing industries 1,171.0 1,015.0 13.3 Government2 - 336.0 - 1Bureau of Business and Economic Research, Michigan Statis- tical Abstract, 4th edition, compiled under the direction of David I. Verway (Graduate School of Business Administration, Michigan State University, 1962), pp. 64-65. 2Not available for 1952. 45 TABLE IX.-Industria1 sources of civilian income received by persons participating in production, selected states, selected years Michigan Massachusetts Wisconsin Source 1950 1960 1950 1960 1950 1960 Million dollars Total civilian income 9,151 15,093 6,008 10,001 4,223 6,985 Percent of total Farms 3.6 2.0 1.5 0.7 12.1 6.2 Mining 0.? 0.7 0.1 0.1 0.3 0.3 Contract construction 4.9 5.3 5.4 5.4 5.4 6.4 Manufacturing 48.6 45.3 38.2 36.1 36.7 38.3 Wholesale and retail trade 17.9 16.8 20.8 18.8 19.7 18.3 Finance, insurance and real estate 2.4 3.3 4.7 5.8 2.8 3.8 Transportation 3.7 3.3 4.0 3.3 4.5 4.1 Communications and public utilities 2.3 2.7 2.6 2.9 2.2 2.5 Services 8.6 10.5 12.0 15.0 8.6 10.5 Government 7.1 10.0 10.0 11.1 7.4 9.3 Other 0.1 0.1 0.5 0.4 0.2 0.2 lUnited States Department of Commerce, Office of Business Eco- nomics, Personal Income by States since 1929, a supplement to the Survey of Current Business (1956), pp. 210-213. zUnited States Department of Commerce, Office of Business Eco- nomics, Survey of Current Business, XLII (August, 1961), p. 19. 46 industrialized, affording ample alternatives for resources used in milk production. A comparison of the income sources for the three states re- veals that only 36 percent of the civilian income in Massachusetts in 1960 originated from manufacturing as compared to 45 percent for Michigan and 38 percent for'lisconsin. Farm income accounted for 2 percent of the civilian income in Michigan, 0.7 percent in.Massa— chusetts, and 6.2 percent in Wisconsin. While this is not conclusive evidence, it does give some idea of the relative importance of the nonagricultural sector of the economy and the alternatives for labor in nonagricultural uses in Michigan. Resource Alternatives in Agricultural Uses Dairy farming competes with both agricultural and nonagri- cultural users of resources for factors of production. Some degree of competition exists within the dairy industry itself between the production of fluid milk and manufacturing milk. The location of milk production with respect to the types of farming areas in the state in- dicates that resources have alternatives in other farm enterprises. Part-time farming also competes for resources on a limited basis. Part-Time Farming Dairy farming is an intensive user of labor and therefore does not readily lend itself to part-time farming.1 In 1959, only 11 percent 1A part-time farm is defined by the census of agriculture as one with farm product sales of $50 to $2,499, the operator is under 65 years of age, and the operator worked off the farm 100 days or more or had a combined family income from nonfarm sources exceeding the value of farm products sold. 47 of the dairy farms in Michigan were classified as part-time farms. A comparison of this percentage with the 31 percent for all farms tends to corroborate the statement that dairy farming is not adaptable to part-time farming. The number of farm operators working off their farms and those working off the farm 100 days or more is shown in Table X. In 1959, more than one-half (54 percent) of the farm operators worked off the farm. Forty-two percent worked off the farm 100 days or more. The trend in both cases has been toward a higher percentage of farm operators supplementing farm income with off-farm employment. It has been estimated that 46 percent of all Michigan farms had off- farm income which exceeded the value of farm products sold in 1959.1 Even though milk production is not a farm enterprise which is often engaged in on a part-time basis, the presence of part-time farming in the state suggests that alternatives for resources do exist in this combination of farm and nonfarm uses. Alternative Farm.Enterprises Data are not available to show the shifts from dairy farming to other types of farming. Hewever, the type of farming areas in Lower Michigan gives some indication of the alternatives available. Type of farming areas are designated on the basis of soil, climate, and markets for the different farm products. Along the western edge of the state, dairy competes with fruit production and truck farming. In the areas of the highest concentra- tion of fluid milk producers (surrounding the metropolitan areas in TIA. Allan Schmid and Fred H. Abel, Michigan Agriculture, 00- operative Extension Service Miscellaneous Series Circular E-22 (East Lansing: Michigan State University, 1962), p. 15. 48 TABLE X.-Number of farm operators working off farms, Michigan, census years 1929-19591 Census year Farm Operators working off their farms Farm operators working off their farms 100 days or more 1929 1934 1939 1944 1949 1954 1959 (number) 60,311 56,782 60,468 60,133 72,494 75,707 60,626 (percent) 35.6 28.9 32.2 34.3 46.6 54.5 54.2 (number) 25,135 25,569 39,792 45,941 49,595 54,705 47,161 (percent) 14.8 13.0 21.2 26.2 31.9 39.4 42.1 lMichigan Department of Agriculture, Cooperative Crop Report- ing Service, Michigan Agricultural Statistics (1961), p. 51. 49 Southern Michigan), the greatest competition is from the enterprises of corn, livestock, and small grains. The competition with cash crops takes on added significance when considering the extent to which dairy farmers are already en- gaged in these enterprises. In 1961, home grown feeds made up 75 percent of the total concentrate ration fed to milk cows.l Corn and the small grains (oats, barley, and wheat) made up 88 percent of .the total concentrate ration fed.2 This indicates that a shift from dairy to cash crops would be relatively easy since conditions are conducive to cash crop farming and since much of the investment in equipment has already been made. Implications The analysis in this chapter has been directed at an evaluation of the alternative uses for resources used in milk production. The differing assumptions of the manufacturing milk formula and the eco- nomic formula relative to alternative uses for the factors of pro- duction are the bases for the analysis. The implications for Class I pricing become apparent when the findings are related to these as- sumptions. The spatial location of milk production is such that resources used in milk production have alternatives available in both farm and nonfarm uses. The trend toward centralization of the population has placed pressure on agricultural land surrounding urban areas to shift to nonagricultural uses. Much of the pressure has been in the form of ’1United States Department of Agriculture, ”Concentrate Rations Fed to Milk Cows, 1961," Milk Production, (march, 1962), p. 11. 21bid., p. 10. 50 high property taxes. The resource, labor, has alternatives in nonfarm employment. During the period 1950-1960, the percent of total civilian income received from farming decreased from 3.6 percent to 2.0. At the same time, the farm labor force decreased 10.7 percent, indicating a movement of labor from the farm sector. Other farm enterprises compete with milk production for resources. The shift from dairy farming to cash crops or to beef feeding operations is relatively easy since much of the investment in equipment has already been made. The analyses in the previous two chapters have focused on the changes, trends, and conditions within which the dairy industry operates and their implications for the type of pricing formula used in establishing Class I prices in the Federal order markets of Lower Michigan. The analysis in Chapter III indicated that the manufacturing milk sector is becoming relatively less important when compared to the fluid milk sector. The analysis in Chapter IV, relative to the alternatives for resources used in milk production, provided evidence that resources used in milk production have al- ternative uses in both farm and nonfarm uses and that dairy farming must compete for these factors. When this evidence is considered in relation to the assumptions of the economic formula that the primary market for milk is the fluid market and that the resources used in fluid milk production have al- ternative uses in both farm and nonfarm uses and in light of the analysis, there appears to be justification for recommending a Class I pricing plan for Lower Michigan built around an economic formula. CHAPTER V A RECOMMENDED CLASS I PRICING-PLAN FOR LOWER MICHIGAN Introduction A third objective of this analysis was to recommend a Class I pricing formula which would reflect the findings of this study rela- tive to the assumptions of the manufacturing milk formula and the economic formula. The analyses, in the previous two chapters, of the changes which occurred in the dairy industry and of the alterna- tives for resources used in milk production suggest the appropriate- ness of an economic formula as a basis for'establishing Class I prices in Michigan. Economic Class I pricing formulas are not new to the dairy industry. They were first used in the Boston milk market in 1948. A few other markets have since adopted them. They have not, however, received widespread use. This is apparently because of the fact that economic circumstances to date have not appeared to warrant their consideration. There are certain basic assumptions which must be adhered to in the use of any formula based on past relationships. First, it is assumed that the forces which influenced Class I prices in the past can be quantified. Second, these forces will remain as relevant influences in the future. Third, Class I milk prices should respond 51 52 to these forces in the future in approximately the same way as in the past. Factors Associated with the Price of Milk There are many factors affecting the Class I price of milk. These may be divided into two categories: (1) factors which affect prices generally and which reflect the broad up and down swings of the general level of commodity prices and (2) local supply and de- mand factors which may result in deviations which differ from those associated with the general price behavior. Problems in pricing Class I milk occur most frequently during periods of rapid rise or fall in the general price level. This price level declined sharply in the early thirties. Milk prices declined more slowly than the wholesale prices of all commodities and wholesale food prices because of the bargaining position of the producer co- operative associations. waever, when the Class I price started de- clining it declined faster and to lower levels than other prices. The adjustment would have been easier, perhaps, if the price of milk had started its downward adjustment more promptly and had not fallen so low. There was another period of adjustment following the sharp in- crease in the general price level when price ceilings were removed in 1946. Serious shortages of milk prevailed in some markets. For these reasons, a formula to be used for establishing the Class I price of milk over a period of time should include at least one factor, the primary purpose of which is, to reflect changes in the general eco- nomic conditions. Changes in the supply of and demand for fluid milk, on a local 53 basis, are of several kinds. First, there are those changes result- ing from changing economic conditions in general and reflected in changes in the general price level. Second, there are changes which affect the dairy industry but do not affect the economy in general. Causes of these variations may be an increase in milk output per cow or a change in food habits favorable to milk and dairy product consumption. A third classification of forces includes changes which effect only the local dairy industry such as weather conditions, changes in local health regulations, or changes in the population of the local marketing area. In addition, seasonal changes in the supply of milk occur regularly each year. There are many factors which might be used for the purpose of reflecting changes of a localized nature in the supply of and demand for milk. Some indicators such as Class I utilization, for example, may tend to reflect changes in both supply and demand while others are associated more closely with the changes which occur in either supply or demand. Changes in the supply of fluid milk may be associated with many factors. The production of fluid milk is affected by the cost of feed, labor, and other items in relation to the prices which farmers receive for.milk. Also, alternative opportunities in other farm enter- prises and in nonfarm uses influence the amount of resources employed in milk production. Changes in the amount and distribution of consumer purchasing power and changes in the price of fluid milk relative to the price of other dairy products and foods are among the most important factors associated with variations in the demand for milk. Changes 54 in the size and composition of the population and in the food habits also influence demand but these changes are gradual and do not account for much of the short-term fluctuations in fluid milk sales. Because of the above forces and their influence on the supply of and demand for milk the following types of movements in Class I price occur. First, there are the wide swings associated with the price structure of the entire economy. Second, the small fluctu- ations, usually of short duration, associated with variations in the supply of and demand for milk in general cause the Class I price to vary. A third type of price fluctuation is associated with seasonal changes in the production of milk. Construction of an Economic Formula In the construction of an economic formula, consideration was given to the following components: (1) time period, (2) time unit, (3) base period for indexes, (4) method of weighting the variables, and (5) the selection of economic variables. These will be discussed in order. Time Period Careful attention should be given to selecting a time period over which factor relationships can be studied. Meny series of data are not available over a long period of time. Therefore, it is sometimes necessary to compromise between the variables and the time period used. Abnormal years such as war years are usually left out. The criterion used as a basis for determining whether to include or ex- clude certain years is that the period should be one in which there 55 occurred little, if any, change in price structure. If evidence of this is not readily ascertainable from observing the data, it is possible to statistically test the coefficient of analyses for dif- ferent periods to determine if the coefficients differ significantly. In attempting to arrive at a satisfactory time period, analy- ses for three periods were made to determine if a price structural change had occurred. The periods were: (1) the entire period of 1935-1961, (2) the entire period except the war years of 1942-46, and (3) the war years of 1942-46. The differences between the co- efficients for the three analyses were not statistically significant.1 It was concluded that results for the entire period should be used as the best predictor for the future. On this basis, the time period 1935-1961 was selected. Time Unit .Most analyses of factors that affect the price or consumption of a given commodity are based on annual data -- either calendar or crop years. Annual data are satisfactory if conditions within the period are homogeneous. The time unit should be of such length as to average out the effect of irregular factors and to insure that a relatively homogeneous set of factors are operating. For products produced continuously throughout the year, available data usually re- late to a calendar year. Since dairy products fall into this cate- gory, the calendar year was selected as the most convenient time unit to use. IThe test used to determine if the coefficients differed significantly is equivalent to the test discussed in Richard J. Foote, Analytical Tools for Studying Demand and Price Structures, U. S. Department of Agriculture Handbook No. 146 (washington: U. S. Government Printing Office, 1958), pp. 180-181. 56 Base Period for Indexes A desirable characteristic of a base period is that it cover a period of fairly stable economic conditions. This means the use of periods other than war, threat of war, or depression. Consideration should also be given to the relationship of the period to present conditions so that recent changes may be clearly viewed. The base period should be long enough to cover several seasonal movements. A period of three years is generally accepted as a logical time period. A 1957-59 base was selected. Many U. S. price series already use this base. It has merit in that it is recent enough to be applicable to present conditions. Method of Weighting Variables After the independent variables are selected, each one.must be assigned a weight. There are two methods commonly used for assigning weights to the variables. The method most commonly used is that of arbitrarily weighting each economic factor to obtain the closest fit between a formula price and the actual price of milk in the past. This is essentially a trial and error technique. The method is easy to understand and use. A second method, which provides a statistical weighting for each independent variable, is that known as the regression method. The regression method employs the technique of least squares multiple regression in assigning weights to the variables. Once the regression coefficients, which are the weighting factors, are determined, the operation of the formula is precisely the same as in the case where weights are assigned arbitrarily. The regression method was selected for weighting the variables included in the recommended economic formula. 57 variables Included in the Economic Formula An important step in constructing an economic formula is se- lecting the independent variables. A number of variables were con- sidered. The final decision regarding the variables which should be included in the milk pricing formula was based on a consideration of the following factors: (1) the degree of association with milk prices in the past, (2) availability of data, (3) limitation of the number of variables used and considerations regarding simplicity, and (4) reflection in the formula price of economic conditions both in the national and local economy. Although it is difficult to cate- gorize the areas of influences of each variable, an attempt was made to include a variable in which the primary function was to: (1) re- flect changes in the general price level, (2) reflect changes in local supply of milk, and (3) reflect changes in the demand for milk in the local market. Because there is no clearcut line of demarcation between these influences the categories are overlapping. The group of variables most difficult with which to work, in terms of the past relationship to fluid milk prices, was the one which indicated a change in consumer demand. There are several factors affecting consumer demand. The ones most commonly used are those which show changes in consumer income. During the past decade, the trend in consumer incomes has been upward while the trend in the price of Class I milk has been downward. The opposite would generally be expected. Another variable often used is an index of the percent- age of producer receipts sold for fluid consumption. A question is sometimes raised as to whether this variable should be used to in- dicate a change in supply or to indicate a change in demand since it 58 varies with both the amount of milk delivered to the market and the amount sold for fluid consumption. However, it does indicate a change in the relationship of supply and demand in the market and this is the relevant point. There are conditions existing in each milk market unique to that market. For this reason, a pricing formula must be one that fits the existing market conditions. It was with this point in mind and in line with the above discussion that the following independent variables were selected as price movers in the economic formula re- commended for Lower Michigan. United States wholesale prices The United States wholesale price index is one of many vari- ables which reflects changes in general economic conditions. In the past, difficulty has been encountered in the dairy industry because fluid milk prices have not adjusted promptly to changes in the general price level. In some cases, shortages occurred. Therefore, a measure of the changes in the general economic conditions is desirable in an economic type pricing formula. Aside from the fact that the United States wholesale price index is the most commonly used indicator of changes in the general price level, it has the added desirable char- acteristic of being readily available on a monthly basis. While changes in the general price level eventually influence the supply of and demand for milk locally, the primary function of the United States wholesale price index in the pricing formula is to make ad- justments in Class I prices which reflect the changing general eco- nomic conditions. 59 Percentage of Class I sales The demand for fluid milk is a function of consumer income, population, tastes and preferences, and prices of other commodities. Changes in any or all of these will bring about a change in demand for fluid milk. The variables most used for reflecting changes in demand are ones which reflect changes in consumer income. The diffi- culty of using an income series for reflecting changes in demand for fluid milk nae been discussed. The negative estimated coefficient obtained is contradictory with the logical assumption of the existence of a positive relationship between income and the demand for fluid milk which is made. An index of the percentage of Class I sales in a market may, indirectly, reflect the influence of income on the demand for fluid milk. At the same time, the composition of the index is such that it may reflect a change in thesupply of milk. The percentage of Class I sales indicates the relationship between the demand for and the supply of fluid milk in the market. Changes in supply and demand conditions in a given market may not be reflected in the other variables in the formula. This index tends to adjust the price of Class I milk as the supply and demand relationships within the market change. For these reasons, the index of the percentage of Class I sales is in- cluded as an important variable in the recommended economic formula. Index of prices paid for manufacturinggmilk The price paid for manufacturing milk reflects the changes in the price of various manufactured dairy products. Since manufactured dairy products are sold in a market which is national in scope, their price is established on a broad base. A second reason for including 60 the index of prices paid for manufacturing milk in the pricing formula is that the index gives an indication of changes affect- ing the supply of fluid milk. The supply of both manufacturing milk and fluid.milk is influenced by many of the same factors. A third reason for including the index is the geographic location of Michigan with respect to the concentration of the manufacturing milk industry in nearhy states. Even though the production of manufactur- ing milk is declining in importance in Michigan, the geographic lo- cation of the state does not permit the neglect of the price paid for manufacturing milk in establishing a Class I price. Index of prices received by Michigan farmers for all farm products except dairy products The index of prices received for all farm products except dairy products serves a double role in the pricing formula. The use of an economic formula assumes that fluid milk production competes with other farm enterprises for factors of production. The index of prices received tends to keep fluid milk prices in line with other farm products since it reflects changes in the prices received for compet- ing products. The index also indicates changes in the costs of pro- duction. Prices received for feed grains and roughages make up part of the index. Changes in the price of feed in relation to the price of milk are often reflected in the amount of fluid milk marketed. The variation in feed prices will be reflected in the Class I milk price through this variable. MOdel A regression model was used to assign weights to each inde- pendent variable. The nature of the data and the type of analysis 61 limited the choice of statistical models. The typical form of a re- gression model is: Yj--0<+/1113+52X23 + ““““ +6pxpj + “3 Where: 'YJ is the jth observation of a dependent variable X11 is the jth observation on the 1th independent variable uJ is the jth observation of a random error when j = l, 2,--N a< and .4 are population parameters. OK is the Y intercept when all Xi's equal zero. / is the slope of the simple regression line between Y and Xi when all other 11's are held constant. There are certain assumptions regarding the statistical prop- erties of the regression model used. If the model is used only to get ”good" estimates of Y, the only assumption needed is that the u's are randomly distributed with uniform variance. With this as- sumption, we can obtain the following estimating equation: '1} : a +b1X1 +b2X2+ ----- +bpxp. T is the estimated value of Y. a and b are estimates of GK and 457 parameters. However, to obtain valid statistical tests of significance about the a and b values, additional assumptions are needed. It is usually assumed that the u's are normally distributed with 0 mean and 672 variance. Even.when some other probability distribution is as- sumed, the u's are assumed to be identically distributed. A further assumption is that the lie are independent of the u's. For each ob- servation G, = Y - E+ b1X11+ b2X21+ b3X31+ 1:41:43 ; where “i = 1, 2,----N and are assumed independent with mean - O, variance : crz. 62 The following estimating formula was obtained when the re- gression model was applied to the data for the period 1935-1961. f: - .68 + .49x1 + .4010; + .07x3 + .0414 (.086) (.134) (.110) (.105) R2:.98 5:3.55 N227 Where: ‘Y Index of the effective Class I price for milk in the Southern Michigan Federal order market Index of U. S. wholesale prices (all commodities) N H II N to I - Index of prices paid for manufacturing milk by midwest condenseries N cs u Index of prices received for all farm.products except dairy products by Michigan farmers Index of percentage of total milk used in Class I sales in the Southern Michigan market. Interpretation The figures in parentheses are the standard errors of the net regression coefficients. The t-ratios are used to test if the coefficients differ from zero. The t-ratio is the ratio of an estimated coefficient to its standard error, i. 8., tbi : bi Sbi The significance probabilities associated with the t-ratios obtained from this regression equation are as follows: .001 for b1 .01 for be .54 for b3 .68 for b4 These values represent the smallest significance level for which the H1 : b1 : 0, where i = l, 2, 3, 4 respectively, would be rejected. S is the standard error of estimate. It is also referred to A as the standard deviation of the residuals (Y - Y). The standard .. error of estimate is an estimate of how well the regression line fits the data. A measure of the "goodness of fit" for the above formula is S = 3.55. The coefficient of multiple determination (R2) is defined as: R2 2 WW‘E’ELEQ. When the coefficient 0 a um of quares of of multiple determination is converted to a percentage basis it indicates the percentage of the variance of Y explained by, or as- sociated with, changes in the independent variables. The R2 of .98 indicates that 98 percent of the variation in Class I milk prices was explained or associated with the changes in four independent variables. When working with a predictive model, as in the case of the Class I pricing formula, the problem often arises as to the level of significance used for determining whether particular variables are useful in explaining the variation in the dependent variable. Usually, the arbitrary levels of 5 percent and 1 percent for* ck (probability of Type I error) are chosen without considering the cost or size of 76? (probability of Type II error). It is impossible to simultaneously minimize Type I and Type II errors. ¢j,and +5? are not independent. Their statistical relationship is dependent upon H5 H1, and the assumption made regarding the underlying prob- ability distribution of the parameters in the population. The decision of which probability level to choose for the Type I error may be de- termined in terms of the costs of each type of error. The cost of one type of error may be high or low relative to the cost of the other type of error. Given a loss function, ck and-{gyshould be 64 chosen so as to minimize the expected loss.1 There are other bases for leaving seemingly insignificant variables in a pricing formula. Reasons relating to particular variables were set forth in the section devoted to a discussion of the selection of variables to include in the formula. It was pointed out that changes in the supply and demand for milk may oc- cur because of conditions unique to the particular market. While the changes are particularly relevant to fluid milk prices in that market, they may not be reflected by the remaining variables. An example is the impact of extreme weather conditions on supply in a particular market. Variables were included to reflect changes in economic conditions both locally and on a national scale. While these two areas are not always independent, there are times when both conditions are not reflected simultaneously by the same vari- able. A similar argument can be made for retaining an independent variable which is highly intercorrelated with another independent variable. The usual procedure is to drop one of the variables or to combine the two highly intercorrelated variables. When two in- dependent variables are highly intercorrelated, the interpretation of the bi's is often misleading. Usually statements are made concern- ing one bi assuming the other variable is constant. When.two inde- pendent variables are highly interrelated, the joint effects as mea- sured by both bi's should be discussed. TIFor a more detailed discussion of the loss function as it relates to choosing a level of significance see Lester'V. Mandar- scheid, An Introduction to Statistical Hypothesis Testihg,.Agri- cultural Economics Mimeo 867 (East lensing: Michigan State Univer- sity, 1962), pp. 1-5. 65 In spite of the low levels of significance, the high inter- correlation with other independent variables in the formula, and the fact that the R2 increases when variables X5 and 14 are deleted, these variables are retained as an important part of the Class I pricing formula.1 The error of eliminating these variables might be greater than the error of leaving them in the formula. Application The estimated Y, using the above model, is referred to as the formula index. The application of the model to Class I pricing in- volves adjusting the Class I price for the base period by using the resulting formula index. The resulting price becomes the basic for- mula price which is subject to adjustments necessary to arrive at a minimum Class I price. The estimated coefficients of the model are applied to the most recent monthly indexes of the variables in order to derive a formula index used in establishing the current monthly Class I price. Evaluation of the Economic Formula One test for a predictive model is a demonstration of how well it predicts outside the sample period. In this case, the sample period used was 1935-61. The period of time since 1961 is not of sufficient length to permit this test. However, there are some de- sirable characteristics of a Class I price which may be observed. These may serve as criteria in evaluating the model. Some of these characteristics are: (1) price certainty, (2) seasonal price behavior, and (3) favorable hfiar-market price relationships. IThe R2 increases from .9776 to .9782 when X5 is deleted and to .9784 when 14 is deleted. 66 Price Certainty Uncertainties prevail in every phase of the production and marketing of milk. The price received for milk may vary sharply from one period to another. The uncertainty associated with wide fluctuations in price often hampers decision making at both the farm and processor levels. Formula pricing has eliminated, to some extent, much of the extreme fluctuations in milk prices and has in- jected a degree of certainty, which is crucial to the allocation of resources. The extreme fluctuations in milk prices following world war II precipitated an interest in economic formulas as a means of preventing any reoccurrence. Most Federal order pricing plans in- clude provisions which prevent wide swings in prices and thus inject a degree of certainty into milk pricing. With the greater degree of certainty associated with Federal order pricing, the milk producer is able to plan a pattern of production which will facilitate the mar- keting process. Seasonal Price Behavior Most Federal orders contain provisions providing for vari- ation in milk prices for different seasons of the year. This is an effort to encourage a pattern of production which is more closely aligned with consumption. .Milk production tends to vary seasonally, while consumption remains relatively constant. There are several reasons why a more even pattern of milk production throughout the year is desirable. The pattern of pro- duction has some effect upon prices received by milk producers. The amount of milk in a given market, above the amount sold for fluid consumption, must be disposed of at a price comparable to manufacturing 67 mulk prices. Manufacturing plants must have sufficient capacity to handle the amount of milk received during the peak production season. In many cases, this means unused capacity during other seasons of the year. The excess capacity represents added investments in fixed assets and increased per unit fixed costs. During the past several years, manufacturing milk prices have been supported at some speci- fied percent of parity. This has been made effective through the establishment of prices at which the government would purchase certain manufactured dairy products. The prices of the manufactured products are set at such levels that allow most processors to purchase manu- facturing milk at the established support price. However, this does not mean that every processor will, or must, pay the support price for manufacturing milk. If processing costs are high, the price paid for milk will be lower. The manufacturing plant with excess capacity will usually have higher fixed costs. Variable costs must then be reduced such that total per unit costs will be in line with that of competitors. The price paid to the producer often suffers the brunt of the reduction in variable costs. Because of the effect of seasonal production on milk prices, it seems logical that an attempt be made to encourage a more even flow of milk to the market. Mest attempts along this line have been in the form of price incentives built into the pricing plan. The schedule of seasonal adjustments discussed in a later section is provided for this purpose. Milk Prices in Surrounding Markets With the advent of new technology in transportation, packaging, refrigeration, etc., price alignment between and among marketing areas 68 has come to be a major consideration in price establishment. Dis- ruptive movements of milk from unregulated areas into order markets .could result in a breakdown of the order pricing machinery. In such a case, a major objective of an order, that of retaining the pro- ceeds of a market for those producers who regularly supply the market with milk, could not be met. This suggests the need fur concern with proper interbmarket price alignment as well as provisions relating to pricing milk which might come from unregulated areas.l Figure 3 shows graphically the relationship of the economic formula Class I price for the Southern.Michigan order to the minimum Class I price in the surrounding markets of Chicago, Toledo, and Northeastern Ohio. The Southern Michigan order price averaged slightly higher than the other market prices during the period 1952-61. However, during the years of greatest difference, 1957 to 1961, the effective prices in the surrounding markets were negotiated prices in- stead of the minimum Class I prices shown. If the effective prices were compared a closer relationship would have existed. Comparison of Historical Class I Price and the Economic Formula Class I Price A comparison of the actual Class I price with the economic formula price is interesting but relatively unimportant in evaluat- ing a formula. The economic formula price, the minimum Class I price, and the negotiated Class I price since 1956 are shown in Figure 4. There was a wide difference between the actual and the economic formula ’1The provision designed for this purpose, the compensatory pay- ment provision, has recently been questioned regarding its constitu- tionality. See Lehigh Valley COOperative Farmers et al. I: United States at 81., 82 S. Ct. 1168 (1962). 69 Prise .——- Southern Michigan 5.50 —————— . Toledo : -——--———~ Northeastern Ohio 5.00 _ Chicago V\\, , ._____. . /,>