TIMING

Data would be requested on a weekly basis from both packers and retailers with the requirement that all data must be in the mail and postmarked by no later than Thursday following the week for which data applies.

SAMPLE SIZE

Preliminary estimates of the sizes of sample needed to obtain U.S. average packer and retailer margins were made for us by SRS statisticians. A 99 percentage confidence level that packer margins will fall within plus or minus 0.3 cents per pound on a live weight basis was used. On a carcass weight basis the 0.3 is about 0.5 cents per pound. The retail sample number was calculated using a 99 percentage confidence level that the retail margin will be correct to within one cent (or) on a retail weight basis. This is about 0.7 cents per pound on a carcass weight basis.

Sample size would be 32 plants for both pork packing plants and beef packing plants. The universe assumed eliminated all plants. killing less than 5 head per hour.

The retail sample size of 42 retail chain divisions was also computed after making several assumptions concerning the universe. Small independent food stores were not considered as part of the universe. Chain divisions rather than individual stores were used. A chain. division is the management unit for all stores of a particular chain in a city or geographic area that fall under the same pricing structure. Note that sample sizes were determined on the basis of the amount of variation we would accept in the margins, not the variation in prices paid or received. This is because variations in quality and type of product handled cause larger variations in product prices than is found in margins themselves. A firm handling a certain type of product in a specific area may pay more (or less) for their product, but in turn sell it for more (or less). While U.S. average prices or values would also be published with the margin series, the statistical properties would be related to the gross margin or difference between prices paid and received.

USDA RESOURCES REQUIRED FOR PROPOSAL 3

Resource needs are hard to estimate as the amount of data and computations required to summarize this data won't be fully realized until it is received.

Presently 4 statistical clerks and one professional are needed in our retail meat price survey. This survey does not include volume information. Expanding from this base it would appear that once the study was underway we would need:

6 statistical clerks for the retail margin data summarization, 2 statistical clerks to utilize the retail data submitted for use in the spread series,

6 statistical clerks to work with the data summarization for the packer margin series (3 for beef and 3 for pork),

1 programmer that would write and maintain programs and. oversee running all programs,

1 professional to work as liaison with all sample firms maintaining sample and orderly inflow of data,

1⁄4 to 1⁄2 professional to supervise margin and price spread series work for beef and pork working with the theory and concepts involved,

1 secretary to handle typing, T&A's, etc.

This would mean that 10 clerks, 1 programmer, 1 secretary and a professional assigned part-time would be needed in addition to the staff currently working on price spreads for meat. These personnel are the anticipated full-time needs after the study is well under way. The initial procurement of the sample, visiting and explaining data needs to the firms selected in the sample, and specification of procedures would require additional personnel as follows:

1 to 2 additional liaison-type professionals to contact the sample firms,

1 additional programmer to help in setting up programs.

D. PROPOSAL 4

An optimal system would measure margins for all beef (and pork) at each level in the industry and would include all the various institutions that move meat to the final consumer. While optimal in the sense of information available, it would require the development of an extensive data gathering and processing system. All retailers, restaurants, institutions and other firms selling to the consumer would have to provide price, volume, and cutting test information. Any wholesalers, jobbers, fabricators, etc. would also have to provide complete data from which margins could be constructed. Meat processors and packers would also be asked to supply complete data information on their operations. Sample sizes would have to be large enough to provide representative data for each firm category and method of handling meat. A very large group of personnel and large computer capability would be required to analyze these data.

E. WHAT THE PROPOSALS DO NOT CONTAIN

These proposals have indicated only the development of gross margin and spread series. Development of cost component data for these margins and spreads are not included. The present annual breakdown into cost components by USDA would be continued. Assuming proposals requiring statistical samples of firms were chosen, an annual survey of costs could be added to the reports requested of cooperating firms.

65-141 O-76-26

APPENDIX IV

WHITE PAN BREAD: COMPONENTS OF RETAIL PRICE PER 1-POUND LOAF, 1973

[General notes of explanation]

These notes have two main purposes: (1) comparing 1973 data with 1972 data, and (2) giving some explanation of how these estimates are derived and sources of data used.

In both years, the spreads at each stage of function are based on ERS estimates of prices and price spreads as we usually develop them. However, in 1972 price spreads data were computed from price estimates which were rounded to one decimal whereas in 1973 price estimates are carried to four decimals. This is necessary to permit more meaningful and accurate comparisons. For example, the data rounded to a tenth cent a loaf showed a spread between the f.o.b. mill sales value of flour and cost of flour f.o.b. bakery of 0.3 cent a loaf in 1972 and only 0.2 cent in 1973. A freight cost per loaf reduction of a third in a year seems indefensible. When these data are computed to four decimals the freight was 0.3041 cent in 1972 and 0.2700 in 1973, still a reduction of over 11 percent which is hard to reconcile in view of reported freight rate increases. However, some obvious possible justifiying factors are:

(1) Trend to more shipments in bulk,

(2) trend to more shipments by truck or lower rates,

(3) trend to fewer small bakeries and relative more large bakeries buying in larger volume,

(4) shift to heavier reporting by mills closer to the market and away from wheat supplies as well as the trend to milling orientation toward the market, and

(5) possible error in reporting.

Cost components are updated to 1973 by use of 1972 estimates, trends in cost factors, and unless they could be derived directly, profits represent the residual. These data are based on concurrent spreads and may not represent spreads actually taken which would include a spread for storage. Separate explanations of cost components. for each concurrent price spread may be helpful for now and future references.

(1) Assembly of wheat represents the spread between farm value and cost of wheat f.o.b. the mill. The spread is for merchandising and transportation. Merchandising is based on assumption that wheat moved into country elevator by truck and out by rail and into a terminal and out by rail on its way to the mill. It is assumed that merchandising spreads covered costs including profits equal to 10 percent of the gross spread. Costs are derived by Allen Schienbein from ERS513 and related ERS files.

(2) Flour miller's spread represents the difference between the cost of wheat f.o.b. mill and f.o.b. mill sales value of flour. Based on NCFM data, this spread is divided 80-20 to processing and wholesaling. Instead of using all costs for 1972 as a base, profits are lowered to percentage of sales in NCFM and other cost items recomputed. Each item is adjusted up by multiplying by respective increases from 1972 to 1973 as provided by Terry Crawford, except net interest and profits. Net interest is derived by multiplying 1972 figure by the cross product of Crawford's factor increase and the increase in wheat price. Profits are derived as the residual when costs are deducted from the spread. (3) Transportation of flour to bakery is the spread between the price of flour f.o.b. the bakery and the derived price f.o.b. the mill as reported by 20 flour milling firms who report sales of bread-type flour to ERS.

(4) Baker-wholesaler spread is the difference between the price of bread received by bakers for bread delivered to the store shelf and the cost of flour f.o.b. the bakery. This deviates from baker-wholesaler spreads normally reported, because included in this spread is the cost of non-flour ingredients to the baker. Farm value of the farm produced products are estimated but cost components for processing and wholesaling these products as well as raw costs and marketing cost of industrial products used are not available. So the cost of these products to the baker is included as a part of the item under processing indicated "other." It amounted to 2.5353 cents in 1973.

Cost components for 1973 are derived by applying index of changes provided by Crawford, with exception of interest, to 1972 cost items. Net interest index is multiplied by increase in cost of ingredients and that index applied to the 1972 figures. Profits are based on quarterly financial data reported by FTC-SEC. The spread, excluding profits and cost of non-flour ingredients, is divided proportionally among the adjusted 1973 cost components and non-flour ingredients added to "other" cost item and profit, which adds to the spread as estimated.

(5) Retailer's spread is the difference between the retail price and the wholesale price as we estimate. Cost components are updated from 1972 to 1973 by using index of increase by Crawford to 1972 estimates, except for net interest. Net interest factor increase is Crawford's factor times the ratio of wholesale price of bread in 1973 to 1972. Profit is the residual after deducting cost components from the price spread.

The retail price of a 1-pound loaf of bread in 1973 averaged 27.6 cents, 2.9 cents above the price in 1972. The 1973 price was 11.9 percent above the 1972 price (tables 1, 2, and 3)1. However, the December price was 28.1 percent higher than the January price, with most of the increase coming after price controls were lifted during the third quarter. Due to restraints on wholesale and retail prices of bread during most of the year and freedom of farm value and flour prices to move up, there was variation in changes in money amounts going to different services performed in putting bread in the consumers hands in 1973.

Farm prices of wheat and prices of flour increased much more in 1973 than did bread prices at wholesale and retail. As is the usual pattern of price movements in times of scarce supplies and strong

1 See pp. V and 296.

demand, prices of raw materials rise more, in relative terms, than finished product prices. This relationship was aided by restraints on finished product prices under economic controls in effect during most of the year.

The farm value of wheat in a loaf of bread increased from 2.9 to 4.1 cents in 1973, or 1.2 cents. This was an increase of 44.9 percent. This was second to the percentage increase in the miller's spread.

The flour miller's spread between the sales value of flour and cost of wheat increased from .67 cent in 1972 to .98 cent in 1973, or .31 cent. This was an increase of 47.3 percent and the largest relative increase for any function, including the production of wheat.

The baker-wholesaler spread for baking and wholesaling bread increased .53 cent per loaf but only 3.3 percent over 1972. This was the lowest increase for any function in the process.

The retailer's spread increased .80 cent and averaged 5.38 cents in 1973. This was an increase of 17.6 percent.

The spread for assembling wheat to the flour mill (transportation and merchandising) increased from .29 to .32 cent per loaf of bread, or 9.7 percent, in 1973. This amounted to 23.0 and 26.1 cents per bushel in 1972 and 1973, respectively.

Based on the spread in prices of flour between the mill door and the bakery, derived from flour sales reported by a sample of milling firms, cost of transportation declined about 11.2 percent in 1972-73. Counteracting rate increases for transportation are possible trends (1) toward milling flour closer to flour demand centers, (2) from bagged to bulk flour shipments carrying lower rates, (3) toward probable lower-rate modes of transportation, and (4) toward larger bakeries, and larger volume purchases.

In view of the variation in changes in spreads by functions, changes in costs or profits, or both, obviously varied also by function in 1973. Changes in costs differed but changes in profits were greater (table 3). Profits in 1973 for assembling wheat and retailing bread about doubled but were up for flour milling three times 1972 levels. The baker-wholesaler, unable to pass on in bread prices increased costs, saw his profits decline from 1972 levels about one-third in 1973.