"specialization ratio" and the "coverage ratio." The first measures the extent to which the plants in an industry specialize in making the product primary to that industry. It is derived by dividing the total output (both primary and secondary products) of the industry's plants into their output of products primary to the industry. Thus, in the above illustration, the specialization ratio is 0.91; that is, shipments of bolts, nuts, washers, and rivets accounted for 91 percent of all products made and shipped by plants in that industry (755÷ 827=0.91). The coverage ratio measures the extent to which the products primary to an industry are shipped by plants classified in that industry. In the illustration the coverage ratio is 0.88; that is, 88 percent of all shipments of bolts, nuts, washers, and rivets were made by plants primarily engaged in the manufacture of these products (755-860=0.88).

Generally speaking, the higher the specialization and coverage ratios, the greater the confidence which can be placed in any comparison between product-industry statistics. In about half of the industries both of these ratios are over 90 percent, and in such cases the industry and commodity approaches yield virtually identical results; and in about 80 percent of the industries both ratios are 75 percent or

over.

1 In this calculation, the value of miscellaneous receipts is omitted.

Tables 1-A-1-G

These tables show the share of total manufacturing activity in the United States accounted for by the largest manufacturing companies. At the "all manufacturing" level concentration is measured in terms of value added by manufacture.

The standard formula for calculating value added by manufacture in 1958 was modified from the one used for the 1954 census and earlier years. During the earlier period, the value added of an establishment was calculated by subtracting the cost of materials, supplies, containers, fuels, purchased electric energy, and contract work from the value of shipments for products manufactured plus miscellaneous receipts for services rendered. This is known as unadjusted value added in census publications. In 1958, the measure of value added was adjusted for each establishment by taking into account the following items: (a) Value added by merchandising (that is, the difference between the sales value and cost of merchandise sold without further manufacture, processing or assembly, plus (b) the net change in finished goods and work-in-process inventories between the beginning and the end of the year. The resulting figure is the adjusted value added.

Thus, the rankings of the 200 largest companies in 1947 and 1954 were based on unadjusted value added; those for 1958 were based on adjusted value added. It is believed that the use of these different measures for determining size had little effect on the ranking and relative shares of individual companies or on the aggregate shares of the 50, 100, 150, or 200 largest companies in each year.

Information on value added by manufacture for all establishments operated in manufacturing by a given company was aggregated irrespective of the industry classification of the individual establishments. The companies were then arrayed by their value added in each census year and figures for the 50, 100, 150, and 200 largest companies in each year aggregated. These tables exclude the companies' establishments classified outside of manufacturing.

(1) In table 1-A, the companies in each category are not necessarily identical for all 3 years since the size array in each year was developed independently.

(2) In tables 1-B, 1-C, and 1-D, the 50 and 100 largest companies in 1 year were identified and the value added for the same companies totaled in each census year and their share of total value added by manufacture in each year computed. Thus, the tables show the changes in shares of value added taking place for a fixed group of companies.

(3) For tables 1-E, 1-F, and 1-G, the 50 largest companies in 1 census year were selected and the rank of these same companies shown in the other census years. Thus, they measure the shifting of position of a fixed group of companies from one census to the next.

7

TABLE 1-A.—Share of total value added by manufacture accounted for by largest manufacturing companies in 1958 compared with share accounted for by largest companies in 1954 and 1947

TABLE 1-B.-Share of total value added by manufacture accounted for by largest manufacturing companies in 1958 compared with share accounted for by identical companies in 1954 and 1947

TABLE 1-C.-Share of total value added by manufacture accounted for by largest manufacturing companies in 1954 compared with share accounted for by identical companies in 1958 and 1947

TABLE 1-D.-Share of total value added by manufacture accounted for by largest manufacturing companies in 1947 compared with share accounted for by identical companies in 1954 and 1958

TABLE 1-E.-1954 and 1947 status of the 50 largest companies in 1958

TABLE 1-F.-1958 and 1947 status of the 50 largest companies in 1954

TABLE 1-G.—1954 and 1958 status of the 50 largest companies in 1947

34

10

6

35

11

3

1

50

Tables 2 and 2-A

These tables are based on the industry approach described in the general introduction. In using these tables (and also table 3), the following should be kept in mind:

(1) The industry categories are those described in the Standard Industrial Classification Manual, 1945 version (old SIC). They are the same as those used in the 1958 Census of Manufactures preliminary industry reports but differ from the revised definitions used in the final industry reports and in the commodity and industry data appearing in tables 4, 5, and 6 of this report.

(2) The top 4, 8, and 20 companies are the largest for each particular year and are thus not necessarily the same for each year.

(3) Since the definition of the company is confined to the sum of the individual establishments operated by the company in each industry, the same company may appear in several industries if it has diversified activities.

(4) The value of shipments includes the value of primary and secondary products as well as miscellaneous receipts. It also includes transfers of products between manufacturing establishments of the same company (interplant transfers), generally at an approximate market value.

(5) Value added rather than value of shipments has been used in making these calculations for a few industries which contain a large and often indeterminate amount of duplication because the products of some establishments in the industry are used to a significant degree as materials by other establishments in the same industry.

TABLE 2.-Percent of value of shipments accounted for by the largest companies in each manufacturing industry, 1958, compared with 1954, 1947, and 1935— Industries arrayed by industry grouping.

GROUP 20. FOOD AND KINDRED PRODUCTS