Introduction => GENERAL This report contains estimates of the value of product shipments for approximately 1,750 classes of manufactured products. These product class estimates are based on reports from a representative sample of about 55,000 manufacturing establishments included in the Annual Survey of Manufactures (ASM). Comparative data from the 1992 Census of Manufactures also are included. In addition, data from the 1991 through 1982, 1977, and 1972 Annual Surveys and Censuses of Manufactures are also included when available and comparable. => SCOPE AND USE OF ANNUAL SURVEY The ASM is a survey conducted by the Bureau of the Census to provide the key measures of manufacturing activity for industry groups and important industries for intercensal years. These key measures, as well as other detailed statistics for manufacturing, were collected in the censuses of manufactures; the three most recent censuses covered the years 1982, 1987, and 1992. An annual survey was conducted for each of the years between censuses starting with 1949. During intercensal periods, these annual surveys provide a continuous series of basic statistics for industries and furnish benchmarks for current business indicators and for measures of industrial production and productivity. The survey also provides significant data in connection with planning for industrial mobilization. => DESCRIPTION OF SURVEY SAMPLE The statistics presented are estimates compiled for a survey which is composed of two components. The mail portion of the survey is a probability sample of about 55,000 manufacturing establishments selected from a total of about 200,000 establishments. These 200,000 establishments represent all manufacturing establishments of multiunit companies and all single- establishment manufacturing companies mailed schedules in the 1987 Census of Manufactures. This mail portion is supplemented annually by a Social Security Administration list of new single-establishment manufacturing companies opened after 1987 and a list of new manufacturing establishments of multiestablishment companies identified from the Census Bureau's Company Organization Survey. The 1989 through current-year ASM sample is similar to the previous sample. For the current panel, all establishments of companies with 1987 shipments in manufacturing in excess of $500 million were included in the survey panel with certainty. There are approximately 575 such companies collectively accounting for approximately 18,000 establishments. For the remaining portion of the mail survey, the establishment was defined as the sampling unit. For this portion, all establishments with 250 employees or more and establishments with a large value of shipments also were included in the survey panel with certainty. A total of about 10,000 establishments was selected from this portion of the universe with certainty. Therefore, of the 55,000 manufacturing establishments included in the ASM panel, approximately 28,000 were selected with certainty. These certainty establishments collectively accounted for approximately 80 percent of the total value of shipments in the 1987 census. Smaller establishments in the remaining portion of the mail survey were sampled with probabilities ranging from 0.999 to 0.005 in accordance with mathematical theory for optimum allocation of a sample. The probabilities of selection assigned to the smaller establishments were proportional to measures of size determined for each establishment. The measures of size depend directly upon each establishment's 1987 product class values and the historic variability of the year-to-year shipments of each product class. Product classes displaying more volatile year-to-year change in shipments at the establishment level were sampled at a heavier rate. This method of assigning measures of size was used to maximize the precision (that is, minimize the variance of estimates of the year-to-year change) in the value of product class shipments. Implicitly, it also gave weight to differences in employment, value added, and other general statistics, since these are highly correlated with value of shipments. Individual sample selection probabilities were obtained by multiplying each establishment's final measure of size by an overall sampling fraction coefficient calculated to yield a total expected sample size. The sample selection procedure also gave each establishment in the sampling frame an independent chance of selection. This method of independent selection permits the rotation of small establishments out of a given sample panel without introducing a bias into the survey estimates. The nonmail portion of the survey includes all single-establishment companies that were tabulated as administrative records in the 1987 Census of Manufactures and all single-establishment company births in 1988 through 1991 below a payroll cutoff that varied by four-digit SIC. These cutoffs were derived from the nonmail payroll cutoffs used for the 1987 Census of Manufactures by adjusting them for inflation. This portion was not sampled; rather, the data for every establishment in this group were estimated based on selected information obtained annually from the administrative records of the Internal Revenue Service (IRS) and the Social Security Administration (SSA). This administrative-record information, which includes payroll, total employment, industry classification, and physical location of the establishment, was obtained under conditions which safeguard the confidentiality of both tax and census records. Estimates of data other than payroll and employment for these small establishments were developed from industry averages. Although this portion contained approximately 160,000 establishments, it accounted for less than 2 percent of the estimate for total value of shipments at the total manufacturing level. The corresponding estimates for the mail and nonmail establishments were added, along with the base-year differences, as defined in the Description of Estimating Procedure section, to produce the figures shown in this publication. => DESCRIPTION OF ESTIMATING PROCEDURE Most of the ASM estimates were computed using a difference estimation procedure. For each item, the difference between the 1987 census published number for an item total and the linear estimate of the totals from the ASM sample was generated. These base-year differences were then added to the corresponding current-year linear estimates, which include the estimates for the mail and nonmail establishments, to produce the current-year estimates shown in this publication. Estimates developed by this procedure usually are more reliable than comparable linear estimates developed from the current sample data alone. => ADJUSTMENTS OF HISTORICAL DATA During census years, the data tabulated for the ASM panel have typically been low in comparison to the census data. For 1982 and 1977, the ASM level was 2.0 to 2.5 percent low in terms of value of shipments. For 1987, the ASM level was about 3.5 percent low compared to the census. New single- establishment companies that are not completely classified after IRS and SSA coding and the timing of the inclusion of new plants and acquisitions of multi-establishment companies to the frame contribute to the low ASM estimate for 1987. Procedures were developed for the 1988 ASM to correct the timing of including new plants of multi-establishment companies. However, the identification of new single-establishment companies remains a problem. Not only was the 1987 ASM low, but also the published ASM estimates between the census years of 1982 and 1987 appear to be low. Recently, the Census Bureau staff benchmarked shipments and inventories data from the Manufactures Shipments, Inventories, and Orders survey to the 1987 Census of Manufactures and the 1988 ASM. The benchmarking procedure provides an approximation of the understatement at the total manufacturing level. Similar estimates at the detailed four-digit industry, five-digit product class, or geographic area level are not available. The chart below shows the ASM inventories and shipments data at the total manufacturing level with the comparable adjustments determined through the benchmarking process: Year End-of-year inventories Value of shipments (million dollars) (million dollars) Published Adjusted Published Adjusted 1988 361 472 361 477 2 682 509 2 682 492 1987 332 376 332 619 2 474 646 2 475 906 1986 311 124 317 567 2 260 317 2 335 881 1985 332 370 329 555 2 280 188 2 334 456 1984 329 593 334 236 2 254 430 2 288 184 1983 308 002 307 675 2 054 899 2 070 564 1982 307 212 307 212 1 960 200 1 960 214 => QUALIFICATIONS OF THE DATA The estimates developed from the sample are apt to differ somewhat from the results of a survey covering all companies in the sample lists but otherwise conducted under essentially the same conditions as the actual sample survey. The estimates of the magnitude of the sampling errors (the difference between the estimates obtained and the results theoretically obtained from a comparable, complete-coverage survey) are provided by the standard errors of estimates. The particular sample selected for the ASM is one of many similar probability samples that, by chance, might have been selected under the same specifications. Each of the possible samples would yield somewhat different sets of results, and the standard errors are measures of the variation of all the possible sample estimates around the theoretical, comparable, complete-coverage values. Estimates of the standard errors have been computed from the sample data for selected statistics in this report. They are represented in the form of relative standard errors (the standard errors divided by the estimated values to which they refer). In conjunction with its associated estimate, the relative standard error may be used to define confidence intervals (ranges that would include the comparable, complete-coverage value for specified percentages of all the possible samples). The complete coverage value would be included in the range: 1. From one standard error below to one standard error above the derived estimate for about two-thirds of all possible samples. 2. From two standard errors below to two standard errors above the derived estimate for about 19 out of 20 of all possible samples. 3. From three standard errors below to three standard errors above the derived estimate for nearly all samples. An inference that the comparable, complete-survey result would be within the indicated ranges would be correct in approximately the relative frequencies shown. Those proportions, therefore, may be interpreted as defining the confidence that the estimates from a particular sample would differ from complete-coverage results by as much as one, two, or three standard errors, respectively. For example, suppose an estimated total is shown at 50,000 with an associated relative standard error of 2 percent, that is, a standard error of 1,000 (2 percent of 50,000). There is approximately 67 percent confidence that the interval 49,000 to 51,000 includes the complete- coverage total, about 95 percent confidence that the interval 48,000 to 52,000 includes the complete-coverage total, and almost certain confidence that the interval 47,000 to 53,000 includes the complete-coverage total. In addition to the sample errors, the estimates are subject to various response and operational errors: errors of collection, reporting, coding, transcription, imputation for nonresponse, etc. These operational errors also would occur if a complete canvass were to be conducted under the same conditions as the survey. Explicit measures of their effects generally are not available. However, it is believed that most of the important operational errors were detected and corrected during the Census Bureau's review of the data for reasonableness and consistency. The small operational errors usually remain. To some extent, they are compensating in the aggregated totals shown. When important operational errors were detected too late to correct the estimates, the data were suppressed or were specifically qualified in the files. As derived, the estimated standard errors included part of the effect of the operational errors. The total errors, which depend upon the joint effect of the sampling and operational errors, are usually of the order of size indicated by the standard error, or moderately higher. However, for particular estimates, the total error may considerably exceed the standard errors shown. Any figures shown in the files in this publication having an associated standard error exceeding 15 percent may be combined with higher level totals, creating a broader aggregate, which then may be of acceptable reliability. => DESCRIPTION OF IMPUTATION PROCEDURE The need for imputation in the ASM arises from several sources. Complete nonresponse, partial nonresponse, inconsistent reported data, and the Bureau's desire to reduce the reporting burden for small manufacturers are the primary factors contributing to the need for imputation. For purposes of the ASM, an item is considered to be an impute if the value was not directly reported on the questionnaire, directly derived from other reported items, directly available from supplemental sources, or obtained from the respondent during the analytical review phase. In imputing records for nonrespondents, two basic approaches are followed. For most larger establishments, operational information is available from the prior-year ASM. Imputation for current-year data for establishments with 1992 data available involves adjusting the corresponding 1992 data by estimates of the year-to-year relative change. These estimates are developed based on employment and payroll data from the Bureau of Labor Statistics at the industry group (three-digit SIC) level and are then applied to the 1990 prior-year establishment values for employment and payroll. The remaining data for the establishment are developed based on their prior-year relationship to these two control items. For nonrespondents that have no prior-year record available including establishments not included in the mail portion of the survey, administrative-record data for annual payroll and employment are used. For all other items, industry average relationships between these control items and the item to be imputed are used. At the total manufacturing level, the overall imputation rate is approximately 7 percent of the published value of shipments figure for 1993. However, this rate varies widely by four-digit industry, five-digit product class, and State total. The median imputation rate for a four-digit industry was 10 percent; for a five-digit product class, the rate was somewhat higher; and for a State total, the rate was comparable to the four-digit industry. => PRODUCT CLASSES A product class is a grouping of individual products of an industry. It is designated by a five-digit code, the first four digits indicating the industry and the fifth digit, the specific group of products. The value of shipments for the class of product codes presented here is on a wherever- made basis; for example, it represents total shipments by all manufacturing industries (the industry in which the product class is primary and other industries in which the product class is secondary). In some cases, by definition, a product class is limited to the products of a particular manufacturing process. Wherever another product class accounts for additional production of the same end products, a cross reference is provided; for example, meat packing plant products: 20116, pork, processed, made in meat packing plants (see also code 20136). Table 2 shows total value of products shipped for products that are primary to more than one industry. The total value of shipments of all product classes belonging to an industry also is shown on the same wherever-made basis. It is designated by the four-digit industry code followed by a dash (2013-, if the industry has more than one product class) and by the four-digit industry code followed by another digit if the industry has only one product class (20210). The four-digit product class aggregate is not the same as the total industry value of shipments.