Click for next page ( 424


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 423
Productivity Measurement in the Men's and Boys' Clothing Industry LEON GREENBERG KramerAssociates, Inc. INTRODUCTION This study covers "men's, youths' and boys' suits coats and overcoats" as defined in Industry 231 of the Standard Industrial Classification Manual. The industry produces a few broad categories of products, e.g., suits, coats, overcoats, uniforms. Within each category there is a considerable variation by fabric, weave, and price line (i.e., quality). The objective of this study is to evaluate the statistics available for measuring productivity, output per unit of input, in the industry. In order to measure trends in productivity one needs to develop two types of measures, an output index and an input index (labor, capital). Indexes of output can be constructed using physical units combined with some form of weight or using deflated values. In either case, the statistics should be in sufficiently detailed categories to take account of changes in the volume of each category produced from one period to the next. If there were no change in the relative mix of products, there would be no problem, but usually this is not the case. In the apparel industries the variety of products and the frequent changes in product specification brought about by changes in fashion can result in serious problems of data comparability over time. Measures of labor input, employment and hours, are available from both the Bureau of the Census and the Bureau of Labor Statistics. Two questions need to be explored. One is the extent of agreement, or dis- agreement, between the data from the two sources and their compat 423

OCR for page 423
424 PAPERS ibility with the output data. The other is the relative accuracy of the two series. Data on capital stock are needed to construct indexes of capital productivity or multi-factor productivity. There are, in general, some major deficiencies in the statistical information on capital stock. While this industry is labor intensive and indexes of capital productivity would not be very significant, this report will examine the available capital stock data. OUTPUT KINDS OF DATA AVAILABLE There are three major sources of output data for this industry, Census of Manufactures (CM), Annual Survey of Manufactures (ASM), and Current Industrial Reports (CIR). Value of industry shipments is available from both CM and ASM, the latter being essentially a sample survey extension of the former. There might be an advantage to using the CM and ASM industry ship- ments data because employment, hours, and capital expenditures data are collected in the same surveys. Thus, there would be direct compati- bility between the output and the input data. However, the output data from these surveys are available only in four broad product classes- men's suits, men's overcoats and topcoats, men's tailored dress and sport coats, and boys' suits, coats, and tailored jackets. Information in this broad degree of classification is quite inadequate for productivity measurements purposes, as we shall see later. The CIR collects and reports data in considerably more detail. They have in the past, for example, collected and reported quantity data by price line, by fabric, by construction for men's suits and for men's sportcoats; boys' suits and coats are reported separately (see Appendix A). Revisions are planned for 1978 that will increase the amount of detail; for example, for men's suits a breakdown by type of fabric for each type of construction, more breakdowns for boys' garments, and more detail for various types of men's garments (see Appendix B). The CIR advantage in terms of product detail is tempered somewhat by the fact that the CIR report is on a product wherever-made basis so that it may not be exactly equal to industry production or shipments. However, the specialization and coverage ratios are high (93 and 96 percent, respectively, in 1972) and have hardly changed between the Census of Manufacturer years 1967 and 1972 (Table 14. So, the CIR

OCR for page 423
Productivity Measurement in the Clothing Industry TABLE 1 Selected Statistics, Industry 231 1970 1972 Value of shipments (million $) 1912.1 1920.0 2396.8 Cost of materials (million $) 862.3 839.8 1064.6 Percent of shipments 45.1 43.7 44.4 Specialization ratio 94 NA 93 Coverage ratio 97 NA 96 NA, not applicable. 425 wherever-made data could be used as a proxy for industry shipments or to calculate percent distributions for application to ASM shipments data. In this industry, work is sometimes contracted out. The value of this "commission" work is included in the industry shipment data, but the result is an overstatement of total value because the commission work is also included in the value reported by the prime manufacturer (or jobber). The C1R product shipments exclude the commission work. The ASM product shipments provide a breakout of commission work, so adjustments would be possible. Commission work as a percent of total value of shipments varies somewhat, ranging from 17 to 21 percent for the 5 years 1972-1976. Over that period, total shipments rose 11.2 percent and shipments less commission work rose 15.4 percent. This means that a productivity measure based on total value of industry shipments, without adjustment for commission work, would understate the 4-year increase by four percentage points (Table 2~.~ The levels shown by the CIR and ASM reports are somewhat different, as might be expected, because of differences in sampling, definition, (e.g., shipments versus production) and inclusion or exclusion of com- mission work. These differences might not be serious if the trends shown by the three types of data were reasonably close to each other. But there are very serious differences in some of the year-to-year changes, and the worst of these is 1974-1975. In that year, CIR production, wherever made, rose 8 percent, ASM shipments on a wherever-made basis rose 4.3 percent, and ASM industry shipments declined 5.2 percent. If industry shipments are adjusted for inventory change, to derive production, the decline is even greater (Table 3~. iBased on current-dollar value data. The difference based on constant-dollar value would be the same since commission work is not available by product class and would have to be deflated by an implicit price index calculated from current- and constant-dollar value of shipments.

OCR for page 423
426 TABLE 2 Commission Work as Percent of Shipments, Industry 231 PAPERS Commission Work Total Shipments Shipments Amount Percent Less Commission 19722230.0 445.5 20.01784.5 19732305.7 484.4 21.01821.3 19742251.2 415.5 18.51835.7 19752346.6 418.8 17.81927.8 19762479.4 420.4 17.02059.0 Percent change 1972-197611.2 15.4 SOURCE: Annual Survey of Manufactures 1976, Value of Product Shipments. According to Census Bureau representatives, the principal reason for this disparity stems from the reporting of men's leisure suits. Ap- parently, many firms had been reporting this new fashion item under product categories not classified in Industry 231. In 1975, the Census Bureau gave leisure suits a "home" in Industry 231, thus contributing to the large increase in production in 1975 as shown by the CIR. According to Census Bureau representatives, the leisure suits, on the one hand, were secondary products of firms not classified in Industry 231; on the other hand, some firms would have been reclassified to Industry 231 because leisure suits had become their primary product. But, once a sample is drawn (from the CM universe), only the large (certainty cell) firms are reclassified. This explanation suggests that (1) if the coverage ratio had been recomputed in 1975, it would have been different than the 1972 ratio and (2) the industry shipments data are an understatement of the true value of shipments. To return to the planned CIR revision, unfortunately it contemplates that a major type of classification will be discontinued. Previously, number of men's suits and number of men's dress and sport coats were each collected by price line. This method of classification has been found to be unrealistic, variant among manufacturers, and often mis- leading, and it has been dropped. In the past, the quality line of men's clothing was identified by "grade"; the grading was based, to some extent, on the degree of hand stitching and other hand tailoring performed on the garment. This method of classification although still used by some of the "old-timers" has become outmoded and is not in general use by the industry.

OCR for page 423
Productivity Measurement in the Clothing Industry TABLE 3 Industry 231 Value of Shipments and Production 427 Product Data Millions of Dollars Industry Shipments and Production ASMt Production Shipments Year CIRC ASM~ Inventory inventory Shipments Beginning End of Pro ASM of Year Year auction 1972176822302397 368 4002429 1973182123062492 400 4622554 19741762e22512387 462 4552380 975190323472263 455 3902198 1976201324792406 390 4062422 Percent Change 1972-19733.03.44.0 8.715.55.1 1973-1974 - 3.2 - 2.4 - 4.215.5 - 1.5 - 6.8 1974-1975 8.0 4.3 - 5.2- 1.5 - 14.3 - 7.6 1975-1976 5.8 5.6 6.3- 14.3 4.1 10.2 1972-1976 13.9 11.2 0.46.0 1.5 - 0.3 u Products wherever made. b Shipments by firms classified in sac 231 SOURCE: Annual Survey of Manufactures, General Statistics for Industry Groups and Industries for 1973-1976; Census of Manu factures for 1972. C Excludes contractor (commission) work. SOURCE: Current Industrial Reports, Apparel Annual Reports. Includes contractor (commission) shipments. SOURCE: Annual Survey of Manufactures 1973-1976; Census of Manufactures 1972. e Some leisure suits and leisure-type sport coats may not be included in the 1974 data. Thus, the 1974 data are not directly comparable to the 1975 data. The consequence of these developments is that there is no direct identification of quality differences among garments within the product classes or groups. This has serious implications for productivity measure- ment since labor requirements are affected by product specifications that are related to quality. Changes in the proportions of output ac- counted for by garments of different quality specification do occur in response to changes in consumer demand. The increased casualness of men's wear, reflected in sport coats and leisure suits, is an obvious example. Moreover, the vagaries of fashion bring about other changes in style that can affect labor requirements. For example, it takes more hours to construct a plaid suit than a solid color suit because of the need to match patterns when collating the various components into the final garment.

OCR for page 423
428 PAPERS The chances of collecting additional information in sufficient detail to adequately reflect the product specification changes are extremely remote, especially for small firms. Industry, union, and management representatives are quite skeptical of such a potential venture. The Census Bureau has had difficulties in the past as indicated to a limited degree by the "risk" (not elsewhere classified) in the annual CIR reports. They are now engaged in extensive discussions with the industry in planning for the 1978 CIR revision. NET VERSUS GROSS OUTPUT Detailed output data (such as they are) are usually available only in their usual form of gross output. If the relationship of material purchased to gross output were to change significantly, it would suggest important changes in vertical integration within the industry (although it could be influenced by changes in the mix of fabrics). This would affect the resultant productivity measures, but is not a serious factor for Industry 231. The ratio of value of materials to value of gross output, in current dollars, changed very little between 1967 and 1972, for example, hovering around 45 percent. Constant-dollar figures are not likely to show signif- icantly different results. WEIGHTED PRODUCTION INDEXE S The preferred type of output index for use as a component of a produc- tivity index is a weighted output index, where the detailed products of the industry are combined with certain weights. While the proceeding discussion indicates a very serious limitation in the method because of the lack of sufficient product detail, it is nevertheless useful to review the results of different weighting schemes. Weights based on worker hours, payrolls, value added, value of production or shipments, and employment may be used, in that order of preference, to construct output indexes for use in the productivity index. If all of them are directly proportionate to each other, then, of course, it makes no difference which weight is used. The Bureau of the Census has constructed production indexes for census years using all five types of weights on both a 1967 and 1972 base. But the indexes are not "pure," that is, the weighting is only part of the process. Product class values at the 7-digit (sac) level are deflated by a price index or unit-value relative computed from CM unit and value data. These are summed to the 5-digit level. The 5-digit

OCR for page 423
Productivity Measurement in the Clothing Industry TABLE 4 Production Indexes for 1972, Industry 231 Indexes of Production (1967= 100) 1967 1972 Type of Weight Weights Weights Value added 109.8 109.9 Value of shipmentsa 110.0 110.6 Employment 108.3 107.6 Production worker hours 109.2 107.8 Payrolls 107.4 107.2 a Shipments adjusted for inventory change. Products deflated at 7-digit level, then combined into 5-digit and 4-digit groups with varying types of weights, as indicated. SOURCE: 1972 Census of Manufactures, Indexes of Production 429 classes and there are only four of them (commission work is excluded)- are combined with the five types of weights (Table 4~. The various kinds of weighting yield somewhat different results. The indexes for 1972 (1967 = 100) range from 107.4 to 110.0 using 1967 weights and from 107.2 to 110.6 using 1972 weights. This means that the annual rates of change in the resultant productivity indexes for the 5-year period could differ by as much as 0.6 of a percentage point under different weighting schemes. There is a very important limitation in the preceding analysis. As indicated, the 7-digit product classes are deflated by a price index or a unit value relative (index); but most of the time, in the past, Census has had to use the unit-value relative for deflation because of the un- availability of adequate price indexes. Unfortunately, the calculation of a unit value in each of the Census periods (to derive a unit-value index) does not take care of the detailed product differentiation that exists within the 7-digit class. It is simply total value divided by number of units. And within the 7-digit grouping, significant and elusive changes in product occur. The important point to note is that a substantial part of the so-called weighted production indexes is obtained by deflating values, however weak or strong that deflation process might be. The Federal Reserve Board indexes of production for Industry 231 are based merely on a count of the number of units produced in the industry without regard to product classification.

OCR for page 423
430 DEFLATED VALUE OF OUTPUT PAPERS An alternative index of production, deflated value of output, would be equal to a unit-value weighted index if all the requirements of detail and compatibility were met. It has already been pointed out that there is insufficient product detail available to construct reliable (and ac- ceptable) unit-value weighted indexes. This problem might be over- come through the use of a deflated value index since the amount of price and product detail needed for deflation is not likely to be as great as the amount of detail needed for a unit-value weighted index. The principle that lies behind industry (or other group) price in- dexes is, of course, that the price movement of the sample of com- modities that are priced is reflective of the price movement of non- priced commodities. Price data are currently available for products of the men's clothing industry through the BUS wholesale price program. Three products are priced, men's lightweight suits, men's regular weight suits, and sport coats. According to the BES, these commodities account for about 85 percent of the industry's output. More precisely, however, the product classes that these commodities represent (men's suits, men's tailored dress and sport coats and jackets) accounted for 85 percent of the value of shipments of products wherever made.2 The product specifications used by BES are broad, so that various types of fabric construction and price line are likely to be included in the sample. Nearly 10 price quotations are obtained for each of the three commodity groups. The BES considers this pricing to be adequate for construction of an Industry 231 price index, beginning with January 1978; coverage was considered inadequate prior to that date. This reviewer has doubts about the current adequacy of the Industry index as a deflator for shipments or production. There is some skepticism in the industry about the ability of BES to price identical garments from one time period to the next, because of fashion/specification changes. BES does not think this is a serious limitation for three reasons: (1) manufacturers are given instructions to price garments with the same specifications from one pricing period to the next and are followed up by commodity analysts, (2) the variations that do occur may be in either direction, that is, they may add to or subtract from the value of the garment, and (3) the variations that might be ignored (by the respondent) and therefore not taken account 2 annual Survey of Manufactures, Value of Product Shipments.

OCR for page 423
Productivity Measurement in the Clothing Industry 431 of are likely to be of small consequence and to have a small cost/price impact. The BES budget for fiscal year 1979 (already approved by the House) contemplates a major expansion in the wholesale price program. Pos- sibly as many as 100 establishments will be in the Industry 231 sample. Attempts will be made to price all of the commodity lines made by each manufacturer. Consequently, beginning in 1980, BES expects to have 200-300 price quotations for this industry. This very substantial expansion of price data will make it possible to construct an industry price index that is quite satisfactory for de- flating the value of output (or shipments). It is important, however, that the sample selection of commodities for pricing should take into account the product classes that will be reported by the Census Bureau so that deflation can take place at the lowest possible level of detail. LABOR INPUT There are two major government sources of labor input data, Annual Survey of Manufactures from the Census Bureau and Employment and Earnings from the BES. Both agencies publish employment and hours but with some differences in what is actually collected. Census collects number of production workers for 1 week in each of 4 months during the year (March, May, August, September) from which they derive an annual average. They collect number of all other employees for only 1 week in the month of March, apparently on the assumption of relative stability in this cadre of workers. BES collects number of all employees and number of production workers for 1 week in each of the 12 months, so they have a larger sample of time periods covered. They also do extensive editing each month and adjust their sample estimates to universe benchwork estimates about once a year. Production-worker hours are collected by the Census Bureau to cover the entire year, that is, to cover all pay periods. Production-worker hours are collected by BES for only 1 week in each of 12 months so their time coverage is less than that of the Census Bureau. Statistically, therefore, each series has its relative advantages and disadvantages. An additional statistical advantage held by the Census series is that its shipments and labor input data came Prom the same survey and are thus directly compatible with each other. There is also a conceptual difference between the two series: BES in- cludes all hours paid for, i.e., vacation, holidays, and sick leave. Census

OCR for page 423
at At a' - 1 v, At o cd o 432 Cal m C,2 c) o - _ ~ - it> 5 - . 0 3 Ct 04 :5 to o ,, ~ o Ct V, . o o ~ - C) o V' ~ ~ ~ rid ~ ~ ~ ~ us _ . . . . . . . . . . _ Do ~ ~ 0 ~ rid 0 ~ up rig ~ ~ Go ~ or ~ ~ ~ us As oO rid ~ ~ ~ ~ Go ~ _ . . . . . . . . . . o ~ or ~ o ~ ~ Us o ~ o ~ Do Go Us Go ~ Go ~ o . . . . . . . . . . Us Cal ~ ~ ~ ~ ~Go Do r~ C~ ~ ~ ~ ~ o ~ ~ ~ ~ U~ ~ o o ~ ~ oo ~ ~ ~ oo ~ oo oo ~o ~ ~ ~_______ o ~o 0 _ ~ U~ oo 0 _ r~ ~ ~ u~ u: ~o r~ ~ ~ ~ ~ r~ _ ~ 00 ~ 0 0 0 U) ~o 0 . . . . . . . . . . u~ oO ~ ~ u~ oo _ _ _ 0 ~ ~ o~ ~ ~ ~ _ _ _ _ oo ~ _ 0 U) 0 ~ ~ C~ ~ . . . . . . . . . . O ~ ~ ~ ~ ~ r~ ~ oo _ 0 _ _ 0 oo oo ________ 0 _ ~ ~ ~r ~r~ \~} r~ C~ _ _ _ ~ _ _ _ _ _ ~

OCR for page 423
_ ~ ~ oo ~ ~ ~ ~ ~ rat ~ cr . . . . . . . . . . . . rot _ ~ oO ~ O _ ~ ~ ~ as - 1 - 1 - - - - 1 1 1 1 1 1 ~ 0 ret oO O Cal ~ us ~ ~ ~ O . . . . . . . . . . . . _ ~ cr _ ~ ~ 0 cr _ 1 - 1 - - 1 1 1 1 1 1 ~ 0 cr oo 0 ran ~ ~ ~ ~ ~ ~ . . . . . . . . . . . . or ~ ~ - - ~ or 1 - 1 - - 1 1 ~ ~ on ~ - ~ oo oo A) try c~] try ~) O ~ oO ~ O ~ ~ ~ ~ _ ro - 1 _ _ _ (~ (~) 1 ' 1 1 1 1 1 cr us ~ ~ ~ 0 ~ urn cot ~ ~ oo . . . . . . . . . . . . _ 0 0 oo rot _ cr us ~ ~ 0 0 - 1 1 - - ~ ~ 1 ' ' 1 1 1 1 ~o ~ ~ o - ~D a~ ~ a~ s: ) 1 1 1 1 ~ ~ ~ ~ o ~D ~- - ~ - c~ oo ~ ~ u~ ~ - ~ oo ~ . . . . . . . . . . a~ oo ~ - ~ ~ r~ ~ 0 0 1 - _ ~ r-) ' ' ' 1 1 1 1 r~ r~ ~ ~ ~ ~ ~ ~ r~ - - - - - - - - _ ~ r~ cr - - - - - - - - - l c~ ~ . - s" cd - ~ c: ~ - ; - o ~ s ~ C} ~ ~D Cd cr, _ L~, o U] _ C~ o _ ~ 0 - Ct C, _ ._ 53 C,2 ,,, _ ~ ._ , ~ g ._ _ s" ._ - 3 C~ ~> ._ o 3 ._ ~3 s~ 3 o s~ z s: ._ . cn ~ a, _ s~ s~ C) -t . _ _ ~ _ ~ ~ o - :~. Ct V' ~ _ Ct C) :^ o Ct Ct C) s" o - 1 _ e~ _ U, ~ s:: V} {e C) .= .. _1 s" o U) 433

OCR for page 423
434 PAPERS hours are "plant" hours, and exclude time away from the plant whether paid for or not. Plant hours, sometimes referred to as "hours worked" would be preferred for productivity measurement but it should be noted that productivity trends are often compared with earning trends, and the latter, based on BES data, are on an hours paid for basis. For the most part, the year-to-year changes shown by the two series are reasonably close to each other. Major exceptions were in 1972 and, to a somewhat lesser extent, in 1975 and 1976. For the two subperiods, 1967-1972 and 1972-1976, the differences were less than 3 percentage points and for the entire 9-year period 1967-1976 the difference was 3.6 percentage points (Table 5~. We might expect Census Bureau average plant hours to increase less than BUS hours paid for (or to decline more) as workers obtain increased vacation and holiday time, but the Census decline was less than that shown by BUS (1967-19761. This could be a consequence of changing labor force structure (e.g., if there are more new workers with less paid leave time) or of statistical abberrations. CAPITAL INPUT Industry 231 is a relatively labor intensive industry. In 1976 production- worker wages were 40 percent of value added compared with 27 per- cent for all operating manufacturing establishments. Capital expenditures were 1.9 percent of value added compared with 8.0 percent for all manufacturing. Consequently, measures of capital productivity are not as significant for this industry as they would be for other, less labor intensive, industries. Nevertheless, measures of capital productivity, and/or multi-factor productivity might be of some interest. The BES maintains two constant-dollar series on capital stock for Industry 231 gross and net. The data are unpublished, but it may be possible for the BUS productivity office to use the series for produc- tivity calculations. Current-dollar stock figures are derived from capital expenditures data in the Census Annual Survey of Manufacturers. BES derives an average life expectancy, separately for plant and equipment, which it uses to convert expenditures (investment) into stock. The gross stocks calculation adds the total investment each year to previously cumulated investment and leaves it at full value until it is subtracted at the end of its life. The net stock method depreciates the value of the investment each year of its estimated life until it reaches zero.

OCR for page 423
Productivity Measurement in the Clothing Industry 435 Indexes for deflating the current-dollar values are derived by use of a weighted price index wherein the weights are obtained from occa- sional studies of the Bureau of Economic Analysis of the Department of Commerce. The weights are assumed (by BES) to be constant, at least until the new data become available. BUS uses 22 different equipment- type groups for its weighting patterns for all industries, but many of them would have a weight of zero in Industry 231. The capital expenditures data are consistent with industry shipments data since they both come from the same source (ASM). However, once an expenditure is put into the time matrix for calculating stock, it re- mains there. So, if an establishment is reclassified from Industry 231 to some other industry, its shipments will be reflected in the ship- ments statistics for the two different industries. On the other hand, its previously derived stock figures will remain in Industry 231. This is not likely to be a serious problem in the industry. As already indicated, the BES capital stock figures are unpublished and not for citation, but they have been made available to the reviewer for analytical purposes. Capital stock, in constant dollars, whether measured on a gross or net basis, increased substantially, more than 75 percent, from 1967 to 1974 (latest date available). At the same time, employment declined substantially. The result was that capital stock per employee more than doubled during the period. It is obvious, from the preceding, that labor productivity and capital productivity estimates would have shown substantially different trends over the past 10 years. It suggests that capital productivity and multi- factor productivity measures should be attempted if labor productivity measures are developed for the industry. SUMMARY AND RECOMMENDATIONS The men's clothing industry, like other apparel industries is highly susceptible to the vagaries of fashion. As a consequence, there are frequent and sometimes significant changes in the output of the industry related either to the mix of products or to their quality. These changes can affect the labor input required for the manufacture of men's clothing. It is not feasible to collect output (or shipments) data in sufficient detail to capture all of the important changes in mix. Moreover, even if such an attempt were made, new fabrics or styles sometimes move into the market so quickly that the data collection apparatus is not likely to function efficiently enough to perceive such changes until they

OCR for page 423
436 PAPERS already result in high volume of production. The period of major change would, therefore, be missed. Examples are permanent press, double knits, and leisure suits. It is even less practical to try to embrace, in a production reporting system, the many small changes in specification that are dictated by style and fashion. Similarly, it would be impractical to attempt to collect the data needed for derivation of unit hour, value added, or value weights. The BUS looked into this matter about 20 years ago and concluded that a weighted production measure for Industry 231, for productivity index purposes, was not feasible. The product-data problems have not really changed very much since then and the conclusions are the same. A weighted production index for use in a productivity ratio for this industry is not feasible and is not recommended. There is much greater promise in the potential development of in- dexes of deflated value of output. Unfortunately, the weakness in cur- rently available price data precludes reliable estimates for past years. However, the planned expansion in the BES wholesale price statistics will result in a substantial increase in price data coverage for Indus- try 231 in 1980. Therefore, it is recommended that plans be made for development of indexes of deflated value of output beginning with 1980 data. It is also recommended that representatives of the offices of prices and productivity at the BUS consult with each other and with the Census Bureau in order that the product sample for price data and the data collection and reporting plans for output are as compatible with each other as is feasible under the two programs. Labor input components (employment, hours) are available from two sources, the Census Bureau and BES. Each source has its own statistical advantages and disadvantages. While it is almost a toss-up as to which source is the preferred one, the Census data have a slight edge for two reasons. They come from the same survey and are directly compatible with the industry output and shipments data. They are conceptually more desirable for productivity measurement since they relate to hours worked or, more precisely, hours at the plant. BUS hours paid for would be affected in different ways by increased leisure time; a decline in hours at work resulting from an increase in paid vacations would not affect average hours, while a decline in the workweek would result in a de- cline in hours. The use of Census hours, is therefore, recommended. The difference between Census and BUS employment and hours, not only in level but especially in trend, are sometimes too great to be "comfortable." This has been a long-time problem and previous at- tempts at reconciliation have not been successful. Nevertheless, it is

OCR for page 423
Productivity Measurement in the Clothing Industry 437 recommended that reconciliation efforts be resumed or continued. Matching data reported by the same respondents to BUS and Census surveys is one avenue. Uniform industry classification, through efforts of the two agencies, is another. We have no particular recommendation for the improvements in capital stock data other than to support whatever efforts have been and will be undertaken to improve the detail, the valuation, and the price deflator mechanisms for the capital expenditure and capital stock estimates for all industries. The available statistics for Industry 231 indicate that a substantial increase in capital stock per worker has taken place. If reliable production indexes are eventually developed, then both labor and capital productivity measures should also be derived.

OCR for page 423
a, o 'e us o v, - o an cd o - c) o of ~ of . o of a' 'e pa ~ - ~ ~c) v] ~ . ho 5- ~ o of - ~ ~ m _ ~ ~ Cat - _ ~ ~ O Z V ~ ~ ._ ~ V2 3- ~ . ~ ~ . - o ~ = 0 ~ rat - o, oo ~ ~ Go r ~~ ~ ~ Go ~ Up _ ~ ~ t- ~ ~ _ ~ a: ~ us ~ ~Go 00 0 ~ O _ ~ ~ ~~ rat ~ 00 _ - _ _ . ) O ~r ~ ~D -t o . ~- ~ ~ oo Ci -- ~ ~D _ r~ ~ ~ o - - x ~ - ~- ~ ~ o oo ~ o oo~ ~ u~ ~ ~ ~ u)~ ~ oo ~ ~ o - - - - - - ~ ~ oo ~ - - - - - ~ o - - c~ o . - U2 e ~o _ C) . O3 o C~ o ~ =<,, 3 & o~" ~'= ~o ~'~ ~ 'Y ~ ~ U~ ~;, ~ ~ O _ ~. ~ ~ =~ _ _ ~ ~ ~ ~ ~ ~) ~ ,,, ,,, ~_ ,~, C) _ _ _ _1 _ _ _ _ _- _ ____ ___ __ ___ 438

OCR for page 423
To oG rid 0 ~ ~ ~ ~ ~Go ~ 0 ~ us ~ _ ~ ~ ~ - ~ An 0 ~ ~ ~ ~(~) _ cat _ _ ~ rat ~ _ cr ~ 0 0 us ~ cot ~00 AL as ~- ~ ~ - - ~ - - - ~ x x x x x x ~ - - - ~ - in oooO O Go or) ~ ~ ~ _ Go ~0 us us _ rat us _ us 00 ~us ~ ~ ~ ~ U) Lr) Go no ~ A) on C'] _1 _ ~_ _ l=,, Z ~ Zz 0 a ~u 3 c; ~8 z 0 f~ ~a oo ~a ~ ~ ~ ~ ~ ~A 1 ~ I 'D (~] c~) <= ~- 1 ~ ~ ~ ~ Lr) ~ ~ 00 00 ~0 _ C~ C ~-) 1 _ C~ ~1 - - O O O O r ~ ~ ~ ~ ~r ~ r~ ro c ~r~ ~ C~ ~ ~ ~ ~ r~ 439

OCR for page 423
- ~ r~oo u ~ 0 ~ ~ ~ ~rid AL - ~ orDo ~Go ~ to -- ~ - ~ ~Lo _ ~ Lr) _ rid to ~- us . . ~- o ~ - - - x x co ~rid Do ~ o ~ ~ ~o ~ ~ ~ ~ 0r~ ~- - --- -~ ~r~ oo _ ~ ~_ .~ - ~ z c~ ~o c} o ~ ~ o ~ - o o o c) 3 ~ c: Ct ~ o , - o o C~ . _ m Ct ~o o C) - C~ 5 ;~2 ~: - o o 3 o - o o 3 - e~ C) CJ x V, ._ ~n C~ o ~ o o Ct C) ~: V' ._ ~D Ct o C-] ~ ._ C) s~ .~ o c: c) ~ m ~ ~ c ~ - - ~- - - - - r~ ~ r ~r~ 440 ~: o - E~ o o P. - U) ; ~4 O C} ,~ o _ C~ ._ C~ - - . ~ ~ C} - . C) ,D C) ~ e~ ; - m _ C~ ro ~ r~ _ _ _ _ ~_ _ _ ~ r ~r~ x - z o <: V' o - z z

OCR for page 423
Productivity Measurement in the Clothing Industry APPENDIX TABLE B1 Working Plan for CIR for 1978: Part D Reference List 441 Quantity Cut or Knit Product Unit of Item Item Code Item Description Measure Code Code Value of Net Shipments (thou- sands of dollars) Section II-Production and Value of Shipments MEN S AND BOYS OUTERWEAR Suits Men's tailored suits, including tailored uniform (exclude casual, ski, and snow suits). (Re- port suits with one pair of pants and suits with two pairs of pants on the basis of manufacturer's wholesale list price for coat and one pair of pants.) (Total) Of total quantity, how many were made from Knitfabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics Woven fabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics Of total quantity, how many were Trios (coat with a matching pair of pants and a contrasting pair of pants) Men's casual (nontailored) suits (total) Of total quantity, how many were made from Knit fabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics Woven fabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics units units units units units units units units Units units units units units units units units units units -

OCR for page 423
442 APPENDIX TABLE B1 Continued PAPERS Quantity Cut or Knit Product Code Item Description Value of Net Shipments (thou- sands of dollars) Unit of Item Item Measure Code Code Boys' tailored suits, except ski, snow, casual, and wash-quits (include uniform), (total) Of total quantity, how many were made from Knit fabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics Coats, Vests, and Jackets, except Work Boys' tailored suit-type separate dress and sport coats, including uniform (total) Of total quantity, how many were made from Knit fabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics Woven fabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics units units units units units units units units units units units units units units Boys' casual (nontailored) type sport coats (total) units Of total quantity, how many were made from Knit fabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics units units units units

OCR for page 423
Productivity Measurement in the Clothing Industry APPENDIX TABLE B1 Continued 443 Quantity Cut or Knit Value of Net Shipments (thou- sands of dollars) Product Code Item Description Unit of Item Item Measure Code Code Woven fabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics Boys' separate vests (excluding sweaters vests; report sweater vests as sweaters) (total) Of total quantity, how many were made from Knit fabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics Woven fabric All cotton and chiefly cotton All man-made and chiefly man-made All wool and chiefly wool All other fabrics units units units units units units units units units units units units units

OCR for page 423