The use of the product data for classifying establishments by industry is only one of the reasons for gathering such information; product statistics constitute basic source material for industrial marketers and government analysts. They are essential for production indexes, price indexes, and input-output studies.

Several considerations went into the design of the Product Reference List or, to state it differently, the commodity coding structure for the industrial census. The Product Reference List used for the previous Providencia Census was taken as the frame of reference, and departures from it were made only for sufficient justification such as product changes during the intervening decade or improved international comparability. Weight was also given in the selection of products (and materials) to the List of Selected Products and Materials developed as a guide for reporting commodity information in the 1973 World Programme. This UN list is limited to products that can be measured in quantitative terms as well as in value terms. It has been utilized in the census planning to select those products for which physical quantity of output would be requested in addition to value.

11

The definition of each product (and materials) category must clearly distinguish between the commodities included there and those included in other categories. The provision of residual categories (other or not elsewhere classified, "n.e.c.") at appropriate points in the structure accommodates items not otherwise provided for in the classification.

For the 1975 Providencia Census, there are separate Product Reference Lists for the

11Ibid, footnote 2.

manufacturing and mineral sectors. As in the 1963 census, there are also Materials Reference Lists for establishments reporting on the long form. As is the case of products, a separate materials list has been developed for manufacturing as distinct from the mineral industries. 4.42 Geographic area classifications.-As discussed earlier, the other primary method of classifying industrial data is by geographic area. For Providencia, the geographical classification follows the definitions of that country's administrative subdivisions. three levels of these subdivisions, in descending order of size and economic importance, are: provinces, counties, and wards. (Refer to exhibit I-2-5, which gives the number of counties and wards for each of the country's 12 provinces.)

The

4.43 Size of establishment classification.--In addition to industry and geography, establishment size is a frequently employed way of classifying economic data. Providencia's previous census publications had size tables with the following class intervals: 1-4, 5-9, 10-19, 20-49, 50-99, and 100 or more persons engaged. Because of the country's economic progress during the ensuing 12 years and because of the interest of data users in the larger producing units, the top size class was expanded for the 1975 census into three classes: 100-249 persons engaged, 250-499, and 500 and over.

4.44 Legal form of organization classification.--In presenting data classified by ownership and legal form of organization, Providencia followed the categories recommended by IASI; namely individual proprietorships, partnerships, corporations or joint stock companies, cooperatives, public corporations, and others. However, they substituted the term "government enterprises" for "public corporations." For purposes of supplemental

analytical reports, a major separation was made between "private enterprises" and "government enterprises". This distinction reflects the strong interest exhibited by developing countries in such data. In addition, data for "private enterprises" were further separated

in the reports according to the legal form classifications listed above. In planning tabulations of this type, it is recommended that each country adopt the classifications most in accord with its national modes of legal organization.

As with the other concepts involved in an industrial census, the selection of the number of items, the degree of detail for each, and which respondents should reply to the questions, required the National Statistical Office to explore a number of options.

Again, the UN principles and recommendations were taken as a major point of reference. Experiences gained in previous censuses conducted by Providencia were carefully evaluated, and the NSO judged which data were critical for detailed industry and small area tabulations, as well as for the establishment of an industrial directory; the NSO decided to gather such information from all establishTabulation of a few items for all

ments.

economic units is a basic statistical investment which has great future value. These items are specified as both long and short form inquiries in exhibit III-2-1, "Data Items Collected in the 1975 Providencia Industrial Census." A wider range of data, including greater detail for some of the key items referred to above, is needed to satisfy the analytical requirements of industry, government and the public. This informational package was obtained from all producing units of significant size. The specific inquiries are listed in the columns headed "Long Form" under Manufacturing and Mineral Industries and those for the Electricity and Gas Industries, where a single form will be used for all establishments.

The use of short forms enabled the NSO of Providencia to stay within its budget and reduced the reporting burden for small

establishments, which typically have more
limited records than the larger ones. A
short form was used for each manufacturing
or mineral establishment with fewer than 10
persons engaged. In the discussion of data
items in this chapter the expression "from
all establishments" means that the item
appears on both the long and short forms,
whereas "from larger establishments only"
signifies that item is a long-form inquiry.
Cottage industries were canvassed with a
special short form. Inasmuch as most items
are common to the entire industrial census
(manufacturing, mining, electricity and gas)
that fact should be assumed to be the case
unless an exception is pointed out in the
discussion of data items which follows.

In the selection of data items for Providencia's Industrial Census, priority was given to the accuracy and completeness of results over an extensive list of inquiries. A questionnaire with too many items can easily have an adverse effect on the validity of data that are critical to a successful census. In general, the items that were selected were those suggested by the UN for countries with developing statistical programs. Recommendations on specific topics were submitted to the NSO by advisory groups and interested individuals. Consultations were held with data users, in both the public and private sectors. Full consideration was given to the statistical needs of the government for the planning and conduct of a comprehensive industrial development program. Taken into account was the aim of producing a set of baseline data from which annual and more frequent surveys could be generated as key

elements in an integrated industrial statistics program for Providencia.

A number of meetings were organized with representatives of respondent groups to obtain their views on the availability of the data and cost of supplying it. Final selection of the topics to be included in the industrial census, however, as well as the phrasing of the inquiries and their arrangement on the questionnaire, were the responsibility of the National Statistical Office.

The items of data to be gathered, tabulated, and published in the census are grouped into the following seven principal subject areas: (a) identification and classification information; (b) employment and earnings, (c) inventories; (d) fixed capital formation, (e) input costs (including detailed materials, fuels, and electricity), (f) gross output (including detailed products), and (g) value added. The list includes items that will be calculated by the NSO rather than called for on the report forms (for example, value added and the energy equivalent of fuels consumed). A number of countries, particularly those with well-developed industrial statistics, will wish to go beyond this program to meet their own special requirements or to serve common regional needs.

2. DESCRIPTION OF DATA ITEMS

A census can yield satisfactory results only if the information that is collected is highly useful to the users. Providencia determined that the information contained in the seven major categories listed above was of major importance to its economic development plans. These seven major groups are discussed below.

2.1 Identification and classification information

Section I of the questionnaire deals with the description of the establishment. Included in the section are items on the (a) name of establishment, (b) its physical location, (c) its economic organization, (d) form of ownership and legal organization, and (e) period of operation and change in ownership (if any) during the census year.

All of these items except for economic organization of the establishment will be obtained from all establishments. Multi-unit operations are not a significant factor for establishments with fewer than 10 persons en(See exhibits III-4-1 to III-4-5.)

gaged.

2.2 Employment and earnings

This section of the industrial census is concerned with the labor element as a major factor of input. It measures the number of persons engaged as of a mid-month week toward the end of the census year together with the weekly payroll for the categories of paid labor. This is asked of all establishments in each sector as is the total payroll for the year. The payroll figures are divided between production workers (operatives) and all other employees. Payrolls include all payments, whether in cash or in kind. Payments in kind include food, clothing, and lodging provided free of charge or at markedly reduced cost by the employer.

Larger establishments are in addition asked to report supplements to wages and salaries (that is, supplementary labor costs not included in payrolls). In recent years, such payments have become an increasingly significant part of total labor remuneration (up to 20 percent of the total labor cost in many countries). The number of production workers employed during a pay period in each of the four quarters of the year is also being asked of

12

larger establishments, as is the number of man-days worked by production workers. In its recommendations for the 1973 World Programme, the UN Statistical Committee stated that mandays worked would be a more practical concept than man-hours worked in the case of developing countries. Experience with the 1963 Economic Census in Providencia indicated difficulty with the standard concept of man-hours worked; therefore, Providencia has adopted the flexibility in the definition of the item now provided by the UN. The National Government of Providencia does not require enterprises to maintain records on the number of hours worked.

2.3 Inventories

This section is designed to measure changes in stocks during the census year, as well as the level of stocks. The information

is being requested only on the manufactures long form and the form for the electricity and gas industries. Inventories are not a significant item in the minerals area. In the manufacturing sector, the smaller establishments find it difficult to report stocks data; however, as a group, they have little impact upon the total level of manufacturing

inventories.

In the electricity and gas industries, the amount of finished product and work-inprocess inventory is minimal; accordingly, no breakdown is called for by stages of fabrication. For manufacturing, the type of inventory detail affords an understanding of the economic forces at work during the census year and provides the necessary adjustment to the value of shipments in the calculation of value added (discussed later in this chapter).

The UN recommendations specify materials and finished product inventory categories as

12Ibid, footnote 2.

priority 1 items but work-in-process as a priority 2 item. Providencia felt that to call for only two of the three components impaired the value added adjustment and introduced a potential source of misreporting by the respondents; therefore, full detail was listed on the manufactures long form.

2.4 Fixed capital formation

New capital formation is a very important element in a country's industrial economy. For Providencia, the phenomenon is being measured by the acquisition of new and used fixed assets defined generally in accordance with UN recommendations. That inquiry, sub-divided between (a) structures and additions to structures and (b) machinery and equipment, appears on all the industrial census forms except for the manufactures short form.

The Providencia Census departs slightly from the UN guidelines in that it includes "transport equipment" with "machinery and equipment" rather than as a separate category. Past censuses and surveys, even in industrialized countries, have revealed a reporting problem when requesting separate figures on transport equipment.

Although the 1963 Providencia Census called for the sales of fixed assets, that item does not appear on the present census forms. Such sales are relatively minor and the item does not enter into the calculation of value added. Sales of fixed assets does affect the computation of "gross additions to fixed assets." The UN 1973 World Programme called for the adoption of that

13

concept only by developed countries and then only as a priority 2 item.

13 Ibid.