PRODUCTIVITY OF TEXTILE LABOUR IN THE GROUP OF COUNTRIES VISITED
I . INTRODUCTION
. 1. The Economic Commission for Latin America undertook the study of the cotton spinning and weaving
industry of some Latin-American countries with a view
to determining the factors influencing labour productivity
in this industry. It was also desired to ascertain the
extent of certain problems which have prevented the
technical development of this industry from following a
course similar to that of more advanced textile industries. Moreover, it was believed that an analysis of the
results of the investigation could give rise to recommendations for the improvement of conditions in the mills.
2. Limitations, both of time and of material resources,
have made it necessary to restrict the study to five countries—
Brazil, Chile, Ecuador, Mexico and Peru—the
textile industries of which offer substantial contrasts in
size, modernity, markets and geographical location. The
inclusion of other industries of considerable interest,
such as those of Argentina and Colombia, would have
improved the sample and added to the variety of aspects
which could be focused in this sector of Latin America's
economy.
3. The measurement of labour productivity and the
analysis of the factors which affect it were chosen rather
than any other study of the industry, because it is believed
that the level of productivity is the best indication of
internal operating conditions in the mills.
In contrast to
cost of production, though closely related with it, productivity is universal in character, not being affected by
differences in the prices of raw materials, machinery and
labour, nor distorted by the rates of foreign exchange and
of interest on investment. Moreover, it can be easily
measured, in a direct manner, in the mills and can be
published without fear of revealing data which the manufacturers might consider confidential.
4. The whole task was carried out by members of the
Commission's staff, assisted by technicians in the countries visited. In order to establish norms for comparison,
that is, standard productivities, the services of a firm of
American consulting engineers1 were enlisted, in addition to the co-operation of several Latin-American
manufacturers.
5. We do not presume to have attained the high degree
of perfection desirable in this type of research, either in
1 Lockwood Greene Engineers Inc., of New York.
2 Chapter VII contains a more detailed description of the
methods employed.
3 The English system of yarn count was used for yarn specifications ; this is the number of groups of 840 yards contained in
one pound of yarn. For fabrics, the relation selected was that
the methodology or in the scope of the work itself, but it
is hoped that the importance of the results will stimulate
the systematic adoption of measurements of this nature
and the development of better methods of analysis.
I I . SUMMAR Y OF TH E METHODOLOGY ADOPTED2
6. The study of each country was made by visiting a
number of mills which differed in size, degree of modernity, characteristics of production, and geographical
location. These units were selected with a view to obtaining sample groups, the composition of which would be as
representative as possible of the country's industry,
within the limitations imposed by the time available for
the investigation.
7. In each of these mills, labour productivity was
measured, process by process, for the greatest possible
number of products. The final results in kilogrammes per
man-hour were obtained by mathematically combining
the measurements made in the processes. These were
expressed separately, for spinning and weaving, by
means of curves, the ordinates of which represent the
productivity and the abscissae correspond to the specifications of the product.3
8. Furthermore, data were obtained regarding all the
factors which might affect productivity, such as the distribution of labour, the quality of cotton, the condition
of the machinery, humidity, the methods of control
adopted, the organization of the processes, the speed of
the equipment, the stoppage frequency for machinery, or
breakage frequency of yarn, and the quality of the intermediary and final products.
9. In order to express the total productivity of a
country's industry, average curves were plotted from
individual mill data, separating the results obtained in
modern industry from those of the older mills.4
10. In the subsequent analyses and comparisons, it was
decided to use units of labour consumption (man-hours
per kilogramme)5 instead of their reciprocals, the units
of productivity (kilogrammes per man-hour), because
the former are more easily handled and can be added
together directly.
11. The problem of comparing labour consumption
for the different types of products, that is to say, both for
between the density of the fabric, in picks plus warp ends per
unit of area and the weight of the cloth per unit of area. In practice the formula shown below was used, that is, the method of
assessment employed by the Mexican customs authorities in
evaluating fabrics. The result has been named "the count of the
fabric".
warp ends
sq. inch + picks
sq. inch -j X width of the fabric in inches X yards
lb.
84
' In the chapters dealing with individual countries, there are
separate graphs for each mill, together with average productivity
curves. The criteria adopted for the classification of mills as
either old or modern is explained in chapter VII..
5 In practice, man-hours per 100 kilogrammes of yarn or fabric
were used.
1
2 Productivity of the Cotton Textile Industry in Latin America
fine and coarse yarns, and for fine and ordinary fabrics,
was solved by determining the ratio between actual average consumption and the consumption which should be
obtained for each product, if the mill (or group of mills)
manufacturing these goods were operating under optimum conditions, with modern equipment, installations of
an adequate size and good management. These relationships, which are comparable for all types of products,
express the influence of all factors which affect productivity and may be considered as indices of the importance
of all these causes together. The analysis was carried out
in different stages, the first of which consisted in separating from the total indices the influences of type of
equipment—of the employment of old instead of modern
machinery—and of size, that is, of the use of installations which are smaller than optimum size, to which
further reference will be made. The remaining influence
was designated by the term "influence of operation", or
the index of importance of the operating factors, since it
expresses the manner in which a mill of a certain type
and size is operated.
12. Since this stage of the analysis calls for the establishment of standard productivities (or standard labour
consumptions), a group of 144 hypothetical standard
mills was set up, including both old and modern mills,
the size of which varied from 2,000 spindles or 40 looms
to 50,000 spindles or 1,000 looms.6 They were assumed
to produce six counts of yarn and six types of fabric,
which would cover practically the whole range of products found in the course of the investigation. According
to their size and type of equipment, these standard mills
were given the best organization possible, both from the
process and labour standpoints, and they were assumed
to be operating with the maximum efficiency attainable,
without impairing the quality of the product. The computed productivity of these standard mills was also
expressed by curves, in order to make it generally comparable to any results obtained in the actual mills.
13. Having obtained a series of theoretical productivity curves for different sizes of mills, it was easy to
determine the influence exercised by size on productivity
(or labour consumption). This information, allied to
considerations of an administrative order,7 was the basis
for establishing the limits within which the optimum size
of Latin-American units is probably found: that is, between 25,000 and 50,000 spindles for spinning mills and
500 to 1,000 looms for the weaving mills. The lower
limits of optimum size were adopted as a basis for all
comparisons because, from there upwards, productivity
increases are very slight.
14. In order to simplify the explanation of the procedure followed during the first stage of analysing the
factors which influence productivity, let it be assumed
that the investigation deals with a single old mill, producing only one type of product, or, in other words, that
only one measurement of labour consumption was taken.
15. The influence of equipment, that is, of the use of
old instead of modern machinery, is defined as the relationship between labour consumption in an old standard
mill and in a modern standard mill, both of which are of
the same size as the actual mill and manufacture the same
8 The basic data and the productivity of these mills will be
found in the annex attached to the report.
' Chapter VII contains a discussion on the optimum size of
mills.
product. It therefore expresses the influence of equipment independently of the effect of other factors.
1
6. The influence of small size in the actual mill is
calculated as the relationship between labour consumption in an old standard mill of the same size as the actual
mill and the consumption of labour in an old standard
mill of optimum size (25,000 spindles or 500 looms),
both of which are manufacturing the same product as the
actual unit.
17. The influence, or the index of importance, of operation is the relationship between labour consumption in
the actual mill and labour consumption in a standard mill
of the same size as that of the actual unit, both of which
are producing the same article and have the same type
of machinery installed. The result expresses the effect of
all the factors which influence productivity, with the
exception of the type of equipment and the small size of
the mill.
18. The product of these three relationships is the
influence, or the index of importance, of all the factors
contributing to low productivity and may be obtained
directly, as the relationship between labour consumption
in the actual old mill and that in a modern standard mill
of optimum size.
19. The influences or indices of importance for a
whole industrial sector in a given country were calculated as weighted averages of the influences corresponding to the productivity observations in all the mills comprising the aforesaid sector, each of these observations
being dealt with in the manner described above.8
20. In the case of sectors which are entirely modern,
it was evidently not necessary to calculate the effect of the
type of equipment, and in determining the other two
influences, exclusive use was made of the standard productivities for modern mills.
21. The second phase of the analysis was the breaking
down of the influences of type of equipment and of operation into their principal components. The influence of the
type of equipment was divided into two parts, namely,
the one caused by the differences in the output per machine unit (spindle or loom) for old and modern equipment ; and the other arising from the difference between
the minimum labour requirements of each type of machinery. This subdivision was solely based on the comparison of outputs and labour consumptions in the standard mills. The influence of operation was first divided
into two parts, the one corresponding to the abnormality
of the output of the machinery, and the other to the
excess of workers employed. As will be shown below,
the analysis was developed in greater detail, with a view
to determining the factors which contribute to such
abnormalities.
22. The analyses of the influence of operation could
not have been made for all the mills visited, during the
time available for the investigation. Therefore, within
the general sample for each country, a smaller sample
was selected consisting of a limited number of mills in
which the speeds, the unit weights of the intermediary
products, the efficiency of the processes, and the amount
8 The details contained in chapter VII show that the observations were not always studied individually in the manner indicated, since this would have involved an elaborate procedure.
The methods employed, however, gave equivalent results.
Chapter I. Productivity in the ; Group of Countries Visited 3
of direct, indirect and miscellaneous labour, were compared with the corresponding values of these elements
in the standard mills of the same size and type of equipment. Adopting the same procedure employed for the
influence of type of equipment and of size, these comparisons were used for determining the influence of each
of the aforesaid elements, that is to say, the labour consumption which their abnormalities would imply if they
prevailed independently of other factors. Labour consumption was expressed as its relationship to standard
labour consumption, that is, the labour which would be
applied in the absence of any abnormality.9 The values
of these influences were controlled by the fact that their
product should equal the value of the influence of operation in the mill studied, this value having already been
ascertained separately by the procedure outlined above.
23. The results of the analysis in the sub-sample mills
of a given industrial sector were averaged and extended
to the general sample for that sector and increased or
decreased by the procedure of logarithmic proportions,
in order that their product should equal the average influence of operation, which had already been obtained
from the data of the general sample.
24. The third and last phase of the analysis of the
influence of operation consisted in assessing, by observations and estimates, and with the help of spot checks—
such as the frequency of machine stoppage and of yarn
breakage—the importance of the factors which cause
the abnormalities of speed, efficiency, unit weight, and
staffing in the mills included in the sub-sample.
25. The value of the influences, that is, of the indices
of importance of the factors which affect productivity,
are expressed in every chapter of this report as a simple
ratio, with the exception of the first chapter, in which
these ratios were multiplied by 100, in order to present
them in the usual form adopted for indices.
III . GENERAL CHARACTERISTICS OF TH E INDUSTRY
26. The cotton textile industry of the countries visited
comprises a total of 786 mills, in which there are
4,602,411 spindles and 147,049 looms, constituting approximately 85 per cent of Latin-American cotton spinning and weaving capacity. Production is almost exclusively limited to articles for domestic consumption in the
individual countries in which, with the exception of
Ecuador,10 present-day demand is fully met, at least for
popular types of fabrics. The industry is featured by the
integration of the spinning and weaving processes, except in the very small units.
27. Brazil's cotton industry is the largest in Latin
America. There are 455 mills, with 3,279,677 spindles
and 100,146 looms. There are mills in the Federal District and in eighteen of the twenty states, but manufacturing is principally concentrated in the State of Sao
Paulo (38.2 per cent of the spindles), Rio de Janeiro-
* The influences, or indices of importance, of abnormalities in
the unit weight of intermediary products (draft schedule), of
speed, of the efficiency of the processes and of direct labour, are
equal to the ratio between standard and actual values of the unit
weight of the intermediary product, of the speed of the machinery, of the efficiency of the processes and of the units of machinery assigned to each tender, respectively. The influence of
i indirect labour is the relationship between standard and actual
ratios of direct labour to direct-plus-indirect labour. The influence of miscellaneous labour is the relationship between standard
and actual ratios of total labour, excluding miscellaneous labour,
Federal District (27.3 per cent), the North-eastern
States (23.6 per cent), and in Minas Gerais (12.2 per
cent). The largest installations in Latin America are
encountered here,11 but like the majority of industries
in this part of the world, the textile industry has a large
proportion of old machinery (91 per cent of the spindles
and 95 per cent of the looms), the period of greatest
expansion having occurred at the beginning of the twentieth century. It was only recently that any attempt was
made to modernize antiquated installations or establish
entirely new mills. Sao Paulo, however, is notable for
the modernity of its equipment (approximately 15 per
cent of the machinery is new), in comparison with the
rest of the country. This is largely because it developed
at a later date than other regions; moreover, the general
atmosphere of intensive progress in this state has proved
an incentive to improvement in all industries. For the
purposes of this investigation the modern mills of Sao
Paulo have been selected to represent the modern sector
of the Brazilian textile industry, whereas the old mills
in the Rio de Janeiro-Federal District region are considered as typical of the older sector of the industry.
28. Sao Paulo supplies the larger part of the raw material consumed by the Brazilian cotton industry. After
domestic consumption has been met, there is still a surplus for export.
29. The modernity of equipment is a striking feature
of Chile's cotton industry (19 mills,12 with 173,534 spindles and 5,012 looms). Seventy-seven per cent of the
spindles and 72 per cent of the looms are entirely new.
This provides a sharp contrast with the rest of LatinAmerican industry and may be explained by the country's delay in developing a textile industry.13
Though a
cotton factory had already been built in Chile in 1867,
72 per cent of the country's present spinning capacity
and 47 per cent of its weaving capacity have been installed since 1938. Its vigorous development in ten years
has enabled domestic production to replace imports almost entirely, though in 1938 the latter supplied 60 per
cent of total consumption.
30. In the early period, the development of the industry was very favoured by exchange controls (since
1932), and by the devaluation of Chilean currency,
which caused a sharp difference between foreign and
domestic prices. Later, it was further encouraged by the
Second World War, which not only eliminated foreign
competition for a period of several years, but also led to
an accumulation of capital that was invested or reinvested in industry. Prevailing conditions are favourable
to the mills, owing to the restriction of imports, the
shortage of foreign exchange, and the fact that total
domestic demand has probably not yet been met.
Seventy-nine per cent of the industry is concentrated
in Santiago and the rest is located in the vicinity of
Concepción (12 per cent) and Valparaiso (9 per cent).
At present, all of the cotton is imported, but attempts
are being made to grow the raw material in Chile.
