Comparison of Establishment Groups in the Manufacturing Sector 1985-1996

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

31F0027MIE96001

Rowena Orok

The last eleven years has brought major challenges to the manufacturing sector. Trade liberalization resulting from GATT rounds of tariff reductions, the US-Canada free trade pact, and the North American Free Trade Pact have all influenced changes in the competitiveness of manufacturing industries. The recession has placed additional demands on industries to rationalize production costs and adjust their resource requirements. Finally, increasing globalization pressures from emerging export-oriented economies have accelerated the pace of structural change for all manufacturing industries.

Did the wave of change have the same effect on all manufacturing establishments regardless of size? This paper analyzes manufacturing establishments in terms of four major areas from 1985 to 1996: manufacturing activity gross output, production cost structure, productivity, and employment structure.

In 1996, the new establishment grouping for each industry in the manufacturing sector was derived from a ranking of establishments from highest to lowest based on the volume of manufacturing shipments. The ranked establishments in each industry were then split into quartiles - largest, upper mid-size, lower mid-size and smallest.

The first part focuses on manufacturing activity gross output. It compares the level and distribution of real gross output in each quartile from 1985 to 1996. The second part analyzes quartile production cost as well as various components such as raw materials, salaries and wages, and fuel and electricity. The third part measures productivity over time by relating real gross outputs to labour inputs. The final part looks at structural changes in quartile employment levels.

The distribution of gross output by quartile for the whole manufacturing sector has remained relatively stable throughout the period, with over 90% accounted for by large establishments (See Chart 1. Distribution of Real Gross Output, by Quartile, Selected Years). At the other extreme, the smallest establishments made up less than 1% of total manufacturing gross output. A closer look at each quartile's output levels (adjusted for inflation) throughout the 11-year period showed that shifts in the business cycle had a greater impact on all lower quartiles.

All establishment groups saw their outputs soar from 1985 to 1989 (See Chart 2. Index of Real Gross Output, by Quartile, Manufacturing Sector, 1985-1996 (1986=100). During this period, all the lower quartiles experienced higher increases in real gross output compared with the largest quartile which only grew by 3%. The lower mid-size and the smallest quartiles gained the most, each expanding their outputs by 10%, followed by the upper mid-size quartile (7%).

Various factors such as the recession, high interest rates, and high Canadian dollar caused deteriorating market conditions in the early 1990s which, in turn, forced all establishment groups to cut back on their outputs. Large and medium establishments had to decrease their outputs by about 5% between 1990 and 1991. The impact, however, was more severe on the smallest quartile, as it reduced its output by over 10%, from $1.49 billion in 1990 to $1.33 billion in 1991.

Large establishments grew more quickly during the period following the recession. From 1992 to 1996, this quartile bounced back faster than any other group, expanding its output by 5%. By 1994, it already exceeded its 1989 output level. In contrast, output levels in the other quartiles were still below their pre-recession levels in 1989.

Total production cost is composed of expenditures on raw materials and supplies, fuel and electricity, and salaries and wages (1). As in the case of real gross output, there were only small changes in the distribution of total manufacturing sector production cost, with all lower quartiles accounting for a combined share of 10% or less of total production cost from 1985 to 1996.

In terms of the major inputs, salaries and wages were the only inputs that showed a different distribution pattern where the largest quartile's shares were consistently below 90%. Raw materials and fuel and electricity inputs, on the other hand, followed the same distribution pattern as that of total production cost (over 90%).

The relatively stable quartile shares at the aggregate level, as discussed above, tend to downplay the occurring trends in each quartile's input composition throughout the period. Thus, the next two sub-sections focus on inputs as a percentage of total quartile production cost.

Raw materials and supplies made up a higher proportion of total quartile production cost, followed by salaries and wages. However, the actual distribution of these inputs varied by quartile. The share of raw materials to quartile production cost increased with establishment size whereas the reverse could be said for salaries and wages. Chart 3. Composition of Quartile Production Cost, 1996 shows that in 1996, for example, raw materials accounted for 77% of total production cost of the largest quartile whereas the same inputs represented only 48% of the smallest quartile's total production cost. In contrast, salaries and wages made up a relatively lower proportion of the largest quartile's total production cost (20%), compared with the smallest quartile's allocation (49%).

The shares of both inputs for all lower quartiles moved together over the period, with an increasing trend in the shares of salaries and wages. Large establishments experienced only slight changes although there was a more visible downward trend in the share of salaries and wages and a corresponding increase in raw materials share after 1991. These trends support the rationale of capital intensity for large establishments and labour intensity for small establishments.

High capacity plants that produce large economies of scale had a negative impact on the largest quartile's labour requirements especially during the period of downsizing and cost reductions. Generally, as large establishments adopted new technology to improve plant efficiency, the number of workers fell. Production workers were displaced by new machinery and equipment as large plants became more automated. In contrast, small and medium establishments did not always have the relative ease of labour substitution. The options available to these establishments were more limited, as new technology was not easily adaptable to smaller plants. Smaller establishments then found it more feasible to hire additional labour inputs that met their cost requirements rather than to undertake huge capital investments.

Fuel and electricity shares remained relatively low for the past 11 years. There were parallel movements in the shares of fuel and electricity for both mid-size establishments, ranging from 2% to 3% between 1985 and 1996. In comparison, the largest and the smallest quartiles had relatively higher allocations for fuel and electricity, 3% to 4%. In general, all quartiles experienced a slight drop in their allocations for fuel and electricity during the period. Increased use of energy efficient equipment as well as plant modernization facilitated the reduction in energy use as in the case of energy-intensive industries such as Pulp and Paper and Cement Manufacturing.

For the past 11 years, how did manufacturing establishments combine various resources (inputs) in order to produce their products (outputs)? Productivity measures are often used to gauge the efficiency of production operations. Various productivity measures have been developed throughout the literature. Some measures take into account all resources used in the production process (total factor/multi-factor productivity measures) while others focus only on a subset of inputs used (partial productivity measures). There are also a range of variables that could be used such as outputs, shipments, value added, wages, number of employees, person-hours used, etc. In this paper, the productivity measure used was real output per worker which takes a look at real gross output as a percentage of labour inputs (2).

Real output per worker measures the contribution of each employee to the total output of the establishment, after taking into account inflationary effects. (See Chart 4. Real Output per Worker, by Quartile, Manufacturing Sector, Selected Years). Real output per worker in the largest quartile has been above $150,000 since 1985. This ratio was at a much lower level and has remained relatively stable throughout the period for the rest of the establishment groups. Large establishments generated more output by improving plant efficiency. The rise in output, accompanied by a decrease in labour requirements translated to a 3% average annual growth in the largest quartile's productivity from 1985 to 1996.

Implementing the use of new technology was not always a possible option for smaller establishments. The amount of capital investment and the existing constraints of smaller plants generally made it more difficult for mid-size and smallest quartiles to use the technology route in generating more output. Instead, they substituted capital with relatively cheaper labour. Thus, productivity growth was slower throughout the period and for the later years, has been declining for the lower mid-size and smallest establishments.

Which types of manufacturing establishments generated more jobs? Did the employment requirements of large establishments evolve in the same manner as small establishments? A closer look at the shifts in employment levels across quartiles give an indication of how establishments of different sizes have adjusted their labour requirements for the last 11 years.

From 1985 to 1996, the average annual rates of growth (or decline) in employment levels for all quartiles were less than 1%. Sub-period growth rates shown in Chart 5. Average Annual Growth Rates in Total Employees, by Quartile, 1985-1996 indicate that from 1985 to 1989, both mid-size quartiles experienced more significant increases (over 7%) in total employees compared to the smaller increases in the largest quartile (2%) and smallest quartile (3%). Downsizing and the overall effects of the recession brought a reduction in labour inputs for all establishment groups. From 1990 to 1991, the smallest quartile was forced to cut back 17% of its labour requirements, more than double the declines which were undergone by all other quartiles. All quartiles maintained these lower levels of labour inputs even after the recession from 1992 to 1996. The largest and smallest quartiles started rehiring and slowly increased their total employees at approximately the same rates. Both mid-size quartiles, on the other hand, were still downsizing and posting negative growth rates in total employees during this period.

The calculation of quartile margins based on value of shipments highlights the relevance of the results presented in this paper. The threshold values allow firms in various manufacturing industries to have a benchmark for performance comparison. Table 1. Quartile Boundaries based on Value of Shipments, Manufacturing Sector, Selected Years shows that in 1996, firms with total value of shipments of $4.8 million or more would belong to the largest quartile.

In this case, Gamma Limited would find out that in 1996, it allocated a relatively higher-than-average proportion of its production cost toward the purchase of raw materials. It could also deduce that the lower-than-average share spent on salaries and wages contributed, in part, to higher real output per worker during that year. Finally, Gamma Limited would be aware that it allocated about the same proportion of expenditures toward energy costs.

In conclusion, establishment structure determined each quartile's response to changing market demands. The observed trends in the four major areas were consistent with the inherent characteristics that defined the different quartiles. However, an analysis of structural changes using other defining variables for each quartile would give a more comprehensive evaluation of the four establishment groups. These would include time series analysis of variables such as amount of person-hours paid, hourly wage rates, other productivity measures and concentration ratios.

This article was written by Rowena Orok. Rowena is a Statistics Canada economist and is formerly of the Manufacturing, Construction and Energy Division.

Further information on Canadian manufacturing can be found in the publication, Manufacturing Industries of Canada: National and Provincial Areas (31-203-XPB). This publication is available annually for $68 per issue in Canada and U.S. $68 outside Canada. Order this publication by telephone: 1-800-267-6677, by fax: 1-800-889-9734, or by Internet. For more information about manufacturing data or time-series call the Disclosure and dissemination Unit, Manufacturing, Construction and Energy Division at (613) 951-9497 or by Internet: manufact@statcan.gc.ca.

Archived Content

31F0027MIE96001

Comparison of Establishment Groups in the Manufacturing Sector 1985-1996

Manufacturing, construction and energy division

Rowena Orok