The title is a play on Timothy Garton Ash’s 2009 book, Facts are Subversive. I want to suggest how statistics are used in the public discourse on the Canadian economy often to distract, either by mistake or on purpose, from underlying economic changes.
Assume that there are three sectors (there are others) of the Canadian economy, agriculture, manufacturing and services, which is roughly how they are measured in terms of share of GDP and employment. Since 1900, the overall economy has grown as have all sectors, although not at the same rate so that their relative shares have changed.
Over time the share of agriculture in GDP and employment has fallen dramatically, although agricultural output has increased. For manufacturing, the share of GDP and of employment increased, but now both overall manufacturing’s share of employment is declining. Good paying jobs are being lost and some people are now working for less money in what were thought of as good paying jobs. This leads to the cry that Canadian workers are becoming hamburger flippers and low paid/unskilled retail employees. (In Ottawa there continue to be help wanted signs in retail outlets, which is curious when people claim to be unemployed in the same area.)
At the same time, there has been a large increase in the share of employment in services, and services output has increased. Some of these are low paying jobs and some are well paid skilled people in construction, the building trades and resource industries, computer programming, professionals like doctors, nurses, engineers and so on. Another statistic refers to the number of temporary foreign workers in Canada, about 300,000 at any one time, doing jobs that unemployed Canadians are either unwilling, or are untrained to do.
Reference to any one of these statistics provides a partial and often distorted view of the economy. Together, they start to put together a story of how the Canadian economy has and is changing. What do they suggest?
Share of labour force by sector and share of GDP (StatCan):
% share of Employment % share of GDP
Year 1950 1990 2012 2012
Agriculture 16 3 2 2
Manufacturing 25 66 13 13
Services 32 71 78 70
In employment terms, agriculture accounts for a much lower share of total employment today than it did 60 years ago. Manufacturing’s share rose from 1950 and has since declined to about 13%, while the share of services shows a rising trend and now rests at 78% of total employment. In GDP terms in 2012, services share is at 70%, manufacturing 13% and agriculture 2%.
Less need for another udder
A remarkable statistic shows that in 2012 the average American cow produced 22,000 lbs of milk each year compared with 5,300 lbs in 1950, an annual increase of 5% for each of the past 60 years. I imagine a Canadian cow would be equally productive, although supply management may require it to work shorter days and take longer vacations. With 1950 model cows, today’s 33 million Canadians could be supplied with a pound of milk per day by about 530 cows. Sixty years ago it would have required over two million cows. A 2012 model cow represents an enormous increase in productivity and a saving for the environment, especially air quality.
In the agricultural sector there has been a marked increased in productivity with the introduction, among other things, of mechanization, together with animal and crop husbandry techniques. A striking contrast is the practice of Doukhobor women providing the women-power for plowing in western Canada around 1900, versus a present-day piece of farm machinery – see photos at Canadian West website, http://www.collectionscanada.gc.ca/canadian-west/052920/05292069_e.html
Fewer farmers now produce more with less, but in the 1920s, farmers worried about where their children would find work. This was soon forgotten as manufacturing grew to produce among other things the tractors, machinery, fertilizers and seeds which allowed for increased productivity. Also refrigeration and canning meant food could be stored and traded internationally. (Bananas from Central America were frozen while being shipped and then gassed to restart the ripening process. This happens today and benefits consumers with imported products.)
As the agricultural sector grew in terms of output, it released labour for employment in the manufacturing and services sectors. Both output and employment grew. In manufacturing the share of employment is now declining especially with the loss of low paying jobs, similar in many ways to what happened previously in agriculture. The reason is a mix of competition from imported manufactured products, and the substitution of capital for labour, for instance the use of robotics on production lines. Robots substitute for labour but require inputs in order to be produced. They will tend to replace less skilled labour but in turn will require inputs to be produced and maintained.
For automobiles, the stages of production or supply chain of inputs, leading to the final product, is divided between countries, between regions of a country and between manufacturing and service sector jobs. Robots need to be manufactured, but their operation requires computer programmers and persons who will maintain the robots which are often service sector jobs.
These changes are always taking place in a dynamic economy. If they did not we would be stuck with Doukhobor plowing technology in agriculture, and a Chaplinesque Modern Times version of production lines.
Finally, consider the service sector, where the share of employment is over 70% and growing. Some see this is as a problem and a reflection of fast food, online shopping, and call centre employment requiring less skills and liable to be outsourced abroad to low wage labour in developing countries. (Some of it is outsourced and when managed incompetently, as in the case of RBC (April 2013), gives all business a deserved black eye.)
A myriad of tasks are undertaken by service sector employees. They include highly qualified professionals who have undertaken years of training and are required to receive ongoing retraining. Doctors and dentists employ assistants who may receive lengthy training. For example a dental assistant, who does more than dental hygiene, now requires to complete a three year course. For nurses it is even longer. Developments in information technology have led to jobs for programmers at different levels of expertise. And so it goes on.
What is happening with the job-type employment numbers is that the service sector as a whole is too broad a category for useful analysis. What is needed is to disaggregate the type of service occupations, for which the data exist in the Labour Force Survey, and examine how and why the demand for different types of occupations are changing. Statistics Canada addresses some of these issues in http://www.statcan.gc.ca/pub/75-001-x/2008109/article/10694-eng.htm
The contemporary cry is that the manufacturing sector is shrinking in terms of its share of output and employment, with “good paying” jobs being lost. At the same time service sector employment is growing, but these are often lower paying jobs. Forgotten is that labour productivity in manufacturing is growing because of often well paid jobs in other sectors such as services.
The fact that Canada today appears to be more a service oriented economy, at least in terms of employment, is not a bad news story, as there are well paid service sector jobs in developed economies, as well as hamburger flippers shopfloor assistants. (The latter is not as glamorous an occupation as portrayed in the TV series Selfridges, where it appears in the 1920s the assistants might work more than eight hours, as a result of having after hours work.)
The moral of this story is not to be blinded by repetition in the public discourse of one statistic, in this case the loss of employment in the manufacturing sector. Job opportunities, some well and some less well paid, in all sectors change continuously. The fact that Canada, like other developed countries, is increasingly a service economy reflects changes in production processes and arrangements in all sectors.
Adjustment is never easy especially for older and less educated workers, and is not instantaneous. Policies which can alleviate the adjustment process relate to education, retraining, use of foreign workers (temporary and permanent), job sharing, outsourcing, but not subsidies, tax incentives and protectionism, which allow existing less efficient manufacturers to remain in business….until the time comes when they will be forced out. Technology, due mainly to the internet, has undermined Canadian cultural protectionism, and something will eventually conquer supply management in agriculture…..although perhaps not in my lifetime.