There are no ‘good’ or ‘bad’ jobs

Posted on November 24, 2015 in Debates – Full Comment
November 23, 2015.   Stephen Gordon

The service sector has a bad reputation. Maybe it’s the name: “service” is derived from the Latin word for “slave,” and a popular impression of a service-sector job is one that involves doing the menial tasks expected of servants. When futurists say that service-sector jobs are the future of the labour market, a popular response is a mixture of dread and resignation.

There is something odd about this reaction: virtually all of our high-paying and high-prestige jobs are in the service sector — doctors, lawyers, teachers, firefighters and so forth. The ranks of the top 1 per cent of high income earners are almost entirely composed of people working in the service sector, and the service sector absorbs the vast majority of university graduates. The well-educated people who do find well-paid employment outside the service sector are invariably those who are in the business of providing services: financial planning, legal advice, engineering design, etc.

Service-sector jobs are not only the future of the Canadian labour market, they’re its present: 80 per cent of workers are employed in the service sector, up from 50 per cent in 1961 and 30 per cent in 1911. Even though the increase in service-sector employment has been accompanied by significant increases in living standard, many regret this trend: the “good” jobs are disappearing. But what does it mean to say that one job is better than another?

The history of economic thought has a long — albeit misguided — tradition of establishing hierarchies for different sectors of the economy. One of the better-known examples what the Physiocrat school developed in 18th-century France. Physiocracy posited that all value was derived from agriculture alone: other sectors simply recycled that which had been created by farmers, without creating additional value. Adam Smith was greatly influenced by the Physiocrats and many of their ideas appear in The Wealth of Nations, but he disagreed with the claim that only one sector could create value.

Another tradition — whose origins were in Canada, of all places — places the manufacturing sector at the centre of economic development. The “Staples Thesis” was never a fully fleshed-out theory; the initial framework proposed by the University of Toronto economist Harold Innis was largely descriptive. But it did give rise to a staples approach that was popular in mid-20th century Canadian economics departments. A recurring theme of this work was the role of the manufacturing sector in economic development. One outcome of this research was the “staples trap” conjecture: an economy with a large resource sector would be deprived of the opportunities for growth that only a large manufacturing sector can provide.

Neither theory gets much traction in current economic thought, but their legacy lives on in public opinion. Much of the complex web of agricultural price supports, income supports and other supports — not least of which is successive Canadian governments continuing to act as enforcers for the dairy cartel — is based on the idea that farming is special, and deserves special treatment.

The same goes for the web of subsidies and special tax treatments for the manufacturing sector. A frequently-made argument is that manufacturing jobs “add value,” leaving the rest of us workers to wonder just what it is we do all day. This extraordinarily self-serving claim is surprisingly widely-accepted as a justification for government support.

These measures have largely been in vain. Technical progress is the proximate cause for the decline in employment outside the service sector — both in the share of the total and in absolute numbers. While new technologies can replace certain service-sector jobs (stenographers are a good example), automation has had a much greater impact in the goods sector. Canada’s manufacturing sector is producing more than it ever has; it just needs fewer workers to do so.

Being able to produce more goods with fewer workers should be an unalloyed benefit — and in the long run, it is — but the short-term costs can be non-trivial. For example, the MIT economist David Autour has found evidence that by eliminating jobs that are structured around routine tasks, automation has “hollowed out” the wage distribution in the United States, leading to a more polarized workforce and the accompanying potential for social tensions. (This pattern appears to be less evident in Canadian data.)

It’s not clear to me, however, that automation is destroying “good” jobs: a job that consists of routine tasks that a machine could do may not be worth saving, regardless of what sector it’s in. Of course, few goods-sector workers made redundant by automation can swiftly transition to a well-paid job in the service sector, and these transition issues should be taken seriously. Even so, it’s not helpful to think of the shift from employment in the goods sector to employment in the service sector as an exchange of good jobs for bad ones.

There are even reasons to be optimistic about the ongoing trend to services. While there are physical limits to the quantity of goods that can be produced, there are no obvious limits to our ability to find new ways to enjoy them. Creating value is more important than producing stuff.

National Post

Stephen Gordon is a professor of economics at Laval University.

< >

Tags: , , , ,

This entry was posted on Tuesday, November 24th, 2015 at 11:49 am and is filed under Debates. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

Leave a Reply