Intro to productivity (that thing Canadians are apparently so bad at)

Presenting: the Econowatch special report on productivity


If you spend any time reading about the Canadian economy, you have inevitably come across the Great Canadian Productivity Puzzle. Canada’s productivity is much lower than that of other countries, and we don’t really know why. Neither do we seem to be able to fix the problem. Policymakers have used every trick in the book to try to boost productivity, but the results have disappointed. Productivity growth matters because it drives up our purchasing power: if it lags, so will our standard of living. And yet—here’s where things get interesting—Canadians are far better off than one would tell looking at our dismal productivity performance over the past 20 years. How did we do it? In this six-part special report, Maclean’s in-house economist Stephen Gordon investigates the mystery. (With a contribution from Econowatch editor Erica Alini.)

Click here to see what’s coming up next.

Here’s what economists mean by productivity—and why you should care

The goal of boosting productivity is to increase households’ purchasing power. In a closed economy—one that does not engage in international trade—the only way to increase domestic income is to increase domestic production. Consumers cannot buy something that hasn’t been made. (The link between income and output is more nuanced in an open economy, but I’ll get to that in a subsequent post.)

Now suppose the population of said economy is growing: The only way for people to become better off is to find a way to increase production by a rate greater than population growth. In other words, per capita incomes can grow only if productivity grows.

There are different types of productivity. There’s labour productivity, which is output divided by the total number of hours worked—that is, the sum of the hours worked by all employees, which gives us the amount produced in 60 minutes by a hypothetical average employee. There’s capital productivity, which is output divided by the amount of capital held (machines, office buildings, infrastructure). And then there is multifactor productivity, a trickier but very important statistic, which tries to capture increases in output that cannot be explained by increases in either labour or capital. The simplest way to measure and understand productivity is labour productivity, so, for now, let’s concentrate on that.

Sarah Conard/Reuters

A couple of myths worth busting

The emphasis on productivity has been known to lead to some confusion. For example, some may view a productivity growth agenda as a thinly-disguised plot to oblige employees to work harder and for longer hours. But improving productivity is not about forcing a construction worker to dig faster and to work through his lunch break, it’s about providing him with a mechanical excavator and the skills to use it. Productivity growth is about being able to create more value with the same amount of effort.

Another popular misconception is that increased productivity means higher unemployment. If the same amount of output can be produced by fewer people, then what happens to those excess workers? This is the “lump of labour” argument, the notion that the quantity of work to be done is a fixed constant. It is also a well-known fallacy: higher productivity increases the demand for labour, because more productive workers are more valuable to employers. Although higher productivity in a given industry may reduce employment in that sector, the increase in total output and income across the economy will create new, better-paying, employment opportunities elsewhere.

How productivity growth drives up wages

From a firm’s perspective, a more productive worker is a more profitable one. For a given number of workers, increased productivity means increased output, sales and revenues. If wages and other costs of production don’t adjust, increased revenues are increased profits.

But the story doesn’t stop here. Since the profits generated by each worker have increased, firms will conclude that the way to keep boosting profits is to hire more workers. In turn, the competitive pressures generated by the increased demand for labour will bid up wages, to the point where the extra revenues generated by hiring an additional worker are completely offset by the higher wages that this worker can command.

Another channel by which greater productivity could lead to higher purchasing power is through prices. More productive workers produce more output, and this increase in supply may drive down the price of the good or service being produced. Either way, real wages—that is, wages divided by the price of goods and services—will rise. The adjustment can be an increase in the numerator (wages), a decrease in the denominator (prices), or some combination of the two.

(Wonkish detail: Labour productivity measures average productivity, but the mechanism described in the preceding paragraphs predicts that real wages should track marginal productivity—the extra output produced by an additional unit of labour. It turns out that although average and marginal productivity are different concepts, real wages track average labour productivity measures reasonably well. The reason why the Cobb-Douglas functional form is so popular in applied work is its property according to which marginal productivity is equal to average productivity.)

Canada vs. Uncle Sam

What ensures that wages respond to labour productivity in the way I described are market forces. In a competitive labour market, the increase in the demand for labour produces upward pressure on wages, and an increase in output supplied to a competitive goods market will drive down prices. That’s what you see in the U.S.:

Canada’s history of protectionism, on the other hand (in addition to restrictions on trade between provinces), has resulted in markets that are generally less competitive than those in the large U.S. market and real wages that do not track productivity growth as closely:

P.S. See the lag between productivity and wages in the upper-right corner of the U.S. chart? For the past few years, productivity rose but real wages didn’t quite follow. The fact that U.S. labour markets have been deeply depressed since the financial crisis likely goes a long way in explaining why.

The takeaways:

What do economists mean by productivity? There are many definitions, but all measures of productivity try to capture the ability to produce more output—which is the same thing as generating more income— with the same levels of inputs.

Why do economists think productivity is a big deal? Increases in productivity are inextricably linked to increases in incomes and wages. Real, per capita income growth can only be sustained by increases in productivity.



  • Part one: Intro to productivity (that thing Canadians are apparently so bad at)

Filed under:

Intro to productivity (that thing Canadians are apparently so bad at)

  1. Great column.

  2. I’m not disputing the conclusions drawn, but the axes of the charts don’t have the same scales. The U.S. chart shows 70+ years of data, with a range from 20 to 120 on the y-axis. The Canadian numbers only show 30 years, with a range from 70 to 110 on the y-axis.

    Shown to the same scale, the Canadian numbers wouldn’t seem to vary as much.

    • Point taken, but the deviations are still larger in Canada on average. The standard deviation of the log ratio is about 10% higher in Canada.

      • Which could also be attributed to a more resource based economy in Canada vs US.

        It’s like the beta of a stock. More risky, higher volatility of a stock has a higher beta (measured against fully diversified stock index).

  3. Hard for Canadians to be productive when companies opt to off shore/outsource work overseas to China, mexico to pump up share prices/dividends.

  4. “Although higher productivity in a given industry may reduce employment in that sector, the increase in total output and income across the economy WILL create new,
    better-paying, employment opportunities elsewhere.”

    Wrong. The increase in total output and income across the economy MAY create new, better-paying employment opportunities elsewhere. The so-called “lump-of-labour fallacy” is a rhetorical decoy (AKA “red herring” “straw man”) that distracts the reader from the analytical legerdemain that follows. Just because something CAN happen doesn’t mean it always WILL. Stephen Gordon has deflected objections to his claim by pre-emptively labeling any deviation as “fallacious,”

    Notice that the alleged “lump of labour” argument is a “popular misconception.”. Translation: Mr. Gordon cannot name anyone in particular who actually believes “that
    the quantity of work to be done is a fixed constant.” This is not surprising given that the bogus fallacy claim was first made 233 years ago and that those who promulgate it almost NEVER identify an actual person who holds the fallacious belief.

    The so-called “lump-of-labour fallacy” is, in short, the swindling economist’s version of the gold ring drop scam. “Oh look, My lucky day! I found this fallacy on the sidewalk! Did you drop it? No? It doesn’t fit me. Here, you take it. Look! it has a hallmark inside so it’s genuine!”

    • You’re not really making a point here, just venting. Many laypeople do use the lump of labour argument in opposition to things like automation or immigration.

      • Name them, then, if there are so many. BTW, it isn’t sufficient to say that people fear that unemployment will result from automation or immigration — such a view does not require the assumption of a fixed amount of work.

        • I don’t think providing an example would make any constructive difference, here. Google for newspaper articles on immigration, and check the comment section if you have honestly never seen such an argument made.

          • Of course you don’t think providing an example would make a constructive difference — because you don’t have an example!

            As Jonathan Chait advised a couple of weeks ago:”If you’re arguing against an idea, you need to accurately describe the people who hold them. If at all possible, link to them and quote their argument. This is a discipline that forces opinion writers to prove that they’re debating an idea somebody actually holds.”

            I have a database of over 500 instances of the fallacy claim being made without any supporting evidence of belief in the fallacious idea. An “idea” no one actually holds is a fiction, not a “fallacy.”

          • Here is the first result for “immigrants taking american jobs”:


            Now, we can iterate through every result this search turns up, and you can tell me how each one does not use the lump of labour fallacy. Am I accurately predicting how this would go?

          • Andrew, There is no necessary connection between “immigrants taking American jobs” and assuming there is “a fixed amount of work.” Think of it as a stock and a flow. If the outflow exceeds the inflow then the stock diminishes regardless of how many new jobs are created (inflow). But obviously if you have it fixed in your mind that “immigrants taking American jobs” = “assuming there is a fixed amount of work” there is nothing I can do to persuade you otherwise.

          • I disagree. What you are describing is not at all the same as the expression “immigrants taking American jobs”. The expression is quite clear: there are jobs: work to be done. This work is to be assigned to someone. Instead of being assigned to Americans, the work is assigned to immigrants (the word “taking” is more crude, but the idea is the same). This means that work is not created by the individuals doing it, it is assigned to workers by someone else. Anyone can understand what the expression means, and it does not mean what you’ve attempted to twist it into meaning.
            It is quite clear: there are “jobs” to be “taken”, and it’s immigrants doing most of the taking.

            I’m sorry, but you seem to be simply insisting that the exact phrase “fixed amount of work” is used, when there are much more common ways of expressing the concept.

          • Stephen Gordon uses the exact phrase, “the quantity of work to be done is a fixed constant.” It is, in fact. precisely that exactitude of the phrase that qualifies the alleged belief as a fallacy. The vague formulations (that people might actually believe) may be wrong empirically but they are not “fallacious.” In the 1890s John Rae gave about as authoritative a version of the fallacy claim as you’re likely to find. His summary statement was that it is “not a simple sum in arithmetic.” That’s true, but it cuts both ways.

          • Of course nobody believes there is an amount of work that does not change over time, which is what you seem to be insinuating by the meaning of “fixed”.

            Nobody believes that. Everybody knows if population of the US was 1 billion there would be more construction jobs, more hair cutting jobs, etc.

            When people say the amount of work is “fixed”, they mean that at a given moment in time, there is a certain amount of work to be done. When you let people in to take that work, the fixed amount of work is divided amongst more people. The additional jobs that may result will only come later. That’s what people mean when they say “fixed”.

            For work to be not fixed, it would mean that when immigrants arrive, they immediately do work that would not have been done otherwise, and they immediately create work that would not exist otherwise. When people say that immigrants are taking American jobs, they are indeed making the claim that this does not happen, so they are indeed claiming that the amount of work is fixed.

          • I’m not the one “insinuating” the absurd belief. Stephen Gordon bluntly states it. If that is not what he means, he should say what he means.

            It’s not really clear to me whether you’re saying that there IS a fixed amount of work at a given point in time or that immigrants DO immediately create work that would not exist otherwise or both.

      • 1873: “The whole system of Trade Unions is founded on the supposition that the amount of work to be done is a fixed quantity, and that it is the business of a trade organization to do all it can, in the first instance, to interpose artificial obstacles in the way of men coming into the trade…”

      • 1876: “The root of the mania which has had such a disastrous effect on the material prosperity of the country, and, above all, of the working classes, is the idea that the amount of work to be done is a fixed quantity, quite independent of any efforts which may be made to encourage and stimulate demand, and that, therefore, the best course is to spread it thin in order to make it go as far as possible.”

      • 1886: “Instead of seeing that work makes work; that busy masons make carpenters busy, and these in turn make busy tailors, shoemakers, grocers and the like, they seem to think there is a fixed amount of work and a fixed “wage fund” as some of our political economists call it, i. e., just so much work to do; just so much money to pay in wages, and therefore if numbers can be kept down, pay can be kept up…”

  5. Canada’s history of protectionism, on the other hand (in addition to restrictions on trade between provinces), has resulted in markets that are generally less competitive than those in the large U.S. market and real wages that do not track productivity growth as closely>

    Very good primer. But, this passage above caught my attention.

    Didn’t Mark Carney, in his swansong speech suggest that provinces, having different inter-provincial exchange rates, help stabilize the country’s economy? In other words, despite a high Can $, lower wages in say Quebec relative to Alberta kept them competitive to foreign manufacturers who might want to also supply the same goods.

    Is that in spite of the inter-provincial trade barriers noted above, then?

    • From Carney speech:

      Movements in provincial real exchange rates are another important part of the adjustment process. Although there is one exchange rate for Canada as a whole (we all use Canadian dollars), price differentials across the country yield different real provincial exchange rates.

      This matters.

      Consider the Alberta/Quebec real exchange rate (Chart 4). When higher energy prices stimulate production and investment in Alberta, extraction, construction and labour costs there rise. This increases the real Alberta exchange rate, making goods and services from the other provinces, including Quebec, more competitive. Interprovincial trade is boosted, spreading benefits from energy price increases throughout the economy.

      • Yes, but there are still myriad barriers to trade which results in reduced competition and efficiency.

  6. Thanks for writing this one. Well said.

      • Thanks, Thwim. Sandwichman is Tom Walker. ;-)

        • Ah.. that explains the database comment then. Still, unless you’re making a strictly logical argument, it’s better to cite peer-reviewed stuff than just posting it — even if it is your own. :)

  7. Re: That US Productivity/Wages chart

    Wages failed to rise with Productivity from the 70’s. The difference between those lines is Profit Extracted by the 1% into their gambling casino. The difference is that money is NOT going into education, infrastructure, health and peace but into class war and energy war and political dominance.

    • That chart uses CPI to calculate real wages; the graphs above use the GDP deflator. The relevant price for the firm is the output price, and using output prices to calculate real wages gives a series that tracks productivity.

      The distinction between producer and consumer prices is important, and I get to it later.

      • Why is that distinction not in your preceding paragraph to that chart? What’s “relevant” for wage-earners in a discussion about Wages and Productivity?

        You have a goal to show a “series that tracks productivity” and the timeline discrepancy between the two charts amplifies the preconception.

        • The difference is largely a reflection of greater wage inequality (“the share of all wages accounted for by the top 1 per cent of wage earners has nearly doubled, from 6.8 per cent in 1973 to 12.9 per cent in 2010” – Mishel), the rapidly growing cost of health insurance and differences between the CPI and the GDP deflator. Larry Mishel of the Economic Policy Institute discusses it here:

          When Stephen Gordon talks about productivity being “the only way for people to become better off” the naive assumption would be that he is referring to “most people” or the “average person” but the figures he uses to show that “wages track productivity growth” have nothing to do with the average person (median) becoming better off — they have to do with the wealthiest becoming better off and thereby raising the average (mean).

          It should go without saying that Darrell Huff discussed this particular “average income” swindle in a whole chapter, “The Well Chosen Average” of his classic, How to Lie with Statistics. Compensation is not “normally distributed” therefore the means and medians are far apart. No doubt Professor Gordon will claim a plausible “technical” rationale for his use of “nominal compensation divided by GDP deflator.” But such a rationale will still gloss over the “why you should care” part. Why should you care that the rich get even richer while the middle class stagnates?

          I’ve posted an expanded version of this comment at EconoSpeak:

  8. In canada I work in the tar sand in fort Mcmurray and we are far to be productive the mentality in Canada regards the safety is so obsess everybody is afraid of working because evwrytime we try to work and be productive your job is in problem.
    The management and analyst are not thinking right if you compare with other country where safety is number 2 but Canada completely oposite so the analyst should get in the field sometimes and get dirty to understand more the reality! We don’t need an university grade to figure that out!