Why do they call it Dutch 'disease,' anyways? - Macleans.ca

Why do they call it Dutch ‘disease,’ anyways?

Stephen Gordon on why Canada’s much-talked-about affliction is actually a blessing


So Dutch consumers are roughly 10% better off than they would have been, but companies have been able to compete only by paring their profit margins.

“The Dutch disease,” The Economist, November 26, 1977

Talk about burying the lede. That sentence appears at the end of the 10th paragraph of the much-referred-to but rarely read article in The Economist that coined the phrase “Dutch Disease.” In  the normal course of things, a 10 per cent increase in consumers’ purchasing power would be the stuff of banner headlines, but, for some reason, The Economist chose to hide that point deep into the story and qualify it with a caveat about how hard it had become for companies to compete. (The answer to that, by the way, is: “So what if producers are struggling?” What really matters is consumer welfare.)

My take on the Dutch Disease debate can be summed up as follows: Why are we calling it a disease?

The Economist was reporting on the difficulties Dutch manufacturers were having in the wake of the discovery and exploitation of natural gas reserves in the Netherlands in the 1970s. The parallels with the recent Canadian experience are obvious: the surge in the prices of oil and other commodities that began in 2002 has been accompanied by a fall in employment in the Canadian manufacturing sector. (Here I’m going to concentrate on the period 2002-2008; the recession complicates the analysis of the last four years.)

No-one can plausibly deny that the increase in resource prices during 2002-2008 led to a reduction in manufacturing employment. But this shift was part of a labour market adjustment that produced broad-based increases in wages and incomes — and broad-based increases in wages and incomes are good things.

The rest of this post is based on a presentation I gave at a the symposium on Dutch Disease organised by the University of Calgary’s School of Public Policy held in Toronto on March 6-7. (It was there that the University of Alberta’s Andrew Leach drew my attention to that remarkable quote.)

The Dutch Disease story is one of sectoral shifts.  Let me explain. Suppose, for now, that there are only two sectors in the economy: manufacturing and resources. (If this level of abstraction bothers you, replace “resources” with “non-manufacturing” in what follows.) If people can work in either sector, wages will be the same in both — otherwise, workers would move to the sector that pays more.

Now suppose that world demand for Canadian resources increases. This increases the demand for labour in the resources sector:

The blue lines are the demand for labour in the two sectors (higher wages reduce the quantity of labour demanded) and the red lines are the supply of labour (higher wages increase the quantity of labour supplied). Market-clearing wages and employment are at the intersection of the labour demand and supply curves — where supply equals demand and there’s no excess supply or excess demand. Before the shock, wages and employment are at levels denoted by the 0 subscript.

The immediate effect of the increase in the demand for labour in the resources sector is to produce upward pressure on wages in the resource sector. Some people stop the story there and conclude that the only people who benefit from high commodity prices are those who work in the resource sector, but this is obviously an incomplete analysis. Manufacturing workers will notice the higher wages on offer in the resource sector and react by moving from manufacturing to resources:

The arrival of new workers in the resource sector attenuates the wage increase there, and — this is the important bit — the departure of workers from the manufacturing sector increases manufacturing wages. The shift of workers from manufacturing to resources will continue so long as the resource sector offers higher wages. When wages are equalized across sectors — at w1 — there is no reason to move from one sector to the other. Note also that w1 is higher than w0: an increase in labour demand in one sector increases wages in both sectors.

Here are the main predictions of the sectoral shifts story:

  1. Employment in the manufacturing sector falls
  2. Employment outside the manufacturing sector increases
  3. Wages increase in both the non-manufacturing and manufacturing sectors

The first prediction obviously fits the data: according to the Labour Force Survey, manufacturing employment fell by 328,000 between 2002 and 2008. So does the second: even though manufacturing employment fell, unemployment rates hit record lows and employment rates hit record highs during Canada’s Dutch Disease period. And here’s what happened to real wages:

The Dutch Disease period was a time when real wage growth accelerated after decades of sluggish improvement: the increase in median real wages in the six years after 2002 was the same as the increase observed over the preceding 18 years.  As far as consumer welfare goes, this is a much more important statistic than the number of manufacturing jobs lost over the period.

There is one curious thing though: manufacturing wages are generally higher than wages in other sectors. How can a decline in employment in a high-paying sector generate an increase in overall wages? The answer is that the manufacturing jobs that disappeared were generally low-paying. (The numbers in the tables are calculated from the Public Use Microdata Files of the Labour Force Survey.)

Hourly wages: 2012 dollars
All jobs:
Manufacturing jobs:
Lost manufacturing jobs:
Mean hourly wage: $21.49 $22.39 $18.00
Median hourly wage: $19.05 $20.42 $16.50
Interquartile range: [$12.53 , $27.38] [$14.60 , $27.98] [$11.58 , $24.08]

(The interquartile range spans the middle 50 per cent of the distribution: its lower bound is the 25th percentile and its upper bound is the 75th percentile.)

The lost jobs were not only relatively low-paying compared to the rest of the manufacturing sector, they were low-paying relative to the economy as a whole. Again, this is consistent with a sectoral shifts story in which workers leave the manufacturing sector in search of higher wages elsewhere.

The next question is whether or not the new jobs created in the non-manufacturing sector did in fact pay better than the manufacturing jobs that were lost. For the most part, they did:

Net change in employment: 2002-2008
Manufacturing Non-manufacturing Total
Lowest quartile
(under $12.53/hr)
-142.5 -140.8 -283.3
Second quartile
($12.53/hr – $19.05/hr)
-80.3 550.0 469.8
Third quartile
($19.05/hr – $27.38/hr)
-49.0 740.1 691.1
Highest quartile
($27.38/hr and up)
-56.5 968.4 911.9
Total -328.3 2117.8 1789.5

The lost manufacturing jobs are mainly concentrated in the bottom part of the wage distribution, and the new non-manufacturing jobs are mainly concentrated in the top part of the distribution. This doesn’t mean that no high-wage manufacturing worker ended up at a low-wage non-manufacturing job, but it does mean that those transitions were not representative of what was going on during the shift out of manufacturing.

There’s another point to consider that I cover in my presentation: the mechanics of the transition. I may get to this later, but for now suffice it to say that the reduction in manufacturing employment between 2002 and 2008 was brought about by attrition: layoff rates held steady while hiring rates fell.

So why does an income-boosting shift of workers from one industry to another get such a bad name? I blame the editors at The Economist. A phenomenon that increased consumers’ buying power by 10 per cent should never have been called a “disease.”



Why do they call it Dutch ‘disease,’ anyways?

  1. Pretty clear why the Economist sees it as a disease – it reduces profits thus is ‘bad’ for business. The increase in wages and purchasing power though should actually be good for business in the long run.

  2. Perhaps because over a number of years you will have manufacturing jobs decline by 10-20-30-40%…(increases in other sectors) So the remaining jobs are high-tech, high value added, but less in number.

    Also , there are all kinds of externalities derived from switching from manufacturing to ressources…and when ressource prices fall and the “dutch disease” stops, what manufacturing has left will not come back for many years, for the capital in the infrastructure is dead and you will have a dire situation in between.

    • I talked about that. Investment and R&D are holding up – look at the last chart in the slides I presented.

      • That doesn’t help if all of the investment and R&D dollars are being funneled into resource extraction related projects.

        • Those numbers *are* for the manufacturing sector.

  3. How one can quote from a 1977 Economist article, and then not look closer at the economy of the Netherlands over the period where the term came from is beyond me.

    The Dutch situation was from a boom in natural gas in the North Sea. Anyone remotely familiar with the oil and gas industry would note that gas development and processing is not even remotely the same as bitumen extraction, processing and extraction. Particularly open pit mining (Syncrude, Shell, Suncor etc.) The latter is much more labour and capital intensive. And the principal demand for labour is during construction – which may include a great deal of lower skilled individuals who might otherwise work in factories. Once the plants/oilsands developments are built, the numbers involved in actually operating the facilities is actually quite low. Even less if you upgrade and refine elsewhere.

    So, the problem (if there is one) with a resource boom in Canada’s oil sands arises when the manic pace of construction stops.

    The Dutch situation was quite different. Gas field development and processing is not nearly as capital/labour intensive. So, not at all the same dynamic. The labour squeezed out of manufacturing by a rising Dutch guilder were not typically in demand by offshore gas development.

    There’s another point to consider that I cover in my presentation: the mechanics of the transition.

    That’s quite a big enchilada. This isn’t the stock market where you can just call up your broker and seamlessly change your investment portfolio. Severance pay, selling off of land and assets…then turnaround a few years later and hire and train a skilled workforce from scratch is daunting – especially if the industry hasbbeen established and is profitable elsewhere due to your earlier moves.

    • Offshore drilling is also pretty capital-intensive. The Netherlands has a large marine engineering sector that designs & builds all those offshore drilling rigs and support vessels. The product doesn’t require as much processing as oilsands, but the exploration and drilling are considerably more demanding than SAGD wells.

      • True enough. But different skill set. Welders and tech people can and do move around. And don’t typically work in factories during slow periods.

      • Without doing some research, the sense I got from my limited understanding of the original “Dutch Disease” was that the upward pressure on the Dutch currency was primarily due to the value of the natural gas (lesser so do to the inflow of capital investment). Once you get the wells, pipelines and plants built, not much more capital spending, but the gas continues to flow keeping the currency high. Though it declines over time, presumably bringing the currency down with it.

  4. Scratch this. I found what I wanted in that last chart.. the way you talk about it didn’t make it clear that it’s actually looking at employment numbers, I thought it was more looking at pay-rates. But it’s combining employment numbers with pay-rates.

    So there was an overall gain in employment of almost 2 million people in 6 years? Wow.. what did prices do, I wonder? And what happens now when the resource is fully extracted?

    • ‘Now’?

      • A turn of phrase. Remove the “now” if it offends.

        • We do what we always do. Adjust to the new situation, whatever it is.

          • “Adjust to the new situation, whatever it is.” Best line of all.

          • So in other words, looking long term isn’t something we should bother with.


          • Half of Canadian workers aren’t yet working at the job they’ll have five years from now. What’s the point of figuring out what jobs people should have 100 years from now?

          • Yeah, why bother looking at the climate when the weather changes every day.

            Please. You can do better than that.

  5. The traditional role of colonies has been to ship away raw resources to the motherland and buy finished goods back from them. The net result of such an exchange is almost invariably the impoverishment of the colony. History is littered with examples of this pattern. The Dutch disease is just a special case of this more general phenomenon.

    • And yet today the roles are reversed. Resources are commanding high prices, and the prices of manufactured goods are on a secular decline.

    • Yes, there is no future in digging holes in the ground. Harper is turning Canada into a colony for multinational corporations. The most reliable wealth creation is in innovation and information-based technologies. That provides the best business opportunities and jobs. We need to emulate countries like Germany and South Korea not Russia and Brazil.

  6. Life (and economics) is never as simple as a diagram representing a system of connections or interrelations among two or more things by a number of distinctive dots, lines, bars, etc.

    As I recall, without skipping over to wikipedia, during the seventies, the Dutch society was taken aback when oil was so scarce that Sundays became ‘auto-free’ Sundays. Finding oil or natural gas within Dutch borders (or within Western European borders for that matter) was/is quite a discovery, because there ain’t too much of it to be found over there. The currency could have increased just because of that finding alone.

  7. If the rest of Canada so chooses, massive new refining industries creating thousands of high paid jobs can be created in BC (to export refined products to Asia and California) and Montreal and New Brunswick (to export refined products to the Eastern seaboard and Europe), can result from continued oilsands development and pipeline building and pipeline reversals. Some Dutch disease.

    • Yes, just ask grade school students what they want to be when the grow up. The most popular answers are, of course: oil refinery worker and open-pit miner…

      Even if the pipelines fall through, these refineries will create about 40,000 jobs. But compare that to the 500,000 export-related jobs that have been lost since the dollar soared 60% from 2003 to 2008.

      The Dutch Disease is a very serious issue because it increases the cost of value-added exports and tourism. That kills business opportunities and jobs.

      We need to strike a balance and diversify. Betting everything on dirty-energy in this century is a dangerous gamble with our economy.

    • Certainly anything is possible, but they would not be being created by market driven forces.

  8. This columns says the increase in (real) manufacturing wages is a good thing. That fails to take into account how increased wages means increased wage costs for businesses. That makes businesses less competitive. That means they pack up and set up shop somewhere else (like the American south or Mexico.)

    So having a bump in real wages does you no good if it puts you out of a job.

    According to the OECD, the Canadian dollar is overvalued by 25%. That makes all value-added exports and tourism 25% more expensive. This is a great burden on these kinds of businesses and the associated jobs (which is why up to 500,000 such jobs have disappeared.)

    There is no free lunch. If Canada adopts the US greenback as our currency, we will have to undergo an “internal devaluation” (wage deflation) like Greece and Spain when they adopted the euro. That means wages will have to come down by about 20% to restore competitiveness.

    Is that supposed to be a good thing? Working for less pay? That will aggravate the housing slump and increase people’s debt burden (both private and public.)

    An overvalued dollar has no benefits except for the rich. It’s better to have a properly valued currency.

    • Dutch Disease in the headlines:

      G&M: Harper government funded study arguing Canada suffers from ‘Dutch Disease’

      National Post: Thomas Mulcair’s Dutch Disease warning supported by OECD report

      G&M: RBC: Shrugging off Canada’s competitiveness shortfall
      “Almost three-quarters of the competitiveness gap is due to the soaring loonie”

      G&M: Strong loonie casts shadow over recovery: Mark Carney

      CBC: Is Canada suffering from ‘Dutch disease’?
      “shifting international trade patterns are responsible for some of the 500,000 manufacturing jobs that have been lost in Canada”

      G&M: ‘This is about as bad as it gets for Canadian exporters’

      G&M: Slowing world economy hits Canada, U.S. trade
      “‘Canada’s trade performance continues to suffer amid a global economic slowdown and a high Canadian dollar,’ Toronto-Dominion Bank economist Francis Fong said in a note to clients.”

      G&M: Mid-sized Canadian manufacturing, up in smoke

  9. “In the normal course of things, a 10 per cent increase in consumers’ purchasing power would be the stuff of banner headlines”

    The problem is the increase in purchasing power as the dollar soared 60% from 2003 to 2008 (when it reached parity) has not translated into lower prices.

    As Mike Moffat of the G&M points out:

    “Prices for retail goods in Canada remain stubbornly high, despite the Canadian dollar rising substantially since 2002. Normally this should lead to falling prices, as a higher Canadian dollar lowers the (Canadian dollar) price of foreign goods. This is not happening. A decade ago, a product that cost $1 (U.S.) in the United States cost, on average, $1.20 (Canadian) in Canada. A decade later, there has been little change.”

    Fact is if one follows the roller coaster ride of the dollar since 1995 it has had little effect on prices (inflation) as it plummeted and soared.

    I think the real reason is demand. If a government decreed an across-the-board increase in wages by 25%, prices would shoot up 25% because businesses charge what the market will bear.

    So when the Canadian dollar fell, in terms of US dollars, businesses had to charge less and take a hit on profits. When the dollar rose, businesses could get away with charging more and make bigger profits.

    The only way prices will equalize out is when the purchasing power advantage evaporates. But by then Canadians will be working for 20% less and their debt burden (public and private) will be 25% more. Only the rich will have a net benefit (25% increase in their wealth.)

  10. Good post. Truly it is pathetic how a 10% gain in purchasing power is simply dismissed. It betrays the ideological leanings of the authors. A true economist would be singing the praises of such success.

  11. An economist that doesn’t get it. It is the rise in C$ caused by the oil exports that hurt our exports, thus reducing manufacturing jobs. So what if the jobs in other sectors paid better? The laid off folks don’t have the means for retraining (all the govt ads on retraining are false as you will know once you try to access the ‘assistance’) to move to the better paying jobs. Let’s face it …we don’t expect the U. of Calgary to produce anything that hurts Alberta’s interest.

    • “So what if the jobs in other sectors paid better?”

      Really? Really? THIS is your rebuttal?

      You should look at the slides I linked to. Or at least read the second-last paragraph of the post.

      I guess I really do have to do a follow up post