OM DETTA AVSNITT
In this podcast Jeremy Leonard and I discuss a paper that he and his colleagues - Mohammad Shakeri and Richard S. Gray - recently published through the Institute for Research on Public Policy. The paper is titled, "Dutch Disease or Failure to Compete? A Diagnosis of Canada's Manufacturing Woes," and is available for download.
Transcript
Brady Deaton Jr.: Welcome to FARE Talk where we set out to provide enduring discussions on contemporary topics relevant to our economy with particular emphasis on food, agriculture, and the environment. My name is Brady Deaton, Jr. of the Department of Food, Agriculture, and Resource Economics at the University of Guelph. I'll be your host.
Today my guest is Jeremy Leonard. Jeremy is a research director at the Institute for Research on Public Policy. He and his co-authors, Mohammad Shakeri and Richard Gray, have just recently released a study titled, Dutch Disease or Failure to Compete, A Diagnosis of Canada's Manufacturing Woes.
Jeremy, welcome to FARE Talk.
Jeremy Leonard: Good to be here, Brady.
Brady Deaton Jr: Jeremy, before we start, talk to me a little bit about the Institute that has published this study.
Jeremy Leonard: Sure, I'd be pleased to. The Institute for Research on Public Policy is a national think tank based in Montreal, Quebec. We're just celebrating our 40th anniversary this year. We were created in 1972 as an independent think tank.
We don't have members and we do not have supporters. We're supported by an endowment fund, which basically allows us to produce studies that are evidence based, and try not to take ideological or political sides in debates. Our role is really to inform and spark debate; pose the questions before we come up with the answers. We study a diverse array of issues including economic issues like the one we're going to talk about today, as well as more social issues like immigration, aging, and a whole host of other issues.
Brady Deaton Jr: All right, and I should say that we will link the listeners up to your site so that they can download this study if they want to read it more fully. Let's start off just by impacting your question. The paper's question, Dutch Disease, or Failure to Compete.
What generally is prompting this question?
Jeremy Leonard: What's prompted the question is two facts over the past several years that are undeniable. One is the fact that the Canadian dollar has strengthened considerably over the last five to six years, going from about 60 to 65 cents to parity today. That's a fact.
Another fact is that the manufacturing sector has been shrinking in Canada over the past five, six, seven years. That shrinkage started happening well before the recession, and it happened at the same time as this appreciation of the currency.
There's a natural question because the issue of the Dutch disease is really just that. The one, appreciation of the currency, causes the other, problems in manufacturing. We're clearly seeing these two things happening at the same time. We thought it was important to sort of dig a little bit deeper on the question, is one really causing the other or are they both happening at the same time for some other reason that we haven't taken account of.
Brady Deaton Jr: One of the things that I was trying to work through in kind of preparing for this podcast, was a better understanding of exactly this term, the Dutch disease. When you first hear it, it sounds like it's something that's damaging to the economy overall or having some kind of miserating effect on the economy. But, in wading through it, and I'd like to get your thoughts on this, it seems more like an effect than a disease. What are your thoughts about that?
Jeremy Leonard: The term Dutch disease comes from Holland as the name suggests. There were discoveries of natural gas off the North Sea off the coast of Holland, and that created tremendous demand for those energy resources. That energy demand can be a good thing for an economy because it brings in dollars; it increases exports; and a whole host of other things.
One of the things it also does it that the people who are interested in buying these resources are using Dutch currency to do it and it causes an appreciation in the Dutch currency which did raise the cost of exports to Dutch manufacturing exporters. And, in fact, the Dutch manufacturing sector did decline considerably.
You could argue that the term Dutch disease is really an effect of something which can at the base have some positive effects. Since then, the Dutch disease has come to be a convenient shorthand to say, it's the exchange rate that's causing problems in the manufacturing sector.
In some sense, in Canada, you can say the same thing about the developments of the oil sands. People can have different opinions about these aspects of it, but one thing that's absolutely clear is that it has brought very large amounts of money and economic resources into the country from the many other countries who are demanding these energy resources.
That still leaves us with the question that we try to address in this study, which is, to what extent does that boom and those economic good times for the energy sector by force mean economic bad times for the manufacturing sector. That was really the goal of the study.
Brady Deaton Jr: Right. So, it's not, isn't bad for the economy. The whole idea of your paper as I read is really focused on this effect on the particular sector, in this case manufacturing. So, is there a negative relationship between a resource boom and manufacturing output. That's really what's going to drive the research that we're going to talk about.
Talk to me a little bit about the theory, the abstraction, that drives the argument for there being a negative effect between a resource boom and manufacturing output.
Jeremy Leonard: Yeah, sure, and I touched on it a little bit, but we can go into it in a little bit more detail. The issue really boils down to the effect that there are some goods in the economy that are tradable and some that are not tradable.
What happens when you have a boom of any sort, whether it's a natural resource boom or any other kind of boom, is you have a sector where there's very rapidly growing demand for services and production. That tends to bid up wages because the higher your demand is, you need to hire more workers, and anyone who has studied Economics 101 knows that in those conditions wages are going to rise and they can rise rapidly.
So you'll have upward wage pressure in the booming sector, which is then going to trickle over into other sectors. People will be drawn into the energy sector, that will mean fewer people wanting to work in the non-energy sectors. The main result is you have this upward pressure on wages.
This upward pressure on wages causes problems in industries that are exporting goods. Their costs are going to go up and they will not be able to raise their prices. Essentially what it causes is an increase in the terms of trade, which causes problems for exporters. We see this has manifested itself in a rising currency.
That's sort of a very complicated way of explaining it, but it really has to do ... It's much more than just about dollars floating around the economy and whether the Central Bank is creating too much money or not creating enough or issues like that. It really has to do with some pretty fundamental economic effects that come about from this resource boom.
We know the economic forces at play and the question then becomes, what's driving those economic forces, and then there are a number of things that can effect that. That's sort of the theory behind it. It's very well developed and you can look at it in terms of how flexible sectors are. In other words, how much labor mobility there is in sectors and things like that.
Invariably what you find is that there's a possibility that the manufacturing and trade intensive sectors will be adversely effected, but it's not a necessity. In other words, you can certainly envision scenarios where you can actually see a resource boom and you wouldn't necessarily see an adverse effect.
Given that the theory doesn't give you a definite answer on whether a resource boom is going to cause problems in manufacturing, we decided we needed to look at the actual numbers in Canada to try to answer that question.
Brady Deaton Jr: All right, so let's get into this. As you pointed out earlier, Canada's energy sector has experienced a boom; oil production has increased; gas has increased; Canadian exports have increased. We've got the first part. Talk to me about how you examine then how this effects manufacturing.
Jeremy Leonard: What we did was we basically did a two stage process. The first thing we wanted to get a handle on was to what extent is the strong Canadian dollar being driven by energy prices. There are many, many things that can affect the exchange rate besides energy prices. It can be prices of other commodities. It can be the stance of monetary policy. For instance, if interest rates are higher in Canada than they are in the United States as they have been for quite a few years, that's going to encourage investors to invest in Canada, which will also put upward pressure on the exchange rate.
So the first stage was to examine the extent to which energy prices are associated with high exchange rates. The answer is that there is a linkage there, but the interesting finding there was that it's not just energy prices that are driving this exchange rate. It's also prices of other commodities like wheat; other commodities that Canada produces like industrial metals. So there is a partial effect.
Stage one was simply to establish what piece of the strengthening of the Canadian dollar could be attributed to rising energy prices.
Brady Deaton Jr.: Before you get to the second stage, first stage you're basically just saying what is the relationship between the Canada and U.S. exchange rate and energy prices and controlling for a bunch of other factors.
Jeremy Leonard: That's right.
Brady Deaton Jr.: What time period are we talking about that you're looking at here?
Jeremy Leonard: We looked at the 1992 to 2007 period. We wanted to look at a longer period than just the boom to get a sense of trends that were happening earlier on. What we found was actually that the effect of energy prices on the exchange rate was about the same and possibly even a little bit less than that for other commodities. That, to us, was an interesting-
Brady Deaton Jr.: That is interesting.
Jeremy Leonard: ... observation. The other thing that we didn't talk about but other researchers have looked at is the fact that there are two things that drive the Canada/U.S. exchange rate. One is the strength of the Canadian dollar, but the other is quite frankly the weakness of the U.S. dollar, which may have little to do with what's going on in terms of resource prices.
We cite in our paper some work that's been done, it's not published yet, but basically indicating there's a piece of the strength of the Canadian dollar that really has to do with the weakness of the U.S. dollar related to the fiscal and economic problems south of the border.
All that to say that the energy piece of the appreciation of the Canada dollar is a lot smaller than the total appreciation that we've seen. Another way of saying that is even if we had not had such an energy boom, we probably still would have seen a strengthening of the Canada/U.S. dollar for other reasons.
I think that's an important observation to make, that we can't just attribute this 40% to 50% in the Canadian dollar 100% to energy because that's not what the data tells us.
Brady Deaton Jr.: This seems to be some evidence against the argument that there's a petrol currency here. If I understand you correctly, you're saying that energy prices effect the exchange rate but so do non energy commodities and other factors and relatively more so than this energy price effect on the exchange rate.
Jeremy Leonard: That's right. That's absolutely right. It comes down to the fact that you can ... you know, many of us have seen these charts where you plot the Canadian dollar against the oil prices and you get a nice match. The fact of the matter is you can actually plot it against a whole of different price indices and you can get a similar match. So one of the points of doing this research was to try to take all of these factors into consideration and ask ourselves, how much is the one effect when controlling for the other.
It would be just as incorrect to call the Canadian dollar a petrol dollar as it would to be to call it a nickel dollar or a wheat dollar or any other commodity you might like to call it.
Again, stage one was really a kind of cautionary tale in the sense that, yes, we do see this positive link as the Dutch disease theory suggests, but we can't draw the conclusion that energy prices are driving exchange rates and by extension, weakness in manufacturing.
Brady Deaton Jr.: That takes us to the second step. If you've established some effect between energy prices and the exchange rate, but there are a lot of other things that are effecting it, and energy prices are a smaller perhaps than expected driver.
And, now we're moving to the second step. Okay?
Jeremy Leonard: Right, because really the ultimate question we want to know is, to what extent have the energy induced strengthening of the dollar adversely effected manufacturing. The way we go about doing that is we actually looked at 80 different manufacturing industries because we wanted to get down at a pretty fine level of detail because you lose a lot of interesting differences and constraints if you look at the sector as a whole.
We looked at 80 different industries. For each of those industries, we essentially estimated outputs using a model that ... You know, output estimation models have certain standard elements to them, but we added this energy price induced appreciation of the exchange rate that we've calculated from stage one.
That was one key element that went into that statistical estimation. The second one that we put in, which was very important, was the trends and output in the corresponding industry in the United States. Now, why did we do that? We did that because we wanted to kind of use the United States as a control country, as a country that is not a large energy exporter, certainly wasn't over the time period we looked at, with the notion being that if the Dutch disease, if this exchange rate induced adverse effect on manufacturers, if that's really the explanation, then we shouldn't see kind of similar trends in output in manufacturing in the United States. They are not afflicted with this Dutch disease.
When I talk about the results, I'll talk about why that's important. It was important for us to have a kind of control case to say, how is output evolving differently in Canada relative to the United States?
We looked at all of these 80 industries. It was quite an arduous task punching all the numbers.
Brady Deaton Jr.: I'm going to say something. For those of you listening, in their paper they actually do a great job of providing a lot of data and a lot of tables that actual allow you to look at the different industries and the various effects on them. That's something that's really well done that I know you're not going to be able to go in details in this discussion.
Jeremy Leonard: Yeah, and really it's because the nuts and bolts of the detailed results are not as interesting as kind of the overall fundamental result of the question we're after, which is what is essentially this variable we're looking at; this kind of Dutch disease term, if you will. Is it significant? Is it statistically significant in these equations?
The answer was obviously nuanced, but I guess a little bit perhaps to our surprise given the popular discourse about the Dutch disease, is that the majority of industries that we looked at really didn't exhibit a strong adverse effect of this energy induced increase in the exchange rate.
What do I mean by that? Even if you accept that energy prices have had an upward influence on the exchange rates, that has not generally been associated with declining manufacturing output once you control for other factors effecting output.
It seems a bit counter intuitive because ... and, that's something we've kind of heard in the political discussion in recent weeks is how can that possible be because we've seen the dollar has strengthened and manufacturing output has declined. So, how can you possibly say that the two aren't linked?
The answer comes back to this notion of the control variable in the United States. There are other factors that have happened that have been going on over the past, during the 2000's, that have been at the same time as the resource. The most important of these is really the rise of China and other low cost producers in global value chains. In fact, most listeners may not know this, but China has actually passed Canada as the United States' number one source of imports into the United States.
This competition is not just happening in traditionally in t-shirts and toys and other things that we've known the Chinese have been exporting to us for decades. They're moving into higher value industries like machinery, like fabricated metals, things feeding into the automotive sector. They're actually starting to compete head-to-head with Canadian exporters in U.S. markets.
This factor, I'll kind of just call it the globalization of manufacturing, has actually been a much more important negative effect on not only Canadian manufacturers, but American manufacturers as well. So what do the results tell us? They tell us that there are certain industries that have indeed been adversely affected by the exchange rates.
The other industries where we don't see an adverse effect from the Dutch disease, these industries are still having problems but they're much more related to trying to compete with the South Korea's and the China's and the Brazil's, and the other emerging markets of the world. Not only within Canada, but more importantly in markets where Canada exports, mostly notably the U.S. market.
Brady Deaton Jr.: Are there any characteristics of those industries that were negatively affected by the exchange rate; do they have any characteristics that are worth noting?
Jeremy Leonard: Yeah, well, I'll just highlight two extremes really. The ones that had the largest negative Dutch disease coefficients in absolute terms were the clothing and textile industries as well as leather products, which is basically namely footwear and handbags. This kind of comes back to what I was saying earlier. These are industries for which there's not much product differentiation. They're sort of commoditized industries. I like to just use the example, a t-shirt is a t-shirt is a t-shirt and there's not really a lot of opportunities to diversify and add value to your product.
Secondly, these industries are very, very trade intensive. They have a very high proportion of their product that is overseas and they also have a very high import penetration rate. It's sort of not surprising that they would be most affected by the exchange rates and least able to adapt to it, I guess is the way I would put it.
These industries, you have to remember, textiles, apparels, and leather together probably make up ... they certainly make up more than 5% of Canadian manufacturing. They are relatively small sectors.
The other sector that was kind of interesting was actually a piece of the chemical industry, the pharmaceutical sector, which also showed a significant negative Dutch disease effect. It's hard to know why this would be the case because pharmaceuticals are often proprietary products and so they may not have direct competitors. It may be that a lot of these companies are global and they may have moved production around, moved production out of Canada. That's just speculation on my part.
The reason I touched on that is because pharmaceuticals is a pretty high value added industry. It does a lot of research and developments, and in general contributes to innovation, certainly in the medical field but also in other endeavors as well. I think that's of some concern because that's one of the things that people worry about with the Dutch disease. If there are certain sectors in manufacturing that are very innovation intensive that contribute to benefits in terms of advancement of knowledge, things that might be beneficial to other sectors. If those sectors are declining because of the high exchange rate, then there is some reason for concern.
Those are kind of two examples of sectors where we didn't find the specific significance. If I can just go on to a couple where we didn't, which was actually-
Brady Deaton Jr.: Sure, sure, that would be great.
Jeremy Leonard: ... because the auto sector was one. That kind of is often held up a poster child of the ... Here is the prime example of what the strong dollar is doing to Canadian manufacturing. Actually when you look at the numbers, you see that the Dutch disease effect really isn't there. It's not statistically significant. Why is this the case?
It comes back to what I was just saying about increase in competition. The interesting thing is that we've seen a decline in the Canadian automotive sector over the 2000's. We've also seen decline in U.S. automotive sector over that same period, which kind of illustrates the point that the problems in the auto sector are not unique to Canada and they're not caused entirely by the high Canadian dollar. What they're being caused by is companies like Kia and Hyundai. Kia and Hyundai weren't even on the radar screen 15 years ago and they have become pretty big players in the automotive market here in North America.
That's just to illustrate that some of our common perceptions of how the exchange rate is effecting various industries really aren't quite in agreement with what the empirical evidence tells us.
Brady Deaton Jr.: When we return back to your paper's title, Dutch Disease or Failure to Compete, I'm reading your results and you tell me if I've got the story that you're telling correct. There are some, but not most, industries in the manufacturing sector that have been effected and perhaps slightly, but most are not, and so the manufacturing woes that Canada is experiencing are largely a failure to compete and not the result of Dutch disease. Is that-
Jeremy Leonard: Yeah. That's a fair way of putting it. I think we have to be fair to say that when you add up all of the industries for which we saw a negative Dutch disease, it's about one quarter of the manufacturing sector that has been significantly adversely effected by the high exchange rates. When one quarter is large or small is a matter of taste, but that's what the numbers tell us.
From our perspective, what that tells me is that if we want to address the bigger problems of manufacturing, which are real. I guess I want to emphasize that in this discussion, and I said it at the outset. The manufacturing sector has shrunk in Canada quite significantly since about 2005. That is a real problem.
The issue is that if we want to resolve that problem, we need to have the diagnosis of what's causing it right. Our results certainly show that the majority of that is being caused by something other than the exchange rates. As I was saying earlier, that something is a failure to compete, which is coming both from an increase in competitive pressures from abroad. It also, quite frankly, simply reflecting the fact that productivity growth in manufacturing, which is really what you need to stay cost competitive, has been very, very sluggish over the past 10 to 15 years.
Just to give you a couple of data points, in the United States, productivity has increased by the order of 2% to 3% per year. In Canada, we're basically flat lining. These numbers don't sound like much, but when you compound them over 10 or 15 years, you just realize that there's no way to be cost competitive unless you can make the investments, change the ways you do things so that you can become more productive and you can compete, quite frankly, with the dollar at any level it might be whether it's at 80 cents or parity or anywhere in between.
Brady Deaton Jr.: Do you have any policy suggestions?
Jeremy Leonard: There are a couple in the paper. I think the first thing to preface is what shouldn't be done. This really talks to some of the stuff that's been talked about in the political discourse. Our study shows that it's really not an either/or zero-sum game in terms of the well-being or the boom of the oil sands and the challenges in manufacturing.
To frame the debate that way and to organize policies around the notion that if only we could keep the energy boom in control or maybe even reduce it, that would solve manufacturing's problems. That just isn't borne out by the data.
That would be something I would not be in favor of, nor would I be in favor of trying to manipulate the exchange rate even if you could. One, I don't think it is easy to do. Secondly, I think it would be counterproductive. Thirdly, it really would not address the core issue.
We get back to this issue of productivity and innovation. This has been a tough nut to crack for policy makers for years. One issue is connecting businesses, especially small businesses, to universities and other sources of potential innovation. There are many, many firms in Canada who do not do any research and development in house because they don't have the resources to do so. So, creating networks to connect them with best practices is one thing to do.
A second kind of more broad, strategic direction is to think about increasing competitive pressures within certain industries. There are still may industries in Canada that are critically important in term of innovation that are protected. Look at telecom as one, although changes are sort of in the works there. Air travel, the financial sector, there are very high barriers to entry. These are all very important sectors, and innovation in these sectors can have implications for productivity and innovation in other sectors.
So increasing competitive pressures generally in the economy. On a related point, it all comes back to competition, is sort of broadening marketing opportunities for Canadian firms. This is something the government has started to do. Kind of diversifying potential markets.
There are a lot of different reasons to want to do this, one of them is to obviously to tap faster growing markets. A second effect that I think is just as important is it puts Canadian firms in contact with firms that have very different cost structures; that have very different management styles; techniques for production; and other differences, which I think can sort of be fertile ground to get Canadian companies to think about maybe there are different ways of doing things; maybe we can do things more cost effectively; and we didn't know we could do it because we never really had to.
Competition ... I mean if necessity is the mother of invention, competition is the mother of innovation, and by extension, productivity. I think in terms of a manufacturing strategy, the government would be much better served to focus on those issues rather than focusing on how do we try to control the exchange rate or the development of the energy sector as a manufacturing strategy.
Brady Deaton Jr.: Jeremy, thank you very much for taking the time to speak us today. I really appreciate it. I really learned a lot. Thanks for the time you and the co-authors took to write the article.
Jeremy Leonard: I appreciate being here, Brady. It's a pleasure.
Brady Deaton Jr.: Thanks for joining us at FARE Talk. We hope you will continue to check our website for updates and the latest podcasts.
