In macroeconomic policy circles the trio of monetary independence, free trade/capital mobility and fixed exchange rates has come to be labelled the unholy trinity, since a country can choose any two of these but cannot opt for all three. Given that Governor John Crow’s celebrated 1988 Eric Hansen Lecture publicly committed the Bank of Canada to monetary independence (in the form initially of price stability but eventually of inflation targeting), and given that in the 1988 free trade election Canadians effectively rati- fied the Canada-US Free Trade Agreement (henceforth FTA), economic consistency thus dictated that flexible exchange rates would be an integral part of Canada’s policy future. But what surely caught the policy authorities by surprise was, and continues to be, the successive bouts of overvaluation and undervaluation of the Canada-US exchange rate, with its always-fascinating, sometimes-troubling interaction with North American economic integration. Addressing and assess- ing these interactive policy episodes between the loonie and the FTA will be the principal focus of the first half of this arti- cle. The remainder will deal at an analytical level with a vari- ety of issues including energy prices and the ”œDutch disease,” productivity, and regional balance, prior to concluding with some further reflections on the unholy trinity. To anticipate one of the central themes of the ensuing analysis, the utter volatility of the Canadian dollar is such that Canada has to find ways to mitigate or otherwise accommodate these exchange rate swings or we must seriously consider fixing the exchange rate and becoming part of a larger currency area.
By way of background material, figure 1 presents data on Canada’s exports to the US and the behaviour of the Canada-US exchange rate, expressed as the number of US cents per Canadian dollar, so that movement from, say, 80 cents to 90 cents represents an appreciation. Figure 2 presents data on manufacturing unit labour costs (ULCs) in Canada and the US, both expressed in US dollars. While these ULCs reflect movements in domestic wages and productivity, the major determinants of the swings in Canadian ULCs are the corresponding swings in the Canadian dollar exchange rate. Note that these ULCs are indexes that are set equal to 100 in 1992. Thus, while figure 2 captures changes in relative com- petitiveness it does not reflect actual manufacturing cost dif- ferentials at any point in time.
The overview of the exchange-rate/FTA nexus begins by focusing on the steep Canadian dollar depreciation in the pre- FTA period, from the mid-80-cent range in 1980 to just over 70 cents in 1986, and relatedly on the consequent near-doubling in Canada’s exports to the US from roughly $50 to $100 billion (see figure 1). This was the era of Presi- dent Reagan’s military Keynesianism, namely the sustained period of econom- ic expansion triggered by defence spend- ing and income tax cuts. The result was a surging US domestic demand, an appre- ciating US dollar, and a corresponding improvement in Canadian ULCs (figure 2), all of which led to Canada’s soaring exports to the US over the 1982-85 time frame. While the Canadian economic policy community was, almost to a per- son, enthusiastic about Canada-US free trade and while the Macdonald Royal Commission famously gave its ”œleap-of- faith” endorsement to free trade, it was this surge in Canada’s export penetration of the US market that finally delivered Canadian business to the FTA cause, at which point Prime Minister Mulroney stepped in and, as they say, the rest is his- tory. Canadian business was concerned that with the US current account deficit in excess of 3.5 percent of GDP, with pro- tectionist sentiment alive and well in Congress, and with our current account balance with the Americans turning sharply into surplus, this export penetra- tion could be curtailed or even rolled back either as a result of direct (intended) intervention or more likely as a result of Canada getting sideswiped by US policies directed toward the Japanese or Euro- peans or both. Indeed, in the 1985 Plaza Accord, the five signatories (France, UK, Germany, Japan and the US) agreed to intervene in foreign exchange markets to devalue the US dollar relative to the yen and mark. In this protectionist context, Canadian business embraced Canada-US free trade, hoping that an FTA would effectively lock in the existing export penetration. And it did. Thus the intrigu- ing first observation relating to the exchange-rate/FTA relationship is that without this early-to-mid-1980s deprecia- tion of the $C there may well have been no FTA.
As the slowdown of exports over the second half of the 1980s reveals, the anticipated benefits of the FTA did not immediately materialize. The reason for this has already been alluded to ”” the Bank of Canada’s adoption of price stabil- ity/inflation targeting as its monetary policy goal. The transition from the then- existing 5 percent plus inflation rate to the eventual 2 percent target was exceed- ingly costly ”” interest rates spiked from (roughly) 7 percent in 1987 to over 12 percent in 1990, and the exchange rate mushroomed from its low of about 71 US cents in the mid-1980s to nearly 89 cents in late 1991. As a result, and as is very evi- dent in Figure 2, Canada’s ULCs (in US dollars) underwent a truly dramatic esca- lation relative to US ULCS, the resulting decrease in Canada’s competitiveness constituting a major factor in triggering the slowdown in export growth in figure 1. This transition was costly in other ways as well. Under the influence of a rapidly appreciating Canadian dollar and interest rates in the teens, Canada’s econ- omy fell into a severe recession, well in advance of the early 1990s global slow- down. Moreover, under the dual impact of falling government revenues and ris- ing government expenditures (e.g., for welfare and UI) on the one hand, and much higher debt-servicing costs on the other, Canada’s outstanding federal deficits and debt spiked markedly upward. All in all, a not very propitious start for the FTA, and again with the exchange rate holding centre stage.
Beginning in 1992, however, the Canadian economy began to fire on all cylinders. The costly transition toward price stability was over and Canadian inflation rates fell below the US rates where they have remained ever since. On the fiscal front, Finance Minister Paul Martin, aided by the above factors and also by the tum- bling of interest rates from the early 1990s highs, tamed the federal deficit by fiscal year 1997-98, setting in train the still ongoing series of decade-long surpluses. Canada also benefited from the prolonged buoyancy of the US economy, propelled as it was by the hi-tech boom and the deft hand of Alan Greenspan at the Federal Reserve tiller. In terms of the FTA, however, the most telling development was once again a dramatic depreciation of the Canadian dollar, initially to the low 70-cent range and then, caught in the crosshairs of the 1998 Asian currency crisis, to a record low 60-cent range by millennium’s end. The result was a dramatic increase in Canadian competitiveness (figure 2), in spite of the fact that Canada continually lagged the US in productivity growth, on which more later.
Not surprisingly, in this economic environment, Canada’s exports to the US literally exploded, more than tripling in nominal terms over the decade of the 1990s. (Note that exports are on a gross basis, not a value-added basis, so that there can be double counting if a product crosses the border more than once. On the other hand, this would constitute evi- dence of a different sort of North American economic integration, namely evidence of a high degree of cross-border production integration.) This caveat aside, this Canadian-dollar-depreciation- triggered penetration of the US market had an enormous influence on the geo- economy and political economy of Canada, because it has ushered in an era where virtually every province trades more with the US than with the rest of Canada. As I’ve noted elsewhere, Canada is less and less a single national economy and more and more an east-west series of north-south economies. To be sure, greater north-south integration was no doubt inevitable, but the 1990s deprecia- tion in the context of the FTA/NAFTA sig- nificantly advanced the process.
The final historical episode in Fiugre 1, roughly from the turn of the millennium through to the present, is arguably the most disquieting of all, since if it persists it has the potential to wholly remake Canada’s internal eco- nomic geography. Underlying this ongoing policy episode is the rise of China and India as economic power- houses and the associated increase in the absolute and relative prices of raw materials compared to the prices of manufactures. Overlaying this was a series of further developments that had substantial implications for the price of oil ”” 9/11, the invasion of Iraq, Hurricane Katrina, the cut in Venezuela’s production and so on. The net result has been the spectacular increase in the landed world price of oil in Canada, from US$20 per barrel in the second quarter of 2001 to over US$70 in the second quarter of 2007.
What is every bit as striking is the sheer magnitude and rapidity of the resulting appreciation of the Canadian dollar (figure 1) and of the relat- ed deterioration in Canadian competi- tiveness (figure 2): Over the 2002-2007 time frame the dollar appreciated by 50 percent (from the low 60-cent range to the mid-90-cent range), and if the data in figure 2 were reorganized to focus on rel- ative costs since 2002 they would also reveal a 50 percent deterioration in Canada’s relative competitiveness.
Small wonder then that central Canadian manufacturing is being clob- bered and that the economic centre of gravity is quickly moving to the ener- gy patch. From a position in the late 1980s when Ontario’s share of Canada’s GDP exceeded its population share by 5 or 6 percentage points (or roughly by 15 percent), Ontario’s 2006 GDP share had fallen below its popula- tion share. In sharp contrast, energy- rich Alberta, with 10.4 percent of Canada’s population in 2006, now has just under 18 percent of Canada’s GDP and it has a per capita GDP nearly 60 percent above that of Ontario.
This completes the brief historical overview of the relationship between the loonie and the FTA. Attention now turns to selected ana- lytical issues associated with a volatile exchange rate in the context of an integrating North America.
The first of these relates to Canada’s intention in the FTA negotiations to include a subsidies code and some binding conditions relating to anti-dumping and coun- tervail duties. Perhaps the Americans would not have agreed to these provisions under any conditions, but the presence of a freely float- ing exchange rate and the possibility that Canada could depreciate itself into a super- competitive position would clearly make them non-starters. Phrased differently, provisions like subsidy codes will require restrictions on exchange rate volatility in order to have any chance of gaining acceptance. Alternatively, exchange rate movements could be incorporated in such provisions. The policy bottom line here is that softwood lumber would not have become anywhere near the issue it has become had the Canadian dollar remained near its mid to high 80-cent level when the FTA was implemented, instead of depreciating to the low 60-cent range. In the event, Canada’s softwood industry has suffered a double whammy of sorts ”” a US-driven softwood agree- ment followed by a 50 percent apprecia- tion of the Canadian dollar.
The remaining analytical issues deal in one way or another with the real- ity that the exchange rate appears to be very prone to overshooting in both direc- tions, that is, prone to successive bouts of overvaluation and undervaluation. To lend some quantitative perspective to this overshooting, the appropriate com- parison is with the purchasing-power- parity (PPP) exchange rate, which has some claim to be viewed as the equilibri- um exchange rate. Canada’s PPP rate vis- aÌ€-vis the US dollar is generally agreed to be in the low to mid 80-cent range (per- haps closer to 85 cents more recently). Thus, the exchange rate extremes in fig- ure 1 (low 70-cent range in mid-1980s; nearly 89 cents in late 1991; 62 cents in 2002; and over 95 cents at the time of writing) represent alternative bouts of undervaluation and overvaluation with respect to PPP.
My version of the mainstream view of this exchange rate overshoot- ing or misalignment in the context over overall macro policy since the FTA would go as follows. While the movement toward price stability/inflation targeting in the late 1980s and early 1990s was indeed costly in terms of output and debt accumulation, this was more than offset by the prolonged 1990s boom and the surge in exports facilitated by the Canadian dollar depreciation. On the reasonable assumption that capital equipment is priced in US dollars, the implications for Canada of the exchange rate depreciation were that capital became more expensive in absolute terms and in relation to labour, so that the incentives facing Canadian firms favoured meeting the increased demand for our exports by adding labour rather than capital. Indeed, this was the era in which our employment rates reached record highs. Readers will also recall that this was also the time frame that the brain-drain issue came to the fore since mobile highly skilled labour (i.e., human capital) was tending to move to the US in search of the much higher US incomes. Therefore, in terms both of physical and human capital our capital-labour ratio was falling relative to that in the US. This remains one of the reasons for Canada’s lagging productivity growth rates.
With the recent sharp and rapid appreciation of the Canadian dollar and the upward spike in Canada’s ULCs, the pendulum has swung fully back in the sense that overvaluation of the currency is now the challenge. There is an important difference, however, between this overvaluation and the one in the immediate aftermath of the intro- duction of the FTA, namely that the Canadian economy is currently perform- ing very well on the overall income and employment fronts. This is so because the implications of the overvaluation on manufacturing employment and output have been more than fully offset by the significant boom in the energy sector, and the resource sectors more generally. By way of an important aside, some of the ongoing manufacturing job losses relate to the excess of marginal jobs cre- ated in the earlier undervaluation period, i.e., there was also overshooting in the market for labour. This caveat aside, the incentives emanating from a Canadian dollar well above PPP are now clearly in the direction of increasing the capital- labour ratio. The bottom line, therefore, would be that our decade-long concern about lagging Canadian productivity is about to be rectified ”” it too has been subject to overshooting ”” and now we are about to deepen capital and to embark on high-productivity growth.
Hence, the mainstream vision, at least as I interpret it, is that exchange rate movements away from PPP have no long-term effects on produc- tivity: while there are obviously lags in the process, we are (or soon will be) essentially at the same position econom- ically as we would have been had there been no overshooting. Beyond this, our macro managers and the policy main- stream would also argue that flexible rates are extremely valuable in their own right because they serve to buffer the Canadian economy from global terms- of-trade shocks (especially energy/ resource shocks), thereby attenuating the impact of such external shocks on Canadian employment and/or output. For example, it is true that while there was a more than three-fold increase in the world price (US dollar) of oil over the 2000-2007 period, the appreciation of the Canadian dollar resulted in only a doubling in the Canadian dollar energy price over this same period, thereby dampening the magnitude of the on- going Alberta boom. By way of a transi- tion toward a different view of these bouts of overshooting one might also note that the havoc being wreaked on central Canadian manufacturing by the accompa- nying appreciation of the Cana- dian dollar can hardly be called ”œbuffering.”
The alternative interpretation, while obviously related, is more disquieting. The starting point is the same, namely that the undervaluation-driven export expansion was accompanied by a lower-than-otherwise increase in physical capital and an out- migration of human capital, both of which served to lower Canada’s relative capital-labour ratio and productivity growth. On the overvaluation side, there are indeed incentives to deepen the capital-labour ratio, but the rapidity and extent of the appreciation-triggered deterioration in out competitiveness (for both bouts of overvaluation in figures 1 and 2) is such that the immediate impacts are less likely to be the deepen- ing of capital and more likely to be downsizing, outsourcing abroad and off- shoring (locating abroad), not all of which will be limited to ”œmarginal” firms. In other words, some of the impli- cations for the non-resource tradeable sectors may well be cumulative, not off- setting, so that Canada risks being left with an economy that is less capital intensive and a labour force that is less human-capital intensive than otherwise would be the case. There is another and more straightforward way of viewing this challenge: given that the vast majority of our manufacturing and service exports are still destined for the US market, attracting domestic and foreign invest- ment to service this market from a Canadian location requires a degree of cost certainty that may well be unattain- able in light of the demonstrated degree of volatility in the Canadian dollar and in our unit labour costs.
This is a convenient segue into a dis- cussion of what has come to be called the Dutch disease, so called because Holland’s North Sea energy exports appreciated its currency to such an extent that Dutch manufacturing was clobbered. There is clearly an ongoing parallel in Canada ”” the rise in energy prices is contributing in a significant way to the appreciation of the Canadian dol- lar which, in turn, is hobbling manufac- turing (and other sectors where Canadian value-added is high) in central Canada and elsewhere. Purists might argue that at most it is only the apprecia- tion above PPP that can legitimately be attributed to the Dutch disease. But this is cold comfort to manufacturers who were comfortably settled in at the low 60- cent equilibrium. In other words, it is the overall volatility of the Canadian dollar that is destructive and energy/resource price swings are arguably at the heart of this volatility. To be sure, one can easily think of other events that could trigger the same implications. For example, were the Chinese to respond to the massive cessation of toy imports and to act on their hint of a compensating ”œrecall” of some of their trillion-dollar-plus holdings of US reserves, bills and bonds the US dollar could be in for a hard time indeed. And were Canada to decide not to allow the Canadian dollar to decline apace with the US dollar, the result would be similar to key aspects of the Dutch Disease. However, for present purposes we will stick to the implications for the Canada of the traditional Dutch Disease.
Beyond tilting prosperity toward the energy patch and away from the manufacturing sector, the Dutch Disease carries with it some other complicating factors that are probably unique to the Canadian setting. The most obvious of these is that energy rents and royalties fall under the provinces’ jurisdic- tion so that, in addition to its manufacturing woes, Ontario (actually Ontarians) will be called upon to fund roughly 40 percent of the energy-related equalization in full knowledge that the province will never be allowed to be a recipient.
Altering the equalization system to include the federal-provincial cash transfers to be allocated on an inter- provincial equity basis (outlined in my June 2007 Policy Options article) would presumably be a step in the right direc- tion, but it does not get at the underlying currency appreciation problem. In this regard, Norway finesses this problem by re-investing the bulk of its North Sea energy revenues back into the interna- tional capital markets, thereby attenuat- ing the appreciation of the Norwegian krone in the first place. Alberta could do this if its citizens so decided and they well might, although (and to provide a tempt- ing twist to a venerable Canadian chest- nut) it is not obvious that we would be well served over the longer term by appealing to the provinces to exercise their provincial spending power in areas of federal jurisdiction! Currently, howev- er, Alberta is doing the very opposite by taking (most of) its energy rents/royalties directly into its consolidated revenue fund and utilizing them to reduce income and sales taxes in the pursuit of creating the lowest tax jurisdiction in North America, let alone Canada, arguably further exacerbating the energy- patch/manufacturing disparity. The point here is obviously not to take Alberta to task for formulating policy within its constitutional parameters. Rather it is to make the straightforward observation that the Dutch Disease is more easily accommodated in those countries where energy rents/royalties accrue to the central government.
Nonetheless, the above constellation of issues are presumably part and parcel of the context underpinning of Peter Lougheed’s recent comments alerting Canadians to a looming war between Alberta and the federal government over environmental issues relating to the oil sands development and the energy sector generally, a struggle that the former Alberta premier thinks will surpass even the bitterness associated with the NEP. My take on this (Policy Options, June 2007) is that, when articulated, Ottawa’s environ- mental principles will become akin to the Canada Health Act principles in that Canadians will embrace them as part of their collective identity and will force their respective provinces into compli- ance. To be sure, this epic drama is likely to unfold irrespective of the exchange rate regime. However, it would seem easier for Ontario to agree to new auto emission standards and for Alberta to enact clean energy principles if both manufacturing and energy were mutually compatible economically within the FTA framework.
This brings us to the heart of the mat- ter. From my vantage point, the fun- damental reality is that the Canadian currency area is far too small to accom- modate both a world-class energy cluster (replete with likely massive foreign investment in pipelines and the tar- sands) and a world-class manufacturing sector under Dutch disease-prone flexible exchange rates. One obvious alternative would be to immerse Canada into some version of an FTA currency area, initially via fixed exchange rates (hopefully with the Federal Reserve supplying US dollars to the foreign exchange market when the US dollar is tending to appreciate, and the Bank of Canada supplying Canadian dollars when our currency is tending to appreciate) and over the longer term per- haps evolving to toward some FTA or NAFTA version of the Euro. Certainly, in the presence of volatile energy prices the major manufacturing and service export- ing areas would benefit from being inside a common FTA currency area.
To see this, consider the Windsor- Detroit or Vancouver-Seattle relation- ships (or any of the cross-border competing cities, or provinces vs. states, for that matter). A rapid increase in the world price of oil will affect Windsor and Detroit in a similar way. The same is true for Vancouver-Seattle, and so on. However, if the Canadian dollar appre- ciates in response to the energy price hike, then both Windsor and Vancouver and, more generally the Canadian side of all cross-border rela- tionships will be thrown way offside competitively vis-aÌ€-vis their US coun- terparts. This is but a different way of stating that the underlying asymmetry arising from energy and resource shocks tends to be east-west in both countries.
Changing the north- south exchange rate does not remove the initial asymmetry; rather it now extends the asymmetry north-south. In other words, it is not the energy price spike that causes the cross- border non-competitive- ness. Rather, it is the exchange rate response to the volatility of energy prices that is the culprit. A common FTA currency area would ensure that any exchange rate response to energy shocks would keep Windsor and Detroit, and Vancouver and Seattle, etc., on a level playing field.
Several observations-cum-conclu- sions are in order. First, the concern with exchange rate volatility is not intended as a criticism of the Bank of Canada. The choice of exchange rate regimes lies with the Government of Canada, not with the bank. And certainly it is not a criti- cism of Governor David Dodge, whose operational transparency and confi- dence-generating capacity have earned him the reputation of the world’s fore- most practitioner of floating-rate infla- tion targeting. However, it is intended as a questioning of Canada’s decision to opt for monetary independence and, therefore, floating exchange rates in an era where a) cost certainty is an impor- tant aspect of succeeding in FTA eco- nomic space and attracting foreign investment and b) severe exchange rate volatility and misalignment appear to be endemic and pervasive for a variety of reasons including the Dutch disease.
The second concluding remark is related: the exchange rate regime and monetary independence are policy instruments. Yet we have effectively raised them to the level of policy goals, and goals by their very nature become difficult to challenge or to subject to substantive analysis. But the goals of monetary policy transcend flexible rates and monetary independence to embrace the larger societal objectives. Even those that argue that inflation control is a goal will have to admit that fixed-rate systems can deliver this as well. By way of a relevant aside, The Economist indicates that 13 most devel- oped countries plus the Euro area all have inflation rates currently under 3 percent. It is also instructive to recall that the remarkable Pearson-era achievements across an incredibly wide range of social policy fronts occurred under Canada’s only post-1950s period of fixed exchange rates (or what is the same thing, the only period where we adopted US monetary policy). While the foregoing analysis is not sufficiently in- depth to make a convincing case for fixed exchange rates, hopefully it is suf- ficiently analytical to argue, first, that we should view monetary independence and flexible rates as instruments and not as goals and, second, we should assess anew and within the current envi- ronment whether they remain the pre- ferred macro policy instruments.
My view on this is surely clear by now: either we implement our mone- tary-independence/flexible-rate regime in ways that mitigate or otherwise accommodate the exchange rate volatil- ity or we must seriously consider fixing the exchange rate and becoming part of a larger currency area. In terms of the ”œunholy trinity,” this would mean embracing free trade/capital mobility and fixed exchange rates, and relying on Federal Reserve inflation targeting rather than Bank of Canada inflation targeting.
By way of a concluding comment, it is appropriate to return to the relation- ship between the loonie and the FTA. This article began by noting that it was the pre-FTA depreciation of the loonie and the resulting export penetration of the US market that brought the business sector on side and served as the catalyst for the FTA (and eventually NAFTA) and for market access to North America. If the 20 years of the loonie/FTA relationship are a prelude, then the scenario will be that swings in Canadian manufacturing unit labour costs will likely generate too much cost uncertainty for Canada to become the location of choice (for domestic and foreign investors alike) in terms of provisioning the North American market. It would be ironic indeed if this exchange rate volatility, and especially the rapid and dramatic appreciations of the loonie, become the mechanisms for curtailing effective access to FTA/NAFTA economic space for selected sectors of Canadian enterprise.