Oil price shocks: Sectoral and dynamic adjustments in a small-open developed and oil-exporting economy

ARTICLE IN PRESS

Energy Policy 38 (2010) 562–572

Contents lists available at ScienceDirect

Energy Policy

0301-42

doi:10.1

� Tel.:

E-m

journal homepage: www.elsevier.com/locate/enpol

Oil price shocks: Sectoral and dynamic adjustments in a small-opendeveloped and oil-exporting economy

Yazid Dissou �

Department of Economics, University of Ottawa Pavillon Desmarais, 55 Laurier Avenue East, Ottawa, Ontario, Canada K1N 6N5

a r t i c l e i n f o

Article history:

Received 12 December 2008

Accepted 2 October 2009Available online 31 October 2009

JEL classification:

C68

F11

F14

N72

Keywords:

Oil prices

Dynamic general equilibrium model

Canada

15/$ - see front matter & 2009 Elsevier Ltd. A

016/j.enpol.2009.10.008

+1613 562 5800�1686; fax: +1613 562 599

ail address: [email protected]

a b s t r a c t

The recent uptrend in oil prices represents both an opportunity and a challenge for small-open

developed and oil-exporting countries. Using Canada as a study case and in contrast to most studies that

use aggregate models, this paper employs a multi-sector, intertemporal general equilibrium model to

provide perspectives on the sectoral, aggregate and dynamic adjustments of a sustained increase in oil

prices. It highlights the transmission channels through which the rise in oil prices affects the domestic

economy. The simulation results suggest that the shock would have positive aggregate impacts, but

would also spur the reallocation of resources and would therefore induce disparities in sectoral

adjustments. The suggested contraction in some industries could not however be attributed to a pure

Dutch disease phenomenon because of, among other factors, the cost-push effect induced by the

increase in oil prices.

& 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Oil prices have increased dramatically in the past 5 years, withpricing benchmarks such as Brent and West Texas Intermediatecrude increasing from $30 US a barrel in early 2004 to more than$100 US a barrel in early 2008. Although, beginning in mid-2008,oil prices have experienced a recent decline because of the worldeconomic crisis, this movement is very likely to be temporary.This study analyzes the impacts of oil price increase on realeconomic activity in a small-open, oil-exporting country usingCanada as a case study. This country is one of the few net-exporters of natural resources among the OECD group ofcountries. Natural resource products represent a non-negligibleshare of Canadian total exports of goods, averaging 24% between2000 and 2004. Interestingly, however, the oil and gas industry(the most important natural resource industry) constitutes lessthan 5% of overall Canadian GDP, while energy-intensive indus-tries constitute a relatively larger share of the same. With such aneconomic composition, a rise in oil prices introduces an interest-ing trade-off for Canada.

While the impact of an increase in oil prices could in general bepositive because of the terms-of-trade improvement, energy-intensive industries could be adversely affected by the sameshock. Since oil is a primary source of energy in the economy, a

ll rights reserved.

9.

rise in its prices represents a classic supply-side shock, wherebyenergy-intensive industries would experience a rise in theirproduction cost. The extent of the negative impact on theeconomy would depend on the export intensity of the energy-intensive industries. In addition to the potential supply-sideshock, non-oil-producing industries could be affected by anappreciation of the real exchange rate that could eventually shiftresources from the affected sectors to the booming sectors. In thiscontext, it is interesting to mention that an increase in the worldprices of any other natural resource can lead to the samephenomenon, which is well documented and known in theliterature as the ‘‘Dutch disease’’. This term historically refers tothe change in the industrial structure in the Netherlands duringthe late 1950s and early 1960s following the discovery of naturalgas reserves. The adjustment was triggered by an appreciation ofthe real exchange rate that resulted in a booming natural gasexport sector while simultaneously leading to a contraction of theexport-manufacturing sector.1

The prospect of resource shifts from the manufacturing sectoris another potential source of concern for policy makers as far aslabour productivity growth is concerned. Indeed, when a boom inthe resource sector shifts resources from the manufacturing sectorthat is the most productive in the economy, aggregate labourproductivity growth may be reduced. This issue has been

1 See Corden and Neary (1982), Corden (1984), and Stevens (2003) for a review

of the concept of Dutch disease.

www.elsevier.com/locate/enpol

dx.doi.org/10.1016/j.enpol.2009.10.008

mailto:[email protected]

ARTICLE IN PRESS

Y. Dissou / Energy Policy 38 (2010) 562–572 563

addressed by several authors, including Gylfason et al. (1999),Krugman (1987), Matsuyama (1992), Rodrıguez and Sanchez(2005), Sachs and Warner (1995), and van Wijnbergen (1984),among others. They find that an export boom in the naturalresource sector could reduce productivity growth in the economy.

In light of the potential adverse impacts on the economystemming from a booming resource sector, it is worth emphasiz-ing that the development of Dutch disease is not systematic. Itsoccurrence and magnitude depend on the structure of the affectedeconomy, namely the degree of diversification and the nature ofinter-industry relationships. It is important to note that Dutchdisease will not pose a problem if the government provides anappropriate policy response to it. The latter depends on severalfactors that could not be discussed in this paper; see for exampleLartey (2008) and Matsen, and Torvik (2005) for some interestingdiscussions.

Several studies have been carried out in most OECD oil-exporting countries to assess the relevance of Dutch disease.2

While most of these studies find little evidence of Dutch disease,Stijns (2003) finds that the increase in energy exports can indeedaffect manufacturing exports. Using a gravity model, he finds thata 1% increase in energy net exports in an energy-exportingcountry can decrease its real manufacturing exports by 8%.

The main challenge of the studies that rely on econometricmethods is an endogeneity problem whereby the pure Dutchdisease effects cannot be separated from the negative impact onexports of an economic slow-down triggered by the increase in oilprices. In contrast, structural simulation models can overcomethis problem by running counterfactual simulations that keepexport demand constant. By doing so, they are able to assess theintrinsic impact of oil price changes on resource reallocation,irrespective of their impacts on economic activity in the rest of theworld. This consideration is very important for policy debate, as itis not clear whether adjustments in non-oil-producing industriesfollowing an increase in oil price are due to price changes alone orto the slow-down in economic activity in other countries as well.

While recent detailed analyses of the potential impacts of oilprice increases on the Canadian economy are scarce, theadjustment of the Canadian economy to a sustained increase inthe world oil prices deserves a careful empirical assessment. Thisanalysis is much needed in an international context where therising growth in demand for natural resource-based products inChina, India and other emerging economies raises the real pricesof these products. Such price increases could lead to a shift ofproductive resources from high-tech manufacturing and knowl-edge-based service industries in Canada. Consequently, a thor-ough understanding of the dynamic and sectoral adjustments ofthe economy in reaction to oil price increases will be an essentialelement in the design of effective policy responses.

The few interesting existing studies on the impacts of oil priceincreases in Canada use macroeconomic models that are morepreoccupied with the aggregate economic impacts than with thesectoral adjustment processes. See, for example, Bayoumi andMuhleisen (2006), Cunado, and Perez de Gracia (2003), Dib(2007), Hunt et al. (2002), and Jimenez-Rodriguez and Sanchez(2005). Similarly, a large body of research on the effects of oilshocks on the US economy also focuses on aggregate impacts only.Finn (2000), Hamilton (2003), Leduc and Sill (2004), Radchenko,(2005), and Rotemberg and Woodford (1996), among others, are afew examples. Since these models include only final goods and donot capture inter-industry transactions, they miss an important

2 See Stijns (2003) for an interesting review of these studies. The policy

response to a Dutch disease depends on several factors; see for example Lartey

(2008) and Matsen, and Torvik (2005) for some interesting discussions on that

issue.

channel through which changes in oil prices can affect theeconomy. Specifically, they are unable to capture adequately thecost-push effects of oil price increases.3

Multi-sector dynamic general equilibrium models can providea very good framework for the analysis of the potential impacts ofan increase in oil and other commodity prices. An interestingcharacteristic of these models is their ability to trace the impactson the entire economy of large policy changes in a particularindustry or sector. Resulting changes in the structure of con-sumption, production and trade could then be understoodcorrectly. General equilibrium models have been extensively usedin Canada and other OECD countries to analyze the potentialimpacts of policies affecting oil prices such as climate changepolicies. See McKibbin et al. (2000), and Rotemberg and Woodford(1996) for further references on this subject.

In this study, we use a multi-sector forward-looking dynamicgeneral equilibrium model of the Canadian economy to analyzethe potential short- and long-term impacts of the increase in theprices of oil in Canada. We analyze the intrinsic impacts of oilprice rises on aggregate and sectoral variables of interest such asGDP, household welfare, sectoral output, employment, invest-ment, imports and exports, prices and real exchange rate. Weconduct some simulations related to the increase in oil prices andperform sensitivity analyses of the results with regards to thevalues of behavioural parameters. We are not aware of any otherstudy that uses a multi-sector dynamic general equilibrium modelto assess the sectoral and growth implications of an increase in oilprices on a developed and oil-exporting country like Canada.

The remainder of the document is as follows. Section 2presents an overview of the structure of the Canadian economy.This section further briefly discusses the model characteristicsand the data. Section 3 analyzes the simulation results andSection 4 provides some concluding remarks.

2. A quick overview of the theoretical structure of the model

In this section, we present a thumbnail description of themodel. The model shares similar modeling philosophy with andincorporates some interesting contributions on multi-sectorintertemporal general-equilibrium modeling from Goulder et al.(1999), and Keuschnigg and Kohler (1994), among others.Although multi-sector intertemporal general equilibrium modelsare now commonly used in the literature, we believe that a shortdiscussion on the main components of the specific model used inthis study would be useful for a good understanding of thefollowing results.4

Eighteen industries and twenty commodities are considered, onthe supply side and the demand side of the economy, respectively.(See Table 1 for a listing of industries and commodities.) In contrastto one-sector macro-models, considering a sectoral disaggregationin analyzing oil price increases is useful, since it provides interestinginsights related to the sectoral adjustments led by the changes inrelative prices while simultaneously accounting for the inter-industry relationships and resource constraints.

Population growth rate as well as technological progress isassumed exogenous, while households and firms derive theirbehaviours from an explicit intertemporal optimization program.In addition to firms and households, government and the rest ofthe world are the other economic agents present in this model.

3 Other studies have analysed the industry impacts of oil price increase in the

US, by using econometric models. See Lee and Ni (2002) and Jimenez-Rodriguez

(2008) are some examples among several others.4 A full listing of the model equations is provided in the Appendix that is

available from the author upon request.

ARTICLE IN PRESS

Table 1Sectoral disaggregation in the model.

Industries Commodities

Agriculture Agriculture

Oil & gas Crude oil

Natural gas

Coal Coal

Other mining Other mining

Power generation Power generation

Gas pipelines Gas pipelines

Lumber Lumber

Wood industries Wood industries

Pulp & paper Pulp & paper

Paper manufacturing Paper manufacturing

Cement Cement

Iron & steel Iron & steel

Smelting Smelting

Chemicals Chemicals

Petroleum refining Petroleum refining

Other manufacturing Other manufacturing

Transport industries Transport industries

Services Services

Non-competitive imports

Y. Dissou / Energy Policy 38 (2010) 562–572564

Canada is considered as a small-open economy that produces bothtradable and non-tradable goods.

5 We chose a nested CES specification because of its regular flexibility property

as demonstrated in Perroni and Rutherford (1995).6 Constant elasticity of substitution.7 Constant elasticity of transformation.

2.1. Households

The representative household has preferences over consump-tion and leisure. It derives income from salaries, returns onfinancial assets and net transfers received from the governmentand the rest of the world. Transfers from the rest of the world areexogenous, while those received from the government areendogenous. The representative household pays sales taxes oncommodities and income taxes on returns to primary factors ofproduction.

The representative household maximizes an intertemporalutility function subject to a lifetime budget constraint. Theinstantaneous utility in the intertemporal utility function is alogarithmic function whose argument is a Cobb–Douglas aggre-gate of the consumption index and leisure as arguments. Throughthe intertemporal optimization process, the representative house-hold determines the optimal path of its consumption spending(consumption index), labour supply and saving. Within eachperiod, using a cost-minimizing rule, it allocates consumptionexpenditures across goods and services within a nested CES utilityfunction. The household preference representation allows for richsubstitution possibilities among commodities especially, on theone hand, between the aggregates of energy and non-energygoods and, on the other hand, among energy goods.

The three first-order conditions of household intertemporaloptimization problem give (i) the well-known Euler equation thatshows the relationship between consumption growth ratebetween two consecutive periods and the real interest rate; (ii)the usual trade-off between labour supply and consumption; and,(iii) the budget constraint that shows the motion law of householdfinancial wealth. Any shock that affects relative prices wouldchange the real interest rate and would thus alter the householdconsumption profile and its composition. In particular, a change inoil prices would trigger some substitution and income effects thatalter the composition of the representative consumer’s consump-tion basket. Detailed discussions on the transmission mechanismsof oil price changes on household behaviour will be discussedfurther in the paper.

2.2. Firms

The representative firm in each industry chooses its optimallevels of labour, intermediate inputs and investment so as tomaximize the firm’s stock market value, subject to bothtechnological and capital accumulation constraints, in the pre-sence of convex adjustment costs. Capital, labour, energy andmaterial inputs are combined to produce output using a constant-returns-to-scale technology, as represented by the nested CESfunctions.5 Akin to the household optimization behaviour, thespecified representation of firm technology allows interestingsubstitution possibilities among the inputs used.

Physical capital stock of the firm changes from one period toanother with capital formation, which accounts for new invest-ment, and physical capital depreciation. In a given period, thecapital stock is considered as given since it is inherited from initialcapital stock and past investment decisions. It follows that in anyperiod the firm will determine the optimal level of labour,intermediate inputs and investment in physical capital that willaffect next period’s physical capital stock. It is important to notethat the constant-returns-to-scale property of the technologydoes not imply a flat short-run supply curve because of capitalinstallation costs.

The forward-looking behaviour of the firm and the presence ofadjustment costs make the firm’s current demand for investmentgoods sensitive to the expectations of future changes in prices.Investment is a function of the marginal version of Tobin’s q andthe adjustment cost parameters.

A change in the relative prices, triggered by an increase in oilprices, would reduce the demand for investment in non-energysectors through the decline in the level of sectoral value addedand the increase in the price of capital goods. This change ininvestment would generate an interesting dynamics in theproduction sector, which is discussed further in Section 3.

The optimal values of the other production factors are chosenaccording to the common rule in static optimization problems thatconsists of equalizing the cost of the factor to its marginal product.

2.3. The government, trade and financial flows

The government derives revenue from taxes on commoditiesand on factor income, consumes goods and services, and enactslump-sum transfers to households. It is compelled to have abalanced budget in each period, achieved by adjusting transfers tohouseholds accordingly. Government real expenditures on com-modities are exogenous, increasing according to the populationgrowth rate including technological progress.

In line with most computable general equilibrium models, thepresent model distinguishes by their origins on both the demandand supply sides. More specifically, total domestic demand of eachcommodity is modeled as a CES-composite6 of imports anddomestic goods while the firm’s gross output is a CET-composite7

of exports and domestic sales. To account for the importance ofCanada’s trade relationship with the US, a second type ofdifferentiation is introduced between traded goods. In otherwords, a distinction is made between exports to the US andexports to the rest of the world (ROW) as well as between importsfrom the US and imports from the ROW.

Because Canada is considered a small country in the worldmarket, it considers foreign prices of imports and exports along

ARTICLE IN PRESS

Table 2Selected sectoral characteristics of the Canadian Social Accounting Matrix.

Source: Statistics Canada and author’s calculations.

Share in total domesticdemand

Share in total supply

Imports Domesticgood

Domesticsupply

Exports


with the world interest rate as given. The latter assumptionimplies that in each period, the net capital inflows to Canada areendogenous and they must offset any imbalance in the currentaccount. Still, the real exchange rate must adjust in order toachieve, over the long run, the sustainability of households’ netclaims over foreign assets. In other words, the real exchange ratewould adjust to avoid a Ponzi game where the country could lendor borrow forever.

Agriculture 11.0 89.0 81.4 18.6

Crude oil 57.7 42.3 36.1 63.9

Natural gas 0.0 100.0 53.9 46.2

Coal 79.5 20.6 23.2 76.8

Other mining 31.8 68.2 44.0 56.0

Power generation 2.5 97.5 87.3 12.7

Gas pipelines 6.7 93.4 69.1 30.9

Lumber 12.9 87.1 38.9 61.1

Wood industries 21.0 79.0 44.9 55.1

Pulp & paper 32.4 67.7 25.1 74.9

Paper manufacturing 30.6 69.4 73.5 26.5

Cement 20.0 80.0 76.8 23.2

Iron & steel 37.2 62.8 70.8 29.2

Smelting 55.7 44.4 27.4 72.6

Chemicals 56.3 43.7 49.7 50.3

Petroleum refining 14.1 85.9 80.1 19.9

Other manufacturing 67.5 32.5 35.3 64.7

Transport industries 12.1 87.9 81.1 18.9

Services 4.5 95.5 94.4 5.7

Non-competitive

imports

100.0 0.0 0.0 0.0

Table 3Selected industry characteristics of the Canadian Social Accounting Matrix.

Source: Statistics Canada and author’s calculations.

Share ofsectoral GDP atfactor cost ingross output

Share in sectoral GDP atfactor cost of

Labour Capital

Agriculture 38.1 41.5 58.5

Oil & gas 67.2 12.9 87.1

Coal 60.7 49.1 50.9

Other mining 56.2 40.8 59.2

Power generation 67.7 24.9 75.1

Gas pipelines 71.9 19.0 81.0

Lumber 34.4 53.3 46.7

Wood industries 37.7 53.8 46.2

Pulp & paper 37.5 40.3 59.7

Paper manufacturing 39.8 65.4 34.6

Cement 43.1 49.1 50.9

Iron & steel 33.7 63.8 36.3

Smelting 25.2 43.3 56.7

Chemicals 33.5 45.2 54.8

Petroleum refining 6.8 48.7 51.3

Other manufacturing 31.2 53.3 46.7

Transport industries 48.1 75.9 24.1

Services 56.0 69.1 30.9

Table 4Selected values of technology parameters used in the model.

Source: Various studies.

Substitution elasticities between

Aggregate of value-added-energy and aggregate of other inputs 0.5

2.4. Equilibrium conditions, data and calibration

In equilibrium, in addition to the requirement that, within anyperiod, all agents respect their budget constraints, domestic pricesand the wage rate adjust to achieve balance between the supply ofand demand for produced goods and labour. Moreover, on anintertemporal level, expected future prices must equal theirrealizations. The market clearing condition applied to the labourmarket assumes that the wage rate is flexible and that labour canmove freely among sectors.

The impacts of policy shocks analyzed in this model will bemeasured with respect to a base-run situation in which theeconomy is assumed to be in an initial steady state characterizedby a constant growth rate of 3.2% (population growth rate and therate of Harrod-neutral technological progress).

Calibration of the model is based on the structure of theCanadian economy as depicted by the social accounting matrix(SAM) in the year 2002 that we built. A SAM provides some usefulinformation on the structural interdependence of an economy byshowing the transaction flows between economic agents andproduction factors.

The year 2002’s SAM was built using that year’s data from theCanadian input–output table, national accounts, trade statisticsand government accounts. Among other characteristics, the SAMfeatures a sectoral disaggregation of the Canadian economy intoeighteen industries and twenty commodities. Tables 2 and 3provide some characteristics of the SAM that might be useful foran understanding of the results. Extraneous parameters requiredfor the model calibration consist of substitution elasticities forhousehold preferences, firm technology and Armington functionsas well as adjustment cost parameters. It is important to note that,since these extraneous parameters are point estimates of the trueunknown parameters most often taken from econometric studies,there is no single ‘‘correct’’ value for each of them. This situation,therefore, calls for some sensitivity analyses of results in order togauge the impact of uncertainty pertaining to the modelparameters.8 The assumed elasticities are deemed conservativeand are based on a literature search on elasticities in the Canadianand US economies. See Tables 4 and 5 for the values of selectedelasticities used in this study.9

As is usual in most computable general equilibrium models,the calibration of the model entails the use of the base-runsituation data and the extraneous behavioural parameters todetermine the values of the unobserved variables and otherparameters, so as to replicate the base-run steady-state equili-brium using the model in the absence of a shock. In thatequilibrium solution, all physical quantities should grow at theexogenous growth rate while relative prices remain unchanged.10

Labour and aggregate of capital and energy 0.8

Capital and total aggregate of energy inputs 0.5

Electricity and non-mobile fossil energy inputs 0.5

Fossil energy inputs 0.5

Refined petroleum products 0.5

Material inputs and mobile energy inputs 0.5

Mobile energy inputs 0.5

Capital adjustment cost parameter 4–12

8 It is noteworthy to recall the reader’s attention to the fact that this problem

is not inherent to the calibration method used in this study alone. It is common to

all analyses that rely on a point estimate of a true unknown parameter.9 The low values of trade elasticities are not surprising, given the level of

aggregation in the model. They are taken from Roland-Holst (1994).10 Stated differently, physical quantities, expressed per efficiency unit of

labour, should be constant in the steady state.

ARTICLE IN PRESS

Table 5Trade substitution elasticities.

Source: Annabi et al. (2003).

Elasticity of substitution between

Aggregate of imports anddomestic good

Aggregate of exports anddomestic good

Imports from the US andimports from the rest ofthe world

Exports to the US and exportsto the rest of the world

Agriculture 1.5 1.5 2.25 2.25

Crude oil 0.7 0.7 1.05 1.05

Natural gas 0.7 0.7 1.05 1.05

Coal 0.95 0.95 1.43 1.43

Other mining 0.95 0.95 1.43 1.43


Gas pipelines 0.7 0.7 1.05 1.05

Lumber 0.6 0.6 0.9 0.9

Wood industries 0.6 0.6 0.9 0.9

Pulp & paper 0.9 0.9 1.35 1.35

Paper manufacturing 0.9 0.9 1.35 1.35

Cement 1.1 1.1 1.65 1.65

Iron & steel 0.9 0.9 1.35 1.35

Smelting 0.8 0.8 1.2 1.2

Chemicals 0.9 0.9 1.35 1.35

Petroleum refining 0.7 0.7 1.05 1.05

Other manufacturing 0.5 0.5 0.75 0.75

Transport industries 1.5 1.5 2.25 2.25

Services 1.5 1.5 2.25 2.25


The calibration of this multi-sector intertemporal generalequilibrium model involves the dynamic aspects as well as thestatic aspects of the model. The methods used in this study followthe ones described in Goulder et al. (1999) as well as Keuschniggand Kohler (1994).

3. Simulations results

3.1. Description of the simulations

In this analysis, we abstract from oil price fluctuations, and arerather interested the effects of their long run trend. Our goal is toderive some insights about the sectoral and dynamic adjustmentsin the Canadian economy brought about by a sustained rise inworld oil prices. We report the results of one main simulationrelated to the increase in world prices of oil wherein we consider apermanent 20% increase in the price of crude oil and refinedpetroleum products. This increase would consist of an immediate20% rise in prices from the reference value to a new value thatwould henceforth remain constant.11 In addition to the mainsimulation, we run two other simulations to perform somesensitivity analyses in order to assess the robustness of thequalitative findings. In the first sensitivity analysis, we consider apermanent 40% increase in the world prices of oil products, whilein the second we assess the impacts of the values of tradeelasticity parameters.

Although numerical models such as the present one have theadvantage of being able to incorporate simultaneous interdepen-dencies that would have been otherwise impossible to consider inother analytical models, they often produce numerous andcomplex results that run counter to one’s initial intuition. Inorder to provide intuitive explanations of the results, we will focuson the main transmission channels at play and make an artificialdistinction in our discussions, firstly between the short- and long-

11 Several studies like Hunt et al. (2002) and Eastwood (1992) have also

considered a permanent increase in oil prices. The choice of 20% is arbitrary and is

done for illustration purpose only.

run impacts and secondly between aggregate and sectoralimpacts. To avoid unnecessary repetition, detailed explanationswill be provided for the main simulation alone.

We would also like to emphasize the fact that the resultspresented in this study are not forecasts; rather, they are theresults of counterfactual simulations, which, ceteris paribus,indicate the impacts on real economic activity of the price shocksconsidered. Finally, unless otherwise mentioned, all results areexpressed in percentage deviations from their base-run values;they are not percentage growth rates between periods.

3.2. Simulation 1: permanent 20% increase in the world (import and

export) prices of oil products

In this simulation, the trajectory of world prices (exports as wellas imports) of oil products is assumed 20% higher than the base-runvalues. Since Canada is a net exporter of oil products, this mayinitially seem to be only a positive terms-of-trade shock, whichwould in general lead to an expansion of the economy. A typicalresponse of the economy to this type of shock according to the‘‘Dutch disease’’ hypothesis, however, would be an export boom thatwould appreciate the real exchange rate by bidding the prices ofdomestic goods up and making imports more competitive.

This need not be the case in the present context. Indeed,because imports represent a non-negligible share of totaldomestic demand for oil products in Canada and increases in oilprices have a negative impact on production cost, the increase inworld prices of oil products would trigger other adjustments inthe economy as well. This would dilute the typical resultspredicted by the Dutch disease theory. Caution is thus called forin the interpretation of the potential implications of the increasein oil prices on an oil-exporting country that has a diversifiedmanufacturing sector, like Canada.

3.2.1. Aggregate impacts

Table 6 presents the aggregate impacts in this simulation whilethe graphs in Fig. 1 show the transitional dynamics of selectedvariables. As economic agents respond to relative prices in thepresent setting, the permanent increase in the world prices of oil

ARTICLE IN PRESS

Table 6Impacts of 20% permanent increase in world prices of oil on selected aggregate

variables.

Source: simulation results.

Percentage deviation from base-run.

After 1 year After 5 years After 10 years

GDP at market prices �0.2 �0.1 0.0

GDP at factor cost �0.3 �0.1 0.1

Petroleum industries 4.9 9.9 13.9

Primary resource industries �1.7 �2.9 �3.9

Manufacturing �1.6 �2.3 �2.7

Other industries �0.4 �0.3 �0.3

Employment �0.2 �0.2 �0.1

Household consumption 0.2 0.3 0.3

Consumption price index 1.3 1.2 1.2

Total real investment 0.9 1.0 1.0

Total real exports �1.1 �1.2 �1.1





Total real imports �0.1 �0.2 �0.3

Petroleum industries �7.5 �7.5 �7.4


Manufacturing 0.0 �0.2 �0.3

Other industries 1.8 1.8 1.7

Real exchange ratea�0.4 �0.3 �0.2

Measure of welfare change 0.24

a A positive sign corresponds to a depreciation of the real exchange rate.


products would trigger some income and substitution effects aswell as changes in the rate of capital accumulation. Aggregate realGDP at factor cost decreases in the first year by 0.3%, but increaseslater on and settles in the long run at a level that is 0.4% higherthan in the base run, as shown in Fig. 1.

Alongside the changes in real GDP at factor cost that reflectchanges in resource allocation in the productive sector (especiallyon income generation), we also provide the results of the impactson real GDP at market prices. The latter impacts, which reflectadjustments on the demand side (expenditures on final demand),are not necessarily identical to those of real GDP at factor costs ina particular year, as expenditures on goods are not based ondomestic income only. Domestic absorption is also affected by netforeign capital inflows. While changes in real GDP at marketprices follow the same pattern as those of GDP at factor cost,figures in Table 6 suggest that their magnitudes are lower becauseof, among other factors, households’ consumption smoothingbehaviour. Indeed, in the present model, households are assumedto consume not according to their current income but according totheir permanent income. Hence, they can smooth their consump-tion stream over time despite a temporary decline in their currentincome. Household real consumption increases in the first periodby 0.2% and, in the long run, settles at a level that is 0.3% higherthan in the base-run situation.

The increase in the world prices of oil products leads to apermanent increase in the consumption price index of about 1.2%.This change, in conjunction with a decrease in nominal wagerates, leads to a decrease in real wage rates, which, in turn, woulddecrease household labour supply. Still, the decline in laboursupply would be modest, as total employment would fall by about0.2% in the first period. Overall, household welfare increases, asevidenced by the positive welfare change of 0.24.12

12 The measure of welfare change used in this dynamic framework

encompasses the household’s entire lifetime. See Chari et al. (1995) for discussions

on this measurement.

As expected, the increase in the world prices of oil productsleads to an appreciation of the aggregate real exchange rate in thefirst period by 0.4% and by a lower magnitude in the subsequentperiods. Total real exports drop in the first year by 1.1%, continuingto decline until the value settles in the long run at a level that is0.8% lower than in the reference situation (Fig. 1). The pattern issimilar for total real imports, although with a smaller magnitude,for instance falling by only 0.1% in the first year. Finally, total realinvestment increases by 0.9% in the first year and increasesslightly in the transitional period before finally settling at a steadystate, which is 0.8% higher than in the base run.

In general, the magnitudes of the impacts on aggregatevariables reported so far are small; they are in line with themagnitude of the initial shock and are not significantly differentfrom the ones obtained in other (macro) models. For example, anIMF (2004) study, which used a one-sector multi-country dynamicgeneral equilibrium model, found that a permanent 20% increasein oil prices would have a negative impact on Canadian GDP (about�0.45%) in the first year. The magnitude of these aggregateimpacts may be misleading as they hide differences in sectoraladjustments and it should be observed that the increase in worldprices of oil products does not affect all sectors equally.

3.2.2. Sectoral impacts

Tables 7a–8b report the impacts on some relevant sectoralvariables in selected years. A discussion on the sectoral anddynamic impacts of the shock will shed some light on theseaggregate effects. On the supply side of the economy, oil priceincreases will induce some resource reallocation effects, withfactors moving from other sectors to oil-producing sectors. Asshown in Table 7a, in the short run, sectoral GDP, employment andreal investment increase in the oil & gas sector as well as inpetroleum refining industries while falling in all other sectorsexcept the power generation industry.13

As argued in the section describing the model, one of thekey advantages of using an intertemporal framework is the abilityto capture consistently the effects of changes in the values offuture variables on investment. The permanent increase in oilprices provides additional incentives to invest in oil industries asthe ratio of the shadow price of capital to its purchasing pricehas increased. In contrast, in the other industries, the increase inenergy cost reduces firms’ incentives to invest; capital formationdecreases in these sectors. As an illustration, investment in oil-producing industries increases by 24% in the first year (Table 7a) andsettles in the long run at a level that is 21% higher than in the baserun. In non-oil-producing industries, investment decreases in boththe short- and long-run. For example, in the first year, investmentfalls by 14.5% and 9.7% in the pulp & paper and smelting industries,respectively.

The adjustment in sectoral capital stocks, alongside thesectoral shift of labour towards oil industries, leads to a changein sectoral value added. GDP at factor cost increases by 5.8% and0.5%, respectively, in the oil & gas and petroleum refiningindustries while falling by 3.0% and 2.5% in the pulp & paperand smelting industries, respectively.

As a corollary to the shift of factors towards the oil industries,exports increase in these industries and fall in others. In reality, asindicated above, non-oil-producing industries are affected notonly from the resource shift towards oil industries, but also by anincrease in their production cost that is brought about by theincrease in oil prices. Consequently, their exports decrease further

13 Power Generation industry includes the production of hydroelectricity.

Through substitution effects, the rise in oil prices induces an increase of

hydroelectricity.

ARTICLE IN PRESS

0.2

0.4

0.6

Impact on GDP at fact cost of permanent 20% increase in world oil prices

0.4

Impact on aggregate real consumption of permanent20% increase

in world oil prices

-0.4

-0.2

00

% c

hang

e fr

om b

ase

run

Years

0

0.2

% c

hang

e fr

om b

ase

run

Years

1

1.2

Impact on aggregate real investment of permanent 20% increase in world oil prices

-0.2

0

Years

Impact on aggregate real exports of permanent 20% increase

in world oil prices

0.2

0.4

0.6

0.8

% c

hang

e fr

om b

ase

run

-1.2

-1

-0.8

-0.6

-0.4

% c

hang

e fr

om b

ase

run

0

Years-1.4

0

Impact on aggregate real imports of permanent 20% increase

in world oil prices0

Impact on aggregate real exchange rate of permanent 20% increase

in world oil prices

-0.4

-0.2

0

% c

hang

e fr

om b

ase

run

Years

-0.4

-0.2

% c

hang

e fr

om b

ase

run

Years

-0.6 -0.6

5 10 15 20 25 30 35 40 45 50

0 5 10 15 20 25 30 35 40 45 50

0 5 10 15 20 25 30 35 40 45 50 0 5 10 15 20 25 30 35 40 45 50

0 5 10 15 20 25 30 35 40 45 50

0 5 10 15 20 25 30 35 40 45 50

Fig. 1. Impacts of 20% permanent increase in world oil prices on various variables.


than would be expected if export prices of oil alone wereconsidered (i.e. without an increase in oil import prices).14 Theincrease in exports of oil products is not sufficient to counter thedecrease in foreign sales experienced by other industries. Thus,

14 In a non-reported simulation with a 20% increase in oil exports prices alone,

without import prices, we found that output contraction in non-oil industries is less

important.

total real exports fall as mentioned in the discussion on aggregateresults.

Regarding the demand side, an increase in the prices of oilproducts leads to a decrease in their domestic use by 3.0% and5.1% for crude oil and refined petroleum products in the first year,respectively (Table 8a). Despite the increase in householdconsumption, total domestic demand falls in non-oil-producingindustries, mainly because of the decline in demand for

ARTICLE IN PRESS

Table 7aImpact of 20% permanent increase in world prices of oil on selected sectoral

industry variables after 1 year.



Grossoutput

SectoralGDP

Employment Realinvestment

Agriculture �2.0 �2.5 �2.1 �5.7

Oil & gas 5.3 5.8 7.9 24.0

Coal �1.7 �2.1 �2.1 �6.8

Other mining �1.6 �1.9 �2.1 �6.0


Gas pipelines 0.2 0.1 0.7 5.2

Lumber �1.9 �1.6 �2.3 �6.0

Wood industries �1.1 �1.1 �1.5 �3.2

Pulp & paper �2.6 �3.0 �2.9 �14.5

Paper

manufacturing

�0.9 �0.9 �1.1 �3.4

Cement �0.8 �1.0 �1.0 �2.2

Iron & steel �1.5 �1.8 �1.8 �4.8

Smelting �2.1 �2.5 �2.6 �9.7

Chemicals �1.8 �2.3 �2.1 �4.9

Petroleum refining �2.5 0.5 2.0 5.1

Other

manufacturing

�1.4 �1.4 �1.6 �6.3

Transport

industries

�1.5 �2.1 �1.2 �2.7

Services �0.2 �0.3 �0.1 0.0

Table 7bImpact of 20% permanent increase in world prices of oil on selected sectoral

industry variables after 10 years.


Industries Percentage deviation from base-run.

Grossoutput

SectoralGDP

Employment Realinvestment

Agriculture �3.3 �3.5 �3.3 �4.2

Oil & gas 15.3 15.9 15.8 22.0

Coal �4.4 �4.4 �4.7 �6.8

Other mining �4.1 �3.9 �4.6 �6.4

Power generation �0.1 0.2 �0.4 �0.7

Gas pipelines 3.5 3.5 3.9 6.0

Lumber �4.2 �3.9 �4.4 �5.6

Wood industries �2.2 �2.0 �2.3 �2.9

Pulp & paper �6.2 �6.0 �6.3 �8.0

Paper

manufacturing

�1.9 �1.6 �1.8 �2.1

Cement �1.1 �1.0 �1.3 �1.5

Iron & steel �2.5 �2.3 �2.7 �3.4

Smelting �6.0 �6.0 �6.3 �9.0

Chemicals �2.3 �2.1 �2.6 �3.0

Petroleum refining �0.2 1.9 3.0 4.0

Other

manufacturing

�2.5 �2.5 �2.6 �3.1

Transport

industries

�1.5 �1.8 �1.2 �1.2

Services �0.1 �0.1 �0.1 �0.1


intermediate inputs. Apparently, output expansion in the oilindustry has not generated sufficient demand for intermediateinputs to counteract the decline in output experienced by non-oil-producing industries.

Since domestic demand is a composite of domestic goods andimports, it is interesting to note that imports are less affectedwhen total demand falls. For example, in the pulp and paperindustry, the decrease of 1.1% in total demand is achieved througha reduction of 1.5% in the demand for domestic goods and a

reduction in import demand of only 0.2%. Some industries, such ascement, experience an increase in their imports while exhibiting adecline in domestic sales. This result is not surprising if oneconsiders the fact that non-oil-producing industries (especiallyenergy-intensive industries) experience not only an increase intheir production cost, but also an appreciation of the realexchange rate.

As mentioned earlier, not all the typical effects of Dutchdisease are observed in this scenario since we do not observe aboom in the non-traded sector, which could be represented in thismodel by the services industry. The reason for this is that, inaddition to the increase in the prices of tradable goods, we find anegative shock that stems from the increase in oil prices (and thusin input prices for the non-traded sector).

3.3. Sensitivity analysis

In order to assess the robustness of the qualitative resultsdiscussed above, we performed some sensitivity analyses byrunning two additional simulations. In the first, we considered a40% increase in world oil prices instead of the 20% increasediscussed earlier. Table 9 reports the aggregate impacts of thissimulation. In general, aggregate variables move in the samedirection, though with greater magnitudes, as the previoussimulation. For example, GDP at factor cost falls more in theshort run (�0.8% versus �0.3% previously) but settles at a higherlevel a few periods later (0.3% versus 0.1% after 10 years). Theappreciation of the real aggregate exchange rate is more severe inthis simulation, resulting in a greater fall in the total real exports.Non-booming industries, as expected, also experience a largerdecrease in their output. Households, on the other hand, benefitmore from the higher oil revenue since they experience a higherconsumption stream leading to a welfare increase.

In the second sensitivity analysis, we respectively decreasedand increased the substitution elasticity parameters (in theArmington CES and CET function) by 25% in all industries.With these modifications, we conducted the simulation using a20% increase in world oil prices again. Table 10 reportsthe aggregate impacts of these simulations. While themagnitudes of the impacts are slightly lower and larger with,respectively, lower and higher values of elasticity, the qualitativeresults obtained for the impact of oil prices on the Canadianeconomy are still valid.

4. Conclusions

In this paper, we have investigated the potential effects of asustained increase in the world prices of oil products on theCanadian economy. We have used a multi-sector intertemporalgeneral equilibrium model that makes it possible to trace out theshort- and long-run adjustment of aggregate and sectoralvariables.

The simulation results suggest that oil shocks are beneficialto the Canadian economy since, because of the improvementin Canadian terms of trade, real GDP increases duringmost periods and the consumption profile is higher in all periods.Household’s welfare change is positive. For example, a permanent20% increase in world oil prices leads, in the long run, to a0.4% increase in GDP at factor cost in comparison to thereference situation (although this variable declines slightly inthe short run). The results also suggest that the magnitude ofthe long-run impact depends on the magnitude of the pricechange.

As expected, the increase in prices of these tradable goodswould shift resources (labour and capital) towards the export

ARTICLE IN PRESS

Table 8aImpact of 20% permanent increase in world prices of oil on selected sectoral trade variables after 1 year.



Total real supply Real exports Real domesticsupply

Real totaldomestic demand

Real imports Sectoral realexchange ratea

Agriculture �2.0 �3.4 �1.6 �1.4 0.2 �1.2

Crude oil 9.5 13.1 2.6 �3.0 �6.8 15.5

Natural gas 2.1 2.4 1.9 1.9 0.0 0.7

Coal �1.7 �1.9 �0.7 0.2 0.5 �1.3

Other mining �1.6 �1.6 �1.5 �1.6 �1.5 0.0

Power

generation

0.1 �2.3 0.4 0.5 3.2 �1.8

Gas pipelines 0.2 �1.4 0.9 1.0 3.2 �3.3

Lumber �1.9 �2.1 �1.6 �1.6 �1.1 �0.8

Wood industries �1.1 �1.4 �0.8 �0.7 �0.1 �1.1

Pulp & paper �2.6 �2.9 �1.5 �1.1 �0.2 �1.6

Paper

manufacturing

�0.9 �1.6 �0.7 �0.4 0.2 �1.0

Cement �0.8 �1.6 �0.6 �0.4 0.4 �0.9

Iron & steel �1.5 �1.9 �1.3 �1.1 �0.8 �0.6

Smelting �2.1 �2.2 �1.8 �1.7 �1.5 �0.5

Chemicals �1.8 �2.2 �1.3 �0.8 �0.5 �1.0

Petroleum

refining

�2.5 3.1 �4.0 �5.1 �10.6 11.7

Other

manufacturing

�1.4 �1.7 �0.8 �0.1 0.2 �1.9

Transport

industries

�1.5 �4.2 �0.9 �0.5 2.4 �2.2

Services �0.2 �1.9 0.0 0.0 1.9 �1.3


Table 8bImpact of 20% permanent increase in world prices of oil on selected sectoral trade variables after 10 years.


Percentage deviation from base-run

Total real supply Real exports Real domesticsupply

Real totaldomestic demand

Real imports Sectoral realexchange ratea

Agriculture �3.3 �5.1 �2.9 �2.7 �0.6 �1.5

Crude oil 20.7 26.7 8.7 �0.6 �6.7 23.5

Natural gas 11.1 15.3 7.2 7.4 0.0 9.8

Coal �4.4 �4.9 �2.6 �0.7 �0.2 �2.6

Other mining �4.1 �4.2 �3.8 �3.8 �3.4 �0.4

Power

generation

�0.1 �1.0 0.1 0.1 1.1 �0.7

Gas pipelines 3.5 2.1 4.1 4.3 6.2 �2.9

Lumber �4.2 �4.7 �3.4 �3.4 �2.0 �2.4

Wood industries �2.2 �2.9 �1.5 �1.2 0.0 �2.4

Pulp & paper �6.2 �7.0 �3.6 �2.4 0.0 �4.2

Paper

manufacturing

�1.9 �3.2 �1.4 �0.9 0.3 �2.0

Cement �1.1 �2.1 �0.8 �0.6 0.5 �1.2

Iron & steel �2.5 �3.1 �2.2 �1.8 �1.4 �1.0

Smelting �6.0 �6.5 �4.7 �3.7 �2.9 �2.4

Chemicals �2.3 �2.8 �1.8 �1.3 �0.9 �1.1

Petroleum

refining

�0.2 7.6 �2.3 �3.7 �11.3 15.4

Other

manufacturing

�2.5 �3.0 �1.5 �0.6 �0.1 �3.0

Transport

industries

�1.5 �4.1 �0.9 �0.5 2.4 �2.2

Services �0.1 �1.8 0.0 0.1 1.9 �1.3



booming sectors and lead to an appreciation of the real exchangerate, which in turn may not benefit traditional manufacturingexports. Contrary to prior studies that used one-sector models,results from this multi-sector analysis suggest that not all

industries will be affected in similar ways. A permanent 20%increase in world oil prices would thus be beneficial to oil-producing industries while it would not benefit the manufactur-ing industries.

ARTICLE IN PRESS


Still, caution should be wielded before jumping to theconclusion that an increase in oil prices would create a pureDutch disease phenomenon in Canada. While this study heldconstant export prices for Canadian goods in the non-booming

Table 9Sensitivity analysis-Impacts of 40% permanent increase in world prices of oil on

selected aggregate variables.



After 1 year After 5 years After 10 years

GDP at market prices �0.5 �0.3 �0.2

GDP at factor cost �0.8 �0.2 0.3





Employment �0.6 �0.5 �0.3

Household consumption 0.9 1.0 1.0

Consumption price index 3.1 3.0 3.0

Total real investment 2.1 2.5 2.4

Total real exports �3.0 �3.1 �2.9





Total real imports 0.3 �0.1 �0.3

Petroleum industries �13.0 �12.7 �12.4


Manufacturing 0.3 �0.1 �0.4

Other industries 4.6 4.5 4.6

Real exchange ratea�1.4 �1.3 �1.2

Measure of welfare change 0.70


Table 10Impacts of 10% permanent increase in world prices of oil on selected aggregate variabl



After 1 year

Low elasticities Base elasticities High

GDP at market prices �0.16 �0.18 �0.

GDP at factor cost �0.27 �0.32 �0.

Petroleum industries 4.42 4.90 5.

Primary resource industries �1.70 �1.74 �1.

Manufacturing �1.46 �1.62 �1.

Other industries �0.34 �0.41 �0.

Employment �0.17 �0.20 �0.

Household consumption 0.19 0.22 0.

Consumption price index 1.31 1.26 1.

Total real investment 0.72 0.88 1.

Total real exports �0.96 �1.10 �1.

Petroleum industries 5.83 6.54 7.


Manufacturing �1.66 �1.84 �1.

Other industries �1.97 �2.46 �2.

Total real imports �0.09 �0.07 �0.

Petroleum industries �6.65 �7.48 �8.


Manufacturing 0.04 0.04 0.

Other industries 1.39 1.80 2.

Real exchange ratea�0.44 �0.38 �0.

Base elasticities’ refers to the simulation with base elasticity values.

Low elasticities’ refers to the simulation with 25% lower than base elasticity values

Base elasticities’ refers to the simulation with 25% higher than base elasticity values


sector, the reported impacts on sectoral variables do account forthe intersectoral reallocation of resources brought about by thecost-push effect of the increase in oil prices. The decline in exportsexperienced by traditional manufacturing industries, especiallythe energy-intensive ones, could accordingly not be attributedpurely to a Dutch disease effect.

Finally, it is worth calling the reader’s attention to the fact thatthe results reported in this study are not forecasts. They are rathercounterfactual simulation results that were obtained whilekeeping several variables constant. For instance, the centralbank’s reaction to price increases (which may be critical toeconomic agents’ behaviour) has not been modeled here. More-over, the induced technological change and the innovationprocesses that could be triggered by the increase in oil pricesand potential supply bottlenecks in resource industries in Canadahave not been considered in this analysis. Finally, the model useddid not account for rigidities in the labour market and for capitalutilization rates. The actual figures of the impact of oil priceincreases on economic activities may therefore be different fromthe ones presented in this study.

Acknowledgements

I would like to thank Serge Coulombe, Patrick Georges,Madanmohan Gosh, David Gray, Someshwar Rao and seminarparticipants at Industry Canada, Bank of Canada, the 47e Congr�esde la Societe Canadienne de Science Economique, the EcoModInternational Conference on Policy Modeling, and two anonymousreferees for helpful discussions and suggestions. Useful commentsand assistance from Chahreddine Abbes, Faisal Deen Arif, MeganBaker, Xiujun Li and James Mancini are also acknowledged. Usualcaveats apply.

es: sensitivity analysis on trade substitution elasticities.

After 5 years

elasticities Low elasticities Base elasticities High elasticities

20 �0.09 �0.10 �0.11

36 �0.07 �0.08 �0.10

25 8.41 9.94 11.27

75 �2.68 �2.90 �3.09

73 �1.96 �2.31 �2.62

46 �0.25 �0.31 �0.37

23 �0.14 �0.17 �0.19

26 0.23 0.26 0.29

21 1.27 1.22 1.18

02 0.78 0.98 1.18

22 �1.01 �1.18 �1.32

09 10.80 13.01 15.00

89 �3.09 �3.35 �3.55

98 �2.41 �2.84 �3.21

91 �1.97 �2.49 �2.98

03 �0.22 �0.24 �0.24

06 �6.44 �7.48 �8.27

43 �2.14 �2.40 �2.62

05 �0.12 �0.15 �0.18

20 1.35 1.75 2.15

34 �0.34 �0.29 �0.25

ARTICLE IN PRESS


Appendix. Supplementary data

Supplementary data associated with this article can be foundin the online version at doi:10.1016/j.enpol.2009.10.008.

References

Annabi, N.B. Decaluwe, Lemelinm, A., 2003. Formes fonctionnelles et parametrisa-tion dans les mod�eles d’equilibre general calculables (MEGC): Revue de lalitterature pour le mod�ele d’equilibre general du minist�ere des Finances duQuebec (MEGFQ), Finances Quebec, Travaux de recherche 2003–03.

Bayoumi and Muhleisen, 2006. Energy, the exchange rate, and the economy:macroeconomic benefits of Canada’s oil sands production. IMF WP/06/70.

Chari, V.V., Lawrence, J. Christiano, Patrick, J. Kehoe, 1995. Policy analysis inbusiness cycle models. In: Cooley Thomas, F. (Ed.), Frontiers of Business CycleResearch. Princeton University Press, pp. 357–391.

Cunado, J., Perez de Gracia, F., 2003. Do oil price shocks matter? Evidence for someEuropean Countries. Energy Economics 25 (2), 137–154.

Corden, W.M., Neary, J.P., 1982. Booming sector and de-industrialization in a smallopen economy. The Economic Journal 92 (368), 825–848.

Corden, W.M., 1984. Booming sector and Dutch disease economics: survey andconsolidation. Oxford Economic Papers 36, 359–380.

Dib, Ali, 2007. Welfare effects of commodity price and exchange rate volatilities ina multi-sector SOE Model. Paper presented at the 41st CEA Annual Meetings,Halifax, Canada.

Eastwood, R.K., 1992. Macroeconomic impacts of energy shocks. Oxford EconomicPapers 44, 403–425.

Finn, M., 2000. Perfect competition and the effects of energy price increases oneconomic activity. Journal of Money, Credit, and Banking 32 (3), 400–417.

Goulder, L.H., et al., 1999. The cost-effectiveness of alternative instruments forenvironmental protection in a second-best setting. Journal of Public Economics72, 329–360.

Gylfason, T., Herbertsson, T.T., Zoega, G., 1999. A mixed blessing: natural resourcesand economic growth. Macroeconomic Dynamics 3, 204–225.

Hamilton, J., 2003. What is an oil shock?. Journal of Econometrics 113 (2), 363–398.Hunt, B., Isard, P., Laxton, D., 2002. The macroeconomic effects of higher oil prices.

National Institute Economic Review 179 (1), 87–103.IMF, 2004. GEM: A New International Macroeconomic Model, International

Monetary Fund, Washington, DC.Jimenez-Rodriguez, R., Sanchez, M., 2005. Oil price shocks and real GDP growth:

empirical evidence for some OECD countries. Applied Economics 37, 201–228.

Jimenez-Rodriguez, R., 2008. The impact of oil price shocks: evidence from theindustries of six OECD countries. Energy Economics 30, 3095–3108.

Krugman, P., 1987. The narrow moving band, the Dutch disease, andthe competitive consequences of Mrs. Thatcher: notes on trade in thepresence of dynamic scale economies. Journal of Development Economics 37,41–55.

Keuschnigg, C., Kohler, W., 1994. Modeling intertemporal general equilibrium: anapplication to Austrian commercial policy. Empirical Economics 19 (1),131–164.

Lartey, E.K.K., 2008. Capital inflows, Dutch disease effects, and monetary policy in asmall open economy. Review of International Economics 16, 971–989.

Leduc, S., Sill, K., 2004. A quantitative analysis of oil-price shocks, systematicmonetary policy, and economic downturns. Journal of Monetary Economics 51(4), 781–808.

Lee, K., Ni, S., 2002. On the dynamic effects of oil price shocks: a study usingindustry level data. Journal of Monetary Economics 49, 823–852.

Matsen, E., Torvik, R., 2005. Optimal Dutch disease. Journal of DevelopmentEconomics 78 (2), 494–515.

Matsuyama, K., 1992. Agricultural productivity, comparative advantage andeconomic growth. Journal of Economic Theory 58, 317–334.

McKibbin, W.J., et al., 2000. Emissions trading, capital flows and the Kyotoprotocol. Energy Journal 21 (Suppl.), 287–334.

Perroni, C., Rutherford, T.F., 1995. Regular flexibility of nested CES functions.European Economic Review 39 (2), 335–343.

Radchenko, S., 2005. Oil price volatility and the asymmetric response ofgasoline prices to oil price increases and decreases. Energy Economics 27 (5),708–730.

Roland-Holst, D.W., Reinert, K.A., Shiells, C.R., 1994. A general equilibrium analysisof North American economic integration. In: Francois, J.F., Shiells, C.R. (Eds.),Modeling Trade Policy: Applied General Equilibrium Assessments of NorthAmerican Free Trade. Cambridge University Press, New York, pp. 47–82.

Rodrıguez, R.J., Sanchez, M., 2005. Oil price shocks and real GDP growth: empiricalevidence for some OECD countries. Applied Economics 37, 201–228.

Rotemberg, J.J., Woodford, M., 1996. Imperfect competition and the effects ofenergy price increases on economic activity. Journal of Money, Credit andBanking 28, 549–577.

Sachs, J.D., Warner, A.M., 1995. Natural resource abundance and economic growth.National Bureau of Economic Research Working Paper No. 5398, Cambridge,MA.

Stevens, P., 2003. Resource impact: curse or blessing? A literature survey. Journal ofEnergy Literature 9 (1), 3–42.

Stijns, 2003. An Empirical Test of the Dutch Disease Hypothesis Using GravityModel of Trade, Mimeo, University of California at Berkeley.

van Wijnbergen, S., 1984. The ‘Dutch disease’: a disease after all?. Economic Journal94, 41–55.

doi:10.1016/j.enpol.2009.10.008

Documents

Oil price shocks: Sectoral and dynamic adjustments in a small-open developed and oil-exporting economy