asian development bank- the challenge of higher oil prices

8/12/2019 Asian Development Bank- The Challenge of Higher Oil Prices

1/21

Te challenge of higher oil prices

Adjusting to higher oil prices: The challenge fordeveloping Asia

O

il prices have risen higheragain.

Benchmark Brent crude oil prices haveaveraged $53 per barrel (/bbl) in 2005

through 31 August. Prices climbed toward the$70 level in late August, or nearly three quartersas high again as at the beginning o 2005. Noall in prices seems imminent: oil price uturesor end-December 2005 and end-December 2006delivery are about $67/bbl (Figure 3.1). Tesedevelopments were not expected. Te AsianDevelopment Outlook 2005, released in April thisyear, assumed an average $41/bbl or 2005 and$39 or 2006.

Te region o developing Asia and thePacific is potentially vulnerable to high oilprices. It is a large net importer o oil (in thissection oil is taken to include petroleum energyproducts excluding natural gas) and much oits rapidly expanding energy needs are met byoil. Developing Asia produces about 11% othe worlds crude oil, but consumes more than20% o it, and this gap is widening. Economiesin developing Asia are nearly as oil intensivein energy consumption and much less energyefficient than most industrial countries. For

each unit o gross domestic product (GDP),measured at market exchange rates, developingAsia consumes nearly five times as much energyas Japan and nearly three times as much as theUnited States (US).

Despite its dependency on oil and a threeoldincrease in nominal oil prices since 2003, theregion has perormed well economically. But pastresilience does not mean that developing Asiais immune to high oil prices. Signs o stress are

indeed starting to surace: inflation is creepingup; uel subsidies are beginning to cast a largeshadow over fiscal prospects in some places; andhigh oil prices may become a prominent actor

that will urther prolong the regions generallyanemic investment demandoutside the PeoplesRepublic o China (PRC)that has prevailed sincethe Asian crisis.

Sustained high oil prices will requirepolicy responses, and the ingredients o theseresponses will vary among countries. In manyoil-importing economies, fiscal and monetaryadjustments will be needed to stabilize impactson prices and output. Where oil consumptionis subsidized, higher prices raise questionsabout the affordability and objectives o oil

price subsidies. For poor countries with limitedborrowing capacity, higher import uel billscould present financing difficulties. But net oilexporters also ace challenges. Governmentsthere will need to consider how best to use the

Figure 3.1 Brent futures prices,

January 2004August 2005

$/bbl

20

30

40

50

60

70

December 2006 deliveryDecember 2005 delivery

AugJunAprFeb2005

DecOctAugJunMarJan2004

Source:Datastream, downloaded 1 September 2005.


2/21

66 Asian Development Outlook 2005 Update

additional revenues generated by oil and, ihigher prices persist, how to manage pressuresor exchange rate appreciation. Over the longerterm, governments across developing Asia will

need to take decisions about the role o oil inmeeting their countrys energy needs. Policychoices need to be guided by a ramework thatpromotes greater energy efficiency and environ-mental sustainability o growth.

Tis part o the Update reviews the possibleconsequences and challenges presented by highoil prices or developing Asia. Afer a brie reviewo why prices are high, their impact on variouseconomies in developing Asia is examined. Policyresponses to structurally higher oil prices arethen surveyed, including a brie discussion on

the risks inherent in subsidizing oil products.Te review ends by setting out key principlesor guiding policy decisions, and offering someconclusions.

Why are oil prices so high?

Figure 3.2 shows nominal and real prices orBrent crude or the past three and a hal decades.Although oil prices continue to set new recordsin nominal terms, in real terms they remainwell below the peak established in the oil shock

o 1979, having generally fluctuated withina broad band o about $2040/bbl. At about$6070/bbl, todays prices are high compared totheir historical average, though they would haveto rise by another $3545/bbl to hit the peak o$107/bbl (2005, first-hal inflation-adjusted) seenduring the month o November 1979 at the timeo the Iranian revolution. Te peak annual averageprice, also in 1979, is $83/bbl (2005, first-halinflation-adjusted), a level closer to todays prices.Te increases in nominal oil prices seen duringthe recent run-up have also been more modestand gradual than the earlier shocks.

O course, oil prices reflect underlying unda-mental orces o demand and supply, and thedemand or oil has seen steady growth, largelypropelled by Asias strong economic perormance.For example, between 1990 and 2003, or theworld as a whole, annual demand or oil grew at1.3%, while or the PRC and India combined itexpanded at 7%. ogether, these two countrieshave accounted or almost 40% o the growth in

demand since 1990. Te impact o rising incomeson oil demand in these two Asian giantstheirincome elasticity o demand or oil is thought tobe about 50% higher than in the rest o the world(Verleger 2005)has been magnified by theircomparatively inefficient use o energy.

Despite substantial increases in oil prices,demand remains robust. Driven by still-stronggrowth in the US and developing Asia, globaloil demand reached 82.8 million barrels per day

(mb/d) in the first hal o 2005, or an increase o1.3 mb/d rom the same period in 2004 (thoughthis increase is substantially less than the surgein the first hal o 2004). For the entire year,demand is now projected to average 83.7 mb/d.As a whole, the increase in developing Asiandemand accounted or nearly hal o globaldemand expansion in 2004. Although oil demandis expected to rise more slowly between 2005 and2006, it will remain robust, with projected growthin the range o 1.52.1 mb/d (depending on theorecasting agency).

Shorter-run influences are also at work onprices. In the ace o severe capacity constraints,refiners have joined the drive to increaseoperating inventory levels. In the first hal o2005, OECD stocks rose to 54 days o orwardconsumption, compared with an average o51 days since the second hal o 2002. Mostcountries lifed their inventories as a consequenceo tight and volatile supply. For much o 2005utures prices have exceeded spot prices, helping

Figure 3.2 Brent crude oil prices, 19702005

$/bbl

0

20

40

60

80

100

120

Real (H1 2005 prices)Nominal

20052000199519901985198019751970

Sources: International Financial Statistics online database,available: http://ifs.apdi.net/imf/ifsbrowser.aspx?branch=ROOT;

Datastream, downloaded 1 September 2005; US Bureau ofLabor Statistics, available: www.bls.gov.
http://ifs.apdi.net/imf/ifsbrowser.aspx?branch=ROOThttp://www.bls.gov/http://www.bls.gov/http://ifs.apdi.net/imf/ifsbrowser.aspx?branch=ROOT


3/21

The challenge of higher oil prices 67

maintain upward pressure on spot prices byreducing the cost o carrying inventories.

Investment in refining capacity has been toolow, and a mismatch has emerged between the

type o refining capacity now required and whatis available. For some time, world oil demandhas been driven by high-quality light crude (oilo low density or containing a low wax content,which makes production and refining easier) andby sweet crude (oil with a low sulur content).Recent additions to production capacity have,though, largely been in the heavy and sourgrades o crude, which are more difficult andcostly to refine.

Tis lack o investment in appropriate refiningcapacity and limited substitution possibilities has

pushed retail prices up. Since the value storedin a barrel o crude rises when final productprices rise, higher retail prices also help lif crudeprices. For example, during July and August,higher gasoline and diesel prices caused byrefinery outages in the US caused those refineriesstill operating to bid up the price o light, sweetcrude so that they could profit rom the highretail prices.

In the first hal o 2005, world supplyincreased to 84.1 mb/d, up by 1.7 mb/d on 2004slevel. At that time OPECs spare capacity had

been reduced to about 2.2 mb/d. However, oncecountries that are prone to supply disruptions,such as Iraq, Nigeria, and Venezuela, are excluded,spare capacity is a meager 1.4 mb/d. Given thisnarrow buffer, events that threaten to disruptsupply are now transmitted very quickly to prices.For instance, an early start o the hurricane seasonthis year along the US Gul Coast was the mainculprit or the price surge in July, while anxietiesover Irans resumption o nuclear activities andears o terrorist attacks on Saudi Arabia lifed theprice urther in early August. Hurricane Katrinapushed up the price in early September.

Looking ahead, there are proven oil reservessufficient to cover current global consumptionneeds or over 40 years. But investment in oilproduction, refining, and distribution inra-structure has been paltry ollowing a protractedperiod o low prices through the late 1980s and1990s. Te oil industry is now moving rom anexploitation phase to an investment phase. In thischanging environment, the rise in long-dated oil

prices reflects expectations o higher long-runmarginal production costs. Long-dated pricesalso incorporate a premium linked to financialrisks. Actual costs are a unction o project

complexity, host-country policies, and a range oother actorsincluding the supply o equipmentand skilled labornone o which is known withmuch certainty. Recent reports o very large costoverruns or the Sakhalin 2 liquefied natural gasproject in Russia and the Athabasca oil sandsproject in Canada vividly illustrate the financialrisks and the difficulties o investment planning.Although the recent surge in long-dated oil pricesmakes investment potentially attractive again,investment o current strong cash flows into newoil production projects remains slow. Investors are

delaying decisions in the ace o cost uncertainty,and new sources o supply will come onstreamonly gradually.

Higher oil prices are expected to stay or theremainder o 2005 and through 2006. A recentstudy by Goldman Sachs projects that oil pricesare likely to be sustained at over $60/bbl overthe period 20062010 (Goldman Sachs 2005).Afer a long period o both low prices and lowinvestment, binding constraints are being eltalong the length o the supply chain. ogetherwith undamental tightness in the current crude

oil supply/demand balance, there are also severalsignificant risks that could cause prices to riseurther or to spike, including robust globaldemand (and unpredicted surges in demand roma large country such as the PRC), weather- andaccident-related disruptions, and heightenedgeopolitical uncertainties.

Why high oil prices matter

As developing Asia consumes more oil than itproduces, higher oil prices are likely to eat intoits income growth. By how much will dependon the extent to which oil prices rise and howlong they remain elevated. For net oil-importingcountries, the impact will depend on a range oactors, including their oil and energy intensityand the ease with which needed adjustments takeplace. For net oil-exporting countries, higher oilprices raise oil sector profits but these benefitsare ofen highly concentrated and can be offsetby negative effects elsewhere. o understand


4/21


possible impacts, it is useul first to look at howdeveloping Asias economies depend on oil.

Oil and energy dependency in Asia

able 3.1 profiles developing Asias reliance onoil in 2003. Dependency is measured in ourways, using five indicators: oil sel-sufficiency;intensity o oil use in energy consumption;energy intensity o GDP, both at market and atpurchasing power parity exchange rates; and percapita oil consumption. Te oil sel-sufficiencyindex measures oil production less consumptionin relation to oil consumption. Tus a value o-1 signifies that a country has no oil productionand is totally reliant on oil imports; a positivenumber means that a country is a net exporter.

Te intensity o oil use in energy consumptionindex measures the share o oil in an economysprimary energy consumption. I a country reliesonly on oil to produce energy, the value o theindex is 1; i no oil is used in producing itsenergy, the value is 0.

Te third and ourth indicators show ameasure o the energy intensity or an entireeconomy (energy consumption divided by GDP).Tis measure is standardized on the energyintensity o the G7 countries. For example, a valueo 2 would mean that the country in question

uses twice the energy as the G7 average perunit o GDP. Tis measure is presented or both

nominal GDP calculated at market exchange ratesand or purchasing power parity-adjusted GDProm the World Economic Outlookdatabase othe International Monetary Fund (IMF). Te fifh

indicator simply divides annual oil consumptionin barrels by a countrys population.

Developing Asia shows considerable diversityin oil sel-sufficiency (able 3.1, Oil sel-suffi-ciency column): several countries are net oilexporters, but many more are totally reliant onoil imports. In addition, its reliance on importedoil has trended up through time: in 2003, 44.7%o oil consumption was imported, compared with

just about 10% in the mid-1980s (Figure 3.3).At the subregional level, South Asia is the

most reliant on imports ollowed by East Asia;

Southeast Asia has also become a net importeras Indonesias production has ailed to keep pacewith consumption. Central Asia and the Pacificare net oil exporters, though the position o thePacific masks the complete reliance on importso all countries but Papua New Guinea (andimor-Leste, which is not included in the Inter-national Energy Annual 2003figures due to lacko data). In Central Asia, the Kyrgyz Republic andajikistan are highly reliant on imports.

Vulnerability to rising oil prices depends notjust on oil sel-sufficiency but also on the intensity

with which oil is used to produce energy. Inthe mid-1980s, oil met about 30% o developing

Figure 3.3 Oil self-sufficiency index, 19802003

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

Developing Asia

South AsiaSoutheast Asia

East Asia

200320011998199519921989198619831980

Notes: 1. The oil self-sufficiency index is oil production less consumption, divided by consumption. No domestic oil production is equal

to -1.0. 2. Prior to 1992, Developing Asia excluded countries in Central Asia. 3. Before 1992, all Pacific countries were net oil importers; inthat year, Papua New Guinea became a net oil exporter.

Source: Energy Information Administration, International Energy Annual 2003, available: www.eia.doe.gov.

-1.0

0.0

1.0

2.0

3.0

4.0

The Pacific Central Asia

20032001199819951992
http://www.eia.doe.gov/http://www.eia.doe.gov/


5/21


Table 3.1 Oil and energy use, developing Asia, 2003

Subregion/Economy Oil self-

sufficiency

Intensity of oil

use in energy

consumption

Energy intensity of GDP

Nominal PPP

Oil consumption per

capita (barrels)

East Asia -0.600 0.310 3.188 0.907 2.4China, Peoples Rep. of -0.361 0.250 4.259 0.877 1.6

Hong Kong, China -1.000 0.628 0.739 0.573 13.9Korea, Rep. of -0.999 0.520 1.891 1.138 16.5

Mongolia -1.000 0.257 10.453 2.677 1.8

Taipei,China -0.991 0.457 1.939 0.971 14.8

Southeast Asia -0.277 0.546 2.684 0.821 2.6Cambodia -1.000 0.932 0.263 0.039 0.1Indonesia 0.074 0.507 2.624 0.801 2.0

Lao Peoples Dem. Rep. -1.000 0.117 3.149 0.608 0.2

Malaysia 0.648 0.445 2.959 1.205 7.5Myanmar -0.512 0.364 2.496 0.332 0.2

Philippines -0.957 0.550 2.125 0.446 1.5

Singapore -0.988 0.888 2.523 2.143 60.5

Thailand -0.685 0.529 2.898 0.840 4.8Viet Nam 0.632 0.460 3.304 0.648 1.0

South Asia -0.690 0.352 3.072 0.573 0.7Afghanistan -1.000 0.533 - - 0.1

Bangladesh -0.919 0.288 1.553 0.301 0.2Bhutan -1.000 0.121 3.866 0.972 0.5

India -0.649 0.343 3.230 0.588 0.8

Maldives -1.000 1.000 1.610 0.480 5.0Nepal -1.000 0.500 1.363 0.225 0.2

Pakistan -0.817 0.383 3.439 0.729 0.8

Sri Lanka -1.007 0.846 1.425 0.338 1.5

Central Asia 1.811 0.210 13.175 3.456 3.5Azerbaijan 1.664 0.415 11.789 2.714 5.5Kazakhstan 3.689 0.216 8.987 2.654 5.4

Kyrgyz Republic -0.819 0.121 12.863 2.570 0.8

Tajikistan -0.986 0.190 21.996 4.683 1.4Turkmenistan 1.542 0.224 9.143 2.867 6.0

Uzbekistan 0.015 0.148 32.550 6.225 2.2

The Pacific 0.667 0.742 1.690 0.513 1.5Cook Islands -1.000 1.000 - - 7.9Fiji Islands -1.000 0.754 1.590 0.722 4.4

Kiribati -1.000 1.000 0.829 0.234 0.8

Nauru -1.000 1.000 - - 27.7

Papua New Guinea 2.366 0.685 1.813 0.459 1.0Samoa -1.000 0.768 1.179 0.321 2.1

Solomon Islands -1.000 1.000 1.542 0.412 1.0

Tonga -1.000 1.000 1.324 0.282 2.9Vanuatu -1.000 1.000 0.619 0.245 1.1

Developing Asia -0.447 0.346 3.227 0.847 1.7Non-oil exporters -0.654 0.342 3.118 0.805 1.7

Memorandum itemsG7 -0.591 0.403 1.000 1.000 18.6Japan -0.978 0.505 0.692 0.796 16.0

United States -0.561 0.395 1.192 1.153 25.1

- = data not available, PPP = purchasing power parity.

Notes:1. The oil self-sufficiency index is oil production less consumption, divided by consumption; a positive number indicates self-

sufficiency. No domestic oil production is equal to -1.0. 2. Intensity of oil use in energy consumption is petroleum consumption divided

by energy consumption. 3. Energy intensity of GDP, for both nominal GDP (at market exchange rates) and GDP measured at purchasing

power parity, is expressed relative to the average of the G7 countries, which is normalized to 1.

Sources:Energy Information Administration. 2005. International Energy Annual 2003. Washington, DC, available: http://www.eia.doe.

gov/iea/; IMF. 2005. World Economic OutlookApril database, available: http://www.imf.org/external/pubs/ft/weo/2005/01/data/index.htm;

World Bank. 2005. World Development Indicatorsonline database, available: http://devdata.worldbank.org/dataonline/.
http://www.eia.doe.gov/iea/http://www.eia.doe.gov/iea/http://www.imf.org/external/pubs/ft/weo/2005/01/data/index.htmhttp://devdata.worldbank.org/dataonline/http://devdata.worldbank.org/dataonline/http://www.imf.org/external/pubs/ft/weo/2005/01/data/index.htmhttp://www.eia.doe.gov/iea/http://www.eia.doe.gov/iea/


6/21


Asias energy needs, or much the same as in 2003(Figure 3.4). However, the oil intensity o energyconsumption is much more pronounced in somecountries than in others (able 3.1, Intensityo oil use in energy consumption column). Anotable eature is that small island economies are

highly dependent on oil or their energy needs.Elsewhere, oil intensity is highest in SoutheastAsia and lowest in Central Asia and the PRC, dueto their use o alternatives, such as natural gas,hydropower, and coal.

Te energy intensity o GDP (able 3.1,Energy intensity o GDP columns) is affectedby several actors, including a countrys climate,size, and stage o development as well as whetherit produces and refines oil. Countries that havecolder climates consume more energy, otherthings being equal, while countries with a largeoil contribution to GDP are likely to be moreenergy intensive. Te energy intensity o GDPalso varies with income levels: across countries, ittends to be low or the poorest but then rises withper capita income, beore tapering off at higherincome levels. Tese eatures o the relationshipbetween energy use and real output (GDP) showup in divergent patterns across developing Asia(Figure 3.5). East Asia has become much lessenergy intensive over time but the energy inputs

into GDP have risen in Southeast Asia whileSouth Asia and the Pacific have been on a flat toslightly declining trend. Te economies o CentralAsia have also as a whole become less energy

intensive, possibly because o changes in economicstructure during their transition to being moremarket oriented.

In comparing developing Asias energyintensity with other countries, it matters greatlywhether GDP is measured in nominal terms atmarket exchange rates or in purchasing powerparity rates. In nominal market terms, developingAsia consumes over three times as much energyas the G7 per unit o output. But in purchasingpower terms, developing Asia is less energyintensive than the G7. Only countries that are

major oil producers or refiners, or which have verycold winters, are more energy intensive than theG7 average. A true picture o energy intensity indeveloping Asia is likely to lie somewhere betweenthe nominal and purchasing power parity-adjustedGDP measures. But it is highly likely that, oridentical activities, or example power production,developing Asia is less energy efficient thanindustrial countries.

Te last column o able 3.1 shows per capita

Figure 3.4 Intensity of oil use in energy consumption,

19802003

0.0

0.2

0.4

0.6

0.8

1.0

Developing Asia

The Pacific Central Asia

South AsiaSoutheast Asia

East Asia

200320011998199519921989198619831980

Notes:1. Intensity of oil use is the share of petroleum in total

energy consumption. Oil is sole energy source at 1.0. 2. Prior to1992, Developing Asia excluded countries in Central Asia.

Source:Energy Information Administration, International Energy

Annual 2003, available: www.eia.doe.gov.

Figure 3.5 Energy intensity of GDP, 19802003

(000 BTU per unit of real GDP)

10

15

20

25

30

35

40

45

50

100

105

110

115

120

125

130

135

140

Developing Asia

The Pacific Central Asia (right scale)

South AsiaSoutheast AsiaEast Asia

200320011998199519921989198619831980

right scale

BTU = British thermal unit.

Notes:1. Energy intensity of GDP is energy consumption in BTU

per US$ of GDP (in 2000 prices). 2. Prior to 1992, DevelopingAsia excluded countries in Central Asia. 3. Before 1992, all

Pacific countries were net oil importers; in that year, PapuaNew Guinea became a net oil exporter.

Source: Energy Information Administration, International Energy

Annual 20 03, available: www.eia.doe.gov.
http://www.eia.doe.gov/http://www.eia.doe.gov/http://www.eia.doe.gov/http://www.eia.doe.gov/


7/21


oil consumption or developing Asia. Tesenumbers show a strong positive association withper capita income, although other actors alsomatter. A measure o Asias potential demand

or oil is captured by the difference between theaverage per capita consumption o developing Asiaand that o the G7. For developing Asia, thesediffer by a actor o more than 11 times.

Oil sel-sufficiency, oil intensity o energyconsumption, energy intensity o GDP, and percapita oil consumption are likely to be closelycorrelated with a countrys susceptibility to oilprice shocks. One way to bring this inormationtogether is to measure the potential impact ohigher oil prices on oil import costs.

Impact of higher oil prices on import bills andexport adjustment

In able 3.2, the potential impact o higher oilprices on the (net) import uel bill is shown. Forthis purpose, oil prices are assumed to rise by75%, which is approximately the increase in pricesbetween the start o 2005 and end-August. Allcosts are expressed as a percentage o GDP. Inthese calculations, higher prices are assumed tolast or 1 year. As oil production and consumptionare taken as given and there is no allowanceor possible adjustments, these are estimates o

potential costs rather than those that are likely.Tis simple, illustrative exercise also gaugesthe potential squeeze on domestic absorptiono traded goods in circumstances where addedimport costs cannot be met by use o oreignexchange reserves or through external borrowing.o the extent that base-value shares o net oilimports have risen since 2002, which is the baseyear or the calculations in able 3.2, potentialimpacts on oil import bills will be larger.

At the subregional level, this exercise suggeststhat South Asia is the most vulnerable to higher oilprices that work through rising uel import bills.South Asia also has the lowest oil sel-sufficiencyindex o all subregions and its GDP, measured atmarket exchange rates, is comparatively energyintensive. Impacts are also substantial or EastAsia and Southeast Asia. A more modest impactor the Pacific is largely attributable to PapuaNew Guinea and its very heavy weight in thePacific aggregatebut or individual Pacific islandeconomies (apart rom imor-Leste), the potential

Table 3.2 Net oil imports, developing Asia

Subregion/Economy Estimated impact of a

75% price rise in the net

oil import bill (% of GDP)

East Asia -1.76

China, Peoples Rep. of -1.11

Hong Kong, China -2.07

Korea, Rep. of -2.85

Mongolia -12.56

Taipei,China -2.28

Southeast Asia -1.23

Cambodia -1.19

Indonesia 0.54

Lao Peoples Dem. Rep. -2.18

Malaysia 2.37

Philippines -3.41

Singapore -4.89Thailand -2.61

Viet Nam -0.69

South Asia -2.31

Afghanistan -1.58

Bangladesh -1.86

Bhutan -2.37

India -2.01

Maldives -8.58

Nepal -3.69

Pakistan -4.17

Sri Lanka -4.43

Central Asia 16.02

Azerbaijan 26.70Kazakhstan 21.33

Kyrgyz Republic -6.42

Tajikistan -26.76

Turkmenistan 25.07

Uzbekistan 0.53

The Pacific -1.02

Fiji Islands -6.20

Kiribati -5.70

Papua New Guinea 3.47

Samoa -5.55

Solomon Islands -6.68

Tonga -7.35

Vanuatu -3.42

Developing Asia -1.53

Notes:1. The base net import shares used in this exercise are

derived from physical data on imports and exports of oil for

2002, the latest year available, and on the prices prevailing

at that time. 2. Net import shares may vary depending on

exchange rates, prices paid for oil, and other factors that affect

output and the supply and demand for oil.

Sources:Staff calculations using data from the Energy

Information Administration, available: www.eia.doe.gov.
http://www.eia.doe.gov/http://www.eia.doe.gov/


8/21


impact on import bills o higher oil prices issubstantial. As a net oil-exporting region, CentralAsia will potentially enjoy larger net exportreceipts rom higher oil prices.

At the country level, Mongolia and ajikistanseem to be the most exposed to the risk o asharp rise in import uel bills. Although potentialcosts could be exaggerated by the data used here,other sources too suggest large potential impacts:6.8% o GDP or ajikistan and 9.8% o GDPor Mongolia (International rade Centre 2005).Potential impacts are also large or the Maldives,Pacific island economies (except Papua NewGuinea and imor-Leste), Kyrgyz Republic, andSingapore, ranging rom 4.5% to 9.0% o GDP.Pakistan, Philippines, Nepal, and Sri Lanka ace

more measured impacts o about 3.04.5% oGDP. For other countries, including the PRC andIndia, potential costs are smaller but by no meansinsignificant. For developing Asia as a whole,imported uel costs could rise by 1.5% o GDP,but this is afer subtracting possible gains by oil-exporting countries.

Another way to measure exposure to higheroil prices is to identiy by how much exportswould need to grow to pay or higher import uelbills. Te data in able 3.3 show the percentagepoint growth in exports that would be needed

to offset the impact o a 75% rise in the uelimport bill on the trade balance. Again, it shouldbe noted that to the extent that oil import costshave risen relative to exports since 20012003, theestimates in able 3.3 may understate the ratiosthat would result rom use o more recent data.Also, this is, once more, a partial calculation andso impacts should be interpreted as potentialrather than likely.

Te estimates in able 3.3 bring out severalpoints. For many net oil-importing Asiancountries, the growth in exports that wouldbe needed to pay or a 75% rise in the costo imported oil is potentially large. Te mostpronounced impacts are in Mongolia and in someSouth Asian countries. Normally, such adjustmentswould occur through a depreciation o thedomestic currency and a shif o resources romnontraded to traded goods activity. Even i higherprices were not sustained and these estimates werehalved, temporary financing needs could still besignificant. In some countries, financing needs

might be met by a drawdown o oreign exchangereserves. But other countries ace more difficultcircumstances. Countries with large external debts,meager reserves, and limited borrowing capacitycould ace financing difficulties.

Tese estimates o the potential suscepti-bility o import bills and trade balances to higheroil prices omit many actors that will affect theeventual impacts. Oil product prices tend to movein step with crude prices but the correlation is notexact. o some degree, thereore, susceptibilitywill depend on the particular product mix o oilconsumption. Producers and consumers will alsoadjust to higher oil product prices, as well as tochanges in income, exchange rates, and interestrates. Important indirect effects will also ollowrom impacts on major trading partners and

Table 3.3 Export growth required to pay for a 75%

rise in fuel prices

Subregion/Economy Export offset, percentage

point changeEast Asia

China, Peoples Rep. of 2.3Hong Kong, China 4.5

Korea, Rep. of 10.5

Mongolia 16.5Taipei,China 5.3

Southeast Asia

Cambodia 4.5

Indonesia -9.8Lao Peoples Dem. Rep. 11.3

Malaysia -3.0

Myanmar -6.8Philippines 6.8

Singapore 2.3

Thailand 5.3Viet Nam -5.3

South Asia

Afghanistan 6.8

Bangladesh 6.0India 15.8

Maldives 8.3

Nepal 26.3Pakistan 18.0

Sri Lanka 8.3

The Pacific

Fiji Islands 6.0Papua New Guinea -5.3

Average 3.0

Note:Based on country averages for 20012003.

Source:Adapted from World Trade Organization. 2005. World

Trade Report. Appendix Table 7, p. 25.


9/21


In a net oil-importing economy,rising oil prices affect output,

inflation, and the balance o pay-ments, as well as the fiscal position,through several pathways.

First, increasing oil prices squeezeincome and demand. At a givenexchange rate, more domestic outputis needed to pay or the same volumeo oil imports. I the domestic cur-rency depreciates in response toinduced payments deficits, thisurther cuts the purchasing powero domestic income over importedgoods. Since important tradingpartners are also likely to sufferincome losses, slower growth oexternal demand aggravates thesedirect impacts. Higher oil pricesalso squeeze aggregate supply, sincerising intermediate input costs erodeproducers profits and may causethem to cut back on output. Lowerprofits may then eat into investmentspending and cause potential outputto all over a protracted period.

Second, higher oil prices presentan inflationary threat. Inflation isdirectly influenced through the weight

o oil products in the consumptionbasket. Secondary or indirect impactsare elt as producers pass throughsome part o higher oil costs to theprice o final goods. Induced effectsollow i higher goods prices leadto higher wage costs that eed backinto prices. But when oil pricesall, nominal wage and other pricerigidities can limit the pass-through tolower final goods prices.

Tird, rising oil prices have fiscalconsequences. I the retail prices o oil

products are subsidized, as they are in

many Asian countries, outlays on uelsubsidies will ratchet up as prices rise.

Tis may prompt cuts in governmentspending; i it does not, larger fiscalburdens will have to be borne. Indi-rectly, fiscal balances will respond tochanges in income and expenditure.

In a net oil-exporting country,the impacts o higher oil pricesare not always the mirror image othose elt by oil importers. Incomesrise in the oil sector, certainly, butdomestic oil consumers (producersand households) may lose. Teeffect on aggregate demand andaggregate income is ambiguous anddepends on a variety o actors. I,or example, most o the additionaloil revenues are saved, or leak romthe economy through profit remit-tances, negative consumption effectsmay dominate. Te way in whichthe fiscal authorities use larger oiltax revenues is crucial. An excessiveexchange rate appreciation couldstunt growth in non-oil sectors.

Precisely how significant thesevarious effects are will depend onmany actors. Te size o oil price

rises is clearly important but so toois the reason or them. I higherprices are a result o strength in theglobal economy, then global demandis clearly less at risk. Te durationo higher prices is also relevant. Ihigher prices endure, accumulatedimpacts will be larger. It also matterswhether consumers and producersexpect higher prices to be temporaryor sustained: i they think that theyare going to last, higher prices arelikely to have larger impacts than i

they are viewed as short-lived.

Te credibility that the authoritiesenjoy in fighting inflation can be vital

in this regard. I rising uel pricesunleash a cost-push inflationaryspiral, as in the late 1970s, thenoutput losses are likely to be mag-nified; but i inflationary impulsesare quickly tamed, and inflationaryexpectations remain firmly anchored,impacts will be more muted. Flex-ibility in pricing and in markets willalso help by encouraging the substi-tution that cuts the oil intensity odemand.

Structural actors are alsoimportant. I oil intensity is high,adjustments are likely to be moredifficult. Importing countries withmeager oreign exchange reserves,poor creditworthiness, and highexternal debts will have greater di-ficulty in coping with the addedfinancing needs o higher oil prices.Where bank or business balancesheets are ragile, higher oil pricesand slower growth may aggravatefinancial distress.

In sum, it is not easy to put allthese pieces together and identiy the

possible impacts o higher oil priceson output, prices, and the balance opayments. In the real world, manychanges occur together, some pullingin opposite directions. Higher oilprices may induce policy responses,which, themselves, influence incomeand prices. I changes are gradual andimpacts deerred, they may prove di-ficult to separate rom other ongoingdevelopments. Identiying impacts ismore complicated still because reper-cussions in one country are likely to

spill over and affect others.

Box 3.1 How higher oil prices impact on growth, inflation, and financial balances

rom policy responses to changing circumstances.Box 3.1 summarizes the different ways in whichhigher oil prices can affect an economy.

The Impact of higher oil prices on growthNumerical simulation methods are neededto unravel the kinds o impacts o higher oil

prices on growth that are described in Box 3.1.Any estimate o the impact o higher oil priceson growth is necessarily contingent on a largenumber o assumptions about the nature o theshock, underlying economic structures andbehaviors, and policy responses. In able 3.4, theresults o simulations o the impact o higher oil


10/21


assumed that higher oil prices start in the thirdquarter o 2005 and are sustained through theourth quarter o 2006. All other actors are heldconstant. In reality o course, many changes occurtogether, so these calculations are indicative and

do not constitute orecasts.Te simulated impacts reported in able 3.4 aresizable or some countries. Te OEF model simu-lations suggest that Philippines, Singapore, andTailand are most susceptible to slower growth ihigher oil prices endure through 2006. All thesecountries are large net oil importers, but negativeimpacts on growth are mitigated by expandedfiscal deficits. In the Philippines and Tailand,fiscal deficits increase by nearly 1 percentage pointo GDP compared to the baseline. Tese fiscalimpacts reflect automatic tax and expenditureadjustments as incomes and prices change, and donot take account o specific oil subsidy schemes,such as the substantial expenditures incurred overthe last 18 months in a number o countries.

Simulated impacts on output growth in thePRC and India are smaller, but not insignificant.Although oil dependency is low in the PRC, themodel traces relatively large negative growthimpacts through external trade. Simulated fiscalimpacts in the PRC are modest. Te impact on

Table 3.4 GDP and budget balance impacts of a rise in the oil price to $70 per barrel, 2006 (percentage points of GDP)

GDP growth, OEF Budget balance,

OEF

GDP growth,

IMF MULTIMOD (2000)

G3 (US, Japan, euro zone) -0.5 -0.3 -0.5China, Peoples Rep. of -1.0 -0.1 -0.6

Hong Kong, China -0.9 -0.1 -

India -1.1 -0.9 -0.8Indonesia (-0.9) -1.1 (0.0) +0.2 0.1

Korea, Rep. of -0.5 -0.9 -1.4

Malaysia (-0.6) -1.1 (0.0) +1.0 -0.3Philippines -1.4 -0.8 -1.3

Singapore -1.3 -0.4 -

Taipei,China -0.2 -1.1 -

Thailand -1.8 -0.7 -1.4

- = not available.

Notes: 1. The baseline is calculated under an assumption of oil prices at $53 per barrel from Q3 2005 to Q4 2006. The simulation is basedon a rise in prices to $70, sustained over the same period. 2. The International Monetar y Fund (IMF) numbers in the GDP growth, IMF

MULTIMOD column result from scaling the impact of a $5 per barrel rise over a $25 per barrel baseline by 1.6, which is roughly equalto a 32% rise in the oil price. This assumes that impacts are linear, which they may not be, and are independent of the base starting

price. 3. The IMF MULTIMOD estimate is for industrial countries, not an average for the G3. 4. For Indonesia and Malaysia, the numbers

in parentheses show the estimated impact on growth and the budget balance when additional oil revenues accruing to government arerecycled.

Sources: Staff calculations using OEF model (available to subscribers: www.oef.com), OEF data release, August 2005; and IMF Research

Department. 2000. The Impact of Higher Oil Prices on the Global Economy. Washington, DC. December, available: http://www.imf.org/external/pubs/ft/oil/2000/.

prices on growth and fiscal balances or selecteddeveloping countries in Asia are summarized.Tese simulations have been conducted using theOxord Economic Forecasting (OEF) model. In theOEF model, higher oil prices squeeze aggregate

demand and supply or net oil importers. Balance-o-payments adjustments occur through the realexchange rate. Indirect impacts are capturedthrough the effect o higher oil prices on tradingpartners growth, which affects exports. Cuts ininvestment may result in a smaller capital stockand permanent output losses, but growth shouldlater return to its original trajectory. Te modelassumes that public sector savings or deficitsadjust passively to the hike in oil prices, and thatinflationary pressures are addressed throughhigher interest rates. Te ocus here, however, ison possible short-run impact effects and not onmore protracted adjustment processes.

In able 3.4, the results o the GDP growth,OEF column show percentage point differencesin GDP growth or 2006 resulting rom a $17/bblhike in the price o oil over this Updates $53/bblbaseline assumption, essentially a rise to $70/bbl.Te results in the Budget balance, OEF columnshow percentage point changes in governmentbudget deficits measured relative to GDP. It is
http://www.oef.com/http://www.imf.org/external/pubs/ft/oil/2000/http://www.imf.org/external/pubs/ft/oil/2000/http://www.imf.org/external/pubs/ft/oil/2000/http://www.imf.org/external/pubs/ft/oil/2000/http://www.oef.com/


11/21


Indias GDP growth is broadly consistent with its oildependency. In India, growth is shielded through alarge measure o fiscal stabilization, and the publicsector deficit expands by 0.9% o GDP in response.

Te OEF model suggests that higher oil priceswould substantially reduce growth in Indonesiaand Malaysia. Indonesia became a net oil importerin 2004, but its dependency on oil imports is stilllow. As Malaysia is a net oil exporter, it benefitsdirectly rom higher prices. Te simulationssuggest that any benefits accruing to oil producersare significantly outweighed by indirect impactson exports as growth slows in major tradingpartners. Tese calculations assume, though, thatadditional fiscal revenues accruing rom higher oilprices are added to government saving. I, instead,

governments target the deficit and recycle oilrevenues, a smaller negative impact on output islikely to ollow. For Malaysia, the negative impacton growth could be as small as 0.6 percentagepoint o growth i the entire fiscal windall isrecycled. For Indonesia, the windall is smaller,but could reduce the potential impact on growthrom 1.1 to 0.9 percentage point. Again, thesecalculations make no allowance or the cost ouel subsidies.

Measured in terms o its oil sel-sufficiencyand oil intensity o energy consumption, Korea is

highly vulnerable to higher oil prices (able 3.1above). Koreas oil dependency and oil intensity oenergy profile is very similar to that o the Phil-ippines and thereore it might be expected thatsimilar impacts are likely. However, compared tothe Philippines and other oil-dependent countries,the estimated reduction in growth or Koreais small. Te reason is that the model predictssubstantial import compression, showing Koreanimports greater sensitivity to the real exchangerate ollowing the rise in oil prices. Impacts arealso moderated through more expansive fiscalaccommodation in Korea.

IMF (2000) has also estimated the possibleimpact on growth o an oil price shock. Teseresults have been used as a basis or imputingthe numbers shown in the GDP growth, IMFMULIMOD column o able 3.4. For mostcountries, the OEF and IMF estimates are broadlysimilar, once allowance is made or the act thatIndonesia is now a net oil importer.

In its April 2005 World Economic Outlook,

and drawing on the results o an associatedbackground paper (IMF 2005), IMF revisitedthe likely impact o higher oil prices on growth.Tis update occurred in a context where the

impact o higher oil prices on global growth in2004 had been muted (Box 3.2). IMF considers atemporary rise in the price o oil to $80/bbl roma baseline o $46. In real terms, and measuredin terms o annual averages, prices at this levelwould be close to the historical high o 1979.For developing Asia, IMF reports that outputlosses could be about 0.8 percentage point oGDP. Adjusted or differences in the scale o theassumed shocks, this estimate is substantiallylower than the OEF model impacts and, indeed,those o IMF (2000) and the International Energy

Agency (2004). But when IMF assumes thathigher prices are sustained over a longer period,as more recent news emanating rom the oilmarkets would seem to suggest, impacts rise to1.3 percentage points o GDP.

Tere is clearly uncertainty about the likelyimpact o higher oil prices. Much depends onassumptions both about the size and duration othe shock, and about how various actors respond.For example, producer behavior in the OEF modelimplies both a rapid pass-through o higher oilprices to final goods, and consumer adjustment to

changes in current income. However, competitivepressures and weaker demand growth may slowor limit the pass-through, and i consumersand producers believe that higher prices willbe temporary, they are more likely to spreadadjustments out.

Despite this uncertainty, there is a consensusthat, or developing Asia, the impact o higheroil prices will be negative. Drawing together thestrands o evidence presented here, it seems thatoil prices at about $70/bbl through to the endo 2006 could cut growth by over 1 percentagepoint in a number o countries. Some countriesin Southeast Asia and South Asia could seegrowth trimmed by the most but there areoffsetting positive actors that vary rom countryto country and that will influence actual growthoutcomes (see Part 2 o this Update). Te pointbears repeating that developing Asia is now betterpositioned to absorb large shocks than it wasat the time o the previous oil price shocks (seeADO 2004 Update, Part 3): external payments


12/21


positions are more secure, monetary policy ismore credible, fiscal strength is greater, andeconomic structures are more flexible and capableo adjusting more quickly than beore. Althoughthe regions appetite or oil continues to grow, theoil intensity o output is drifing lower.

In the next section, policy responses to higheroil prices are considered.

Policy responses to higher oil prices

No one size fits all response to higher oil pricesexists. Across developing Asia, circumstances varygreatly and countries need to respond in different

ways. For net oil importers, the challenges posedby higher oil prices will differ depending on theirmacroeconomic conditions, available financialresources, degree o access to internationalcapital markets, impact on trading partners,economic structure, and uel-pricing policies.For net oil exporters, structural actors will alsobe important, including their oil reserves, theownership structure o the oil sector, oil taxation,the governments financial position, and the publicsectors absorptive capacity. Matters are morecomplicated still, or all countries, because thereis ofen a considerable measure o uncertaintyabout how long higher prices are likely to endure.

Between 2004 and 2005, GDPgrowth in developing Asia is

expected to slow by about 0.8 per-centage point, rom 7.4% to 6.6%.Broadly, this is a reversion totrend. Several actors, in additionto higher oil prices, may have con-tributed to sofening (see Part 1):some economies have been affectedby a cyclical downturn in the elec-tronics sector; slower growth oglobal trade has trimmed exportgrowth; and in several countries,fiscal and monetary policies havebeen less accommodative.

It is difficult to be precise aboutthe part that the various actorshave played in moderating growth,though there are several reasonsto think that the role played byhigher oil prices has been muted.First, or much o 2004 and inearly 2005, impacts ran largelyrom global demand (notably, USand PRC demand) to oil prices,not the other way round, therebylimiting the negative effects o

higher oil prices on consumerand investor confidence. Second,over this period the escalation inoil prices was gradual, suggestingthat impacts, too, may stretch outover time. Tird, i consumersand investors had regarded higher

oil prices as largely temporary,they would not have significantly

adjusted their spending plans. And,ourth, consumers across devel-oping Asia were to a significantextent shielded rom the effects ohigher prices by discretionary risesin government uel subsidies andby firms limiting the pass-throughto final prices through cuttingmarkups. Te box figure indicatesthat the recent upsurge in oil prices

has had little impact in acceleratingconsumer price inflation.

IMFs World Economic Outlookin April 2005 assumed that averageoil prices over 2005 would be about

23% higher than in 2004. On thisbasis, IMF calculated that higher

oil prices might cut global growthin 2005 by 0.20.5 percentage pointin a context where global growth isexpected to slow by 0.8 percentagepoint. As developing Asia is a largenet oil importer and the globalestimate includes gains or net oilexporters, it might be expectedthat the effect o higher oil priceson slowing growth in developingAsia might be at the upper end othe IMF range. Indeed, as oil priceshave climbed well beyond the IMF

projection or 2005, it would betempting to conjecture that outputlosses in developing Asia might belarger than the IMF upper bound.I this were in act the case, littleo the slowdown in 2005 could beattributed to other developments,including a dip in the growth oworld trade volumes. More likely,the extensive use o uel subsidiesseen in 2004 and so ar in 2005,helped by generally strong fiscal

and oreign reserves positions,has contained output losses. Tis,though, raises the question o whatis likely to happen as subsidies arescaled down or removed, a processthat is now under way in severalcountries.

Box 3.2 Oil prices, inflation, and GDP growth, 20042005

Box figure Average inflation rates,20022005

0

10

20

30

40

Developing AsiaBrent crude

2005200420032002

%, year on year

Sources: Datastream, downloaded

15 August 2005; Asian Devel opment Outl ook

database; staff estimates.


13/21


One small benefit o such uncertainty, though,is that it will generally commend a measuredresponse, which can be reversed without incurringlarge costs. Looking to the long term, policies

that influence oil consumption and use must beconsistent with broader development objectives.

Oil subsidies and taxation

Many governments across developing Asiadirectly subsidize oil products, including kerosene,liquefied petroleum gas (LPG), and, generallyto a lesser extent, diesel and gasoline. In somecountries retail prices are openly subsidized andin others they are regulated or controlled throughstate-owned distribution channels. Indirectsubsidies are also common, and are seen where

products that have a high oil content, principallyelectric power, are provided at prices below theirtrue cost. Even in countries where there areno open or indirect subsidies, taxation is ofenmodest. Excise and customs taxes on oil productsare a potentially important source o fiscal revenuethat need to be maintained at an appropriate levelboth or budget revenue and the proper long-termallocation o the countrys investment capital. Inthe recent run-up in oil prices, however, somecountries have markedly reduced such taxes in anattempt to protect consumers.

Box 3.3 summarizes the experience o eightcountries with uel subsidies. Subsidies in thesecountries have so ar limited the pass-through romhigher crude oil prices to the retail prices o variousoil products and thereore to final goods. Tis hascertainly helped contain the inflationary impacts orising crude prices, but in the absence o detailedstudy very little is known about exactly whobenefits rom these subsidies and by what amount.

Beyond concerns about the impact o higheruel prices on the general population, the rationaleor oil subsidies and discretionary increases insubsidies is not particularly clear. Subsidies do noteliminate the negative effects o higher oil priceson potential output. Demand must still adjustto the deterioration in the external paymentsposition. Subsidies also add to the fiscal burdenand represent an opportunity cost (in terms o thealternative uses to which scarce fiscal resourcescould have been put). In Indonesia, or example,the fiscal cost o oil product subsidies in 2005will be larger than budgetary allocations or

education and health combined. Raising subsidiesor reducing excise taxes as oil prices rise createsdeeper distortions, too. Subsidies underwriteuel and energy inefficiency, retard the devel-

opment and diffusion o cleaner technologies,and contribute to harming the environment. Terent created by subsidies also encourages uelsmuggling and other illegal activities.

Rising fiscal deficits, driven in part by growinguel subsidies, have led some countries to scaledown or withdraw subsidies. For example, on12 July, having incurred fiscal costs o about$2.2 billion over an 18-month period, theGovernment o Tailand announced that alluel subsidies would be removed by February2006, and immediately ended all diesel subsidies.

Malaysias Government, which had earliersuspended excise taxes on gasoline and diesel,has now declared its intention to scrap subsidieson these two products. Malaysia has adopted agraduated approach and has so ar lifed gasolineand diesel prices three times in 2005.

In Indonesia, too, diesel subsidies were cutearlier in 2005, but subsequent increases in theprice o crude oil mean that expected budgetarycosts o all subsidies have swollen and now exceedtheir 2005 appropriation. Other countries haveproblems. In Bangladesh, the state-owned oil

distributor, Bangladesh Petroleum Corporation,is accumulating very large operating losses whilethe oil bill is putting pressure on oreign exchangereserves. In India, the ederal Government hasexpressed concern about recently announcedlosses at major refining and oil marketingcompanies. Without doubt, similar pressuresare being elt in other countries that are heavilyreliant on imported uel while selling it domes-tically at below imported cost.

Removing uel subsidies clearly meets withormidable political resistance in some countries.But i subsidies are retained and higher oil pricesdo not recede, their fiscal costs will mount. Oneapproach might be to remove subsidies first onthose uel types on which the poor do not depend.In most countries in developing Asia, gasolinesubsidies are not provided or are relatively small,but, equally, taxation is ofen relatively modestgiven the income levels o gasoline consumers.Although diesel subsidies are widespread, and thepoor do not directly consume much diesel, the


14/21


Governments in developing Asiahave been trying to cushion

consumers rom the impact osoaring oil prices by subsidizingretail uel prices, based on the beliethat higher oil prices, as in pre-

vious episodes, will be temporary.However, it is becoming increas-ingly clear that higher oil pricesmay be here to stay or some time.Many Asian governments now aceincreasing pressure on their budgetsrom rising subsidy bills. Tis boxillustrates the extent o the strainon the fiscal positions in eight

countries in Asia.

Bangladesh

Since the 1970s, the petroleumsector has been served mainlyby the state-owned BangladeshPetroleum Corporation (BPC),which imports crude oil andpetroleum products and operatesthe state refinery. Te prices oBPCs petroleum products havegenerally been administered.However, instead o contributing

to state revenues, BPC has in actbeen losing heavily in recent yearsbecause it sells below cost.

Te Government has loweredtaxes on uels used by the poor,such as kerosene and diesel, whiletaxes on gasoline remain muchhigher. In January 2003, it approvedprice increases on BPCs retail sales,effectively reducing consumptionsubsidies. Tis is partly reflectedin the decline in BPCs losses orFY2003. Te move also helpedreduce smuggling into India romBangladesh. However, with the con-tinued increase in crude oil and inpetroleum product prices, the Gov-ernment has made only relativelysmall increases in some domesticprices, thus at the same time raisingcertain categories o effective uel

subsidization. Tis has generatedlarger losses or BPC, which are

entirely financed by commercialand external borrowings. As a resulto the Governments policy, dieseland kerosene were effectively beingsubsidized at 18.2% and 19.1%,respectively, o import/border pricesin FY2004, translating into a totalsubsidy o $170.5 million duringthat fiscal year.

For FY2005, it is estimated thatBPC losses were $445.4 million(about 0.7% o GDP). Sincecustoms and excise taxes were cut

in the FY2006 budget to reduce thecompanys losses, the Governmentis acing an immediate worseningo its fiscal position, in addition toits quasi-fiscal obligation stemmingrom BPCs large accumulatedlosses.

Peoples Republic of China

Domestic prices or crude oil andrefined products are in principlelinked to international prices withadjustments made afer a 1-month

lag. Tis mechanism, however, hasnot been consistently ollowed,especially or refined products, bythe authorities that control prices.Moreover, price policies have notbeen consistent throughout thecountry, with a smaller degreeo adjustment in domestic pricesto rising global oil prices in thesouthern part o the country.

Increases in retail prices haveallen substantially behind increasesin crude oil costs. Tis policy hasmuted the impact o rising globaloil prices on inflation and on pro-ducers such as armers who usediesel. However, it also means thatPRC oil refiners have incurredlosses reported at CNY4.19 billion(about $510 million) in the first halo 2005 as a result o the widening

gap between their costs or crudeoil and receipts or oil products.

Some small refiners are reportedto have cut or stopped productionbecause o the losses, and othershave diverted oil products to pro-itable markets abroad. Refineryoutput by the state-owned oil com-panies rose by only 0.5% in the first7 months o this year rom a yearearlier. Te pricing policy is onecause o shortages and reportedhoarding o oil products.

India

Te domestic petroleum prices arein practice still essentially admin-istered; particularly sensitive arekerosene and LPG since theseare used as cooking uel by manyrural poor. In the FY2004 gov-ernment budget, the subsidy orkerosene and LPG was estimated at$776.5 million. Te effective subsidybill actually reached $4.8 billion, asstate-owned distributors shoulderedthe $4.0 billion in un-recoveredcosts (losses) on sales o these

products. Tere is no indicationthat the subsidy bill or FY2005will decline, as subsidy estimatesor the first quarter alone have runup to $2.2 billion and without priceadjustments would exceed about$9.3 billion or 1.1% o GDP in theyear.

Refiners and retailers have notbeen allowed to raise LPG pricessince June 2004, and keroseneprices since April 2002. Marketingcompanies subsidize Rs92 o everyLPG cylinder and Rs11 o everyliter o kerosene. As a result, energysector losses are mounting. In thefirst quarter o FY2005, Indian Oil,Bharat Petroleum, and HindustanPetroleum suffered losses o$12.3 million, $98.5 million, and$116.1 million, respectively.

Box 3.3 Oil subsidies and fiscal strain


15/21


Indonesia

Indonesia became a net oil

importer in 2004. While it importsat market prices, state-owned Per-tamina sells petroleum productsto consumers at subsidized prices.As o April 2005, the Governmentowed Pertamina about $2.6 billionin uel subsidies, putting pressureon the companys cash flow andon its ability to pay or importedpetroleum products. Tis hasaffected oil supplies to the country,which now aces petroleumshortages. Recent parliamentarydelays in approving the Govern-ments revised budget have urtherdelayed partial payment o uel sub-sidies to the company.

Even though petroleum productprices (except kerosene) wereincreased by 29% in February,the Government estimates thatthe subsidy bill will balloon to$12.5 billion (about 4.7% o GDP)by the end o the year i currentcrude oil prices persist. Last year,

subsidies cost the Government$7.4 billion (2.9% o GDP). In theabsence to date o urther cuts insubsidies, government interventionis reduced to pushing the popu-lation to limit consumption. Carowners are also encouraged to useexpensive, nonsubsidized premiumuel, which currently accountsor only 4% o domestic gasolineconsumption. elevision stationsnow close at 12 midnight, in amove intended to curtail nighttime

energy consumption.

Malaysia

On 1 August, Malaysia increasedprices o premium gasoline by6.6% to RM1.62 per liter, regulargasoline by 6.8% to RM1.58 perliter, and LPG by 3.6% to RM1.45per kilogram, in an effort to cut

rising subsidies. Diesel prices werealso lifed by 18.5% to RM1.28 per

liter, except or fishers, who willreceive increased subsidies to offsetthe price rise. Even as this is theourth increase since October 2004(and the third this year), prices inMalaysia remain among the lowestin Southeast Asia.

Fuel subsidies cost the Gov-ernment $1.3 billion last year, and,despite the latest price increase,are expected to cost $1.7 billion in2005. In addition, tax exemptionson gasoline will cost the Gov-ernment an additional $2.1 billion,bringing this years subsidy bill to$3.8 billion (about 2.9% o GDP).

Nepal

In 2003, the Government createdan independent committee to setuel prices, ollowing heavy lossesat the state-owned oil monopoly,Nepal Oil Corporation. While thecommittee was mandated to adjustprices in line with international

trends, it has rerained rom doingso, perhaps in the hope that priceswings will ultimately cancel them-selves out. Te last price adjustmentwas only made in January 2005,and consequently the oil monopolyhas been suffering losses o overNRs500 million a month. In thesecond hal o FY2005, its lossesreached $29.4 million. I domesticprices are not adjusted, its lossesor FY2006 may exceed FY2004s$56 million (about 0.8% o GDP).

Thailand

In the wake o rising oil prices andinflationary pressures, oil subsidiesthat draw on the oil stabilizationund started on 1 January 2004.However, as sustained high oilprices began rapidly to depletethe und, the Government made

gradual moves to reduce the sub-sidies. First, the subsidy on gasoline

was removed in November 2004.In March 2005, diesel prices wereraised by B3 per liter. Ten, thediesel subsidy was reduced to B1.30per liter in June and eventuallyremoved on 12 July 2005. Never-theless, the Government still spent$2.2 billion in 18 months on uelsubsidies (about 0.9% o GDP overthis period).

Subsidies on diesel alone costaround B300 million a day duringthe spending peak. At present, theoil und is more than B80 billion indeficit. Te Government still con-tinues to subsidize the price o LPG,at a cost o around B500 million($12.6 million) per month.

Viet Nam

Viet Nam is Southeast Asias third-largest oil producer, though itspends more than hal o its crudeexport revenues on importingpetroleum products since it has no

major refineries. In addition, theGovernment subsidizes retail prices,spending about 2% o GDP on thisin 2004.

In order to reduce subsidies andcurb smuggling into Cambodiaand the PRC, in August 2005 theGovernment, or the third time,increased diesel, gasoline, andkerosene prices. In spite o this,the Government is still expected tospend about $350 million on sub-sidies in the second hal o 2005.

In the first hal, oil importers lost$440 million, and so subsidies areexpected to cost $790 million, or1.6% o GDP in 2005. Te Gov-ernment is ully covering theselosses.

Sources: National press reports,JulyAugust 2005.

Box 3.3 (continued)


16/21


poor indirectly rely on it, particularly or transpor-tation. But many non-poor also benefit rom dieselsubsidies, and the case or phasing out is strong.

For those uels that the very poor rely on the

situation is more vexed, and a range o actorsneeds to be careully considered. In principle,it may make sense to replace uel subsidiesby income subsidies, but income-targetingapproaches, e.g., vouchers, may prove difficultand costly to implement. In some situations,the removal o subsidies may not make mucheconomic sense i the alternative is that poorpeople turn to other uel sources, particularlybiomass, which result in heavier environmentaldamage and costs to health.

Governments also need to be careul in

considering the distributional impact o subsidies.Sometimes, as e.g., with diesel, subsidies arecaptured by the non-poor. Tis can happen wherethere are both monopoly control over distributionand regulatory ailure. For example, the relativelylarge share o kerosene in total oil productconsumption (see the appendix table to this part)in countries where kerosene is heavily subsidizedis an apparent indication o problems in targetingsubsidies. A decision to remove or scale backsubsidies may be politically more palatable i somepart o the fiscal savings is visibly earmarked or

development programs that are ast disbursingand that directly benefit the poor.As many decisions on energy production and

use are taken by the private sector, it is importantthat oil prices reflect ully their social and envi-ronmental opportunity costs. Tis requires goingbeyond just removing subsidies on oil products. Oiltaxes could provide an important source o budgetrevenue. Moreover, tax rates need to ensure thatoil products are priced to ully reflect the negativeexternalities that they create in terms o pollution.

Te price o oil products will be a majordeterminant o Asias uture demand, not just oroil but or alternative sources o renewable energyas well (Box 3.4). I oil is not suitably taxed, or isinappropriately subsidized, incentives to developand adopt more energy-efficient technologies willbe blunted and conservation will be hampered.Tis is a major reason why, in the past, developingAsia has not always adopted energy-efficienttechnologies, preerring cheap but less energy-efficient alternatives.

Macroeconomic policies

For net oil-importing countries, higher oil priceswill require that domestic demand adjusts to adecline in potential output. Te role o macro-

economic policies should be to ease neededadjustments to demand and supply and to guardagainst the possibility o a destabilizing inflationaryspiral. Different economies will have varyingdegrees to maneuver in their policy responses.

In countries where the monetary authorityenjoys credibility and where inflationary expec-tations are well anchored, monetary policy may beable to accommodate some o the direct impact ohigher oil costs on final goods prices. But i higheroil prices threaten to percolate through to risingwages in a second round o cost increases, or

inflationary expectations become heightened, themonetary authorities should consider tightening.Tis will help guard against the risk that higheroil prices unleash a cost-price spiral, magniyingoutput losses over a protracted period. Tis wasthe experience across much o Asia during thefirst and second oil price shocks, though thistime around, preemptive tightening o monetarypolicy, as seen in timely measures taken in thePhilippines and Tailand, should help containinflationary impacts.

Fiscal policy can help buffer the output losses

entailed by higher oil prices. Its role should beto assist in smooth adjustments and to providea measure o temporary relie, but it cannotinoculate an economy against higher oil prices.Normally, fiscal stabilization should occur auto-matically. Any discretionary response should belimited, especially as it may be difficult later onto remove expenditure programs and subsidiesor to restore oil taxation to previous levels ioil prices subsequently all. Attempts to shieldconsumers and producers rom the impact ohigher oil prices through discretionary fiscalsubsidies, as is happening in many countries, canhave a high opportunity cost both in fiscal termsand in terms o broader efficiency considerations(see above). For countries whose initial fiscalposition is weak, even automatic stabilizationmay prove difficult. I a larger deficit cannot beaccommodated, adjustments will need to be moreabrupt.

Tose countries acing external paymentsdifficulties will generally have less scope to


17/21


The box figure gives a snapshoto retail prices o super gasoline

and diesel in November 2004 acrossa sample o 30 Asian developingcountries. Te data were compiledby German echnical Cooperation(GZ). It shows three sets o coloredvertical lines that define benchmarkprices that broadly indicate national

pricing policies or transportationuels. Te red lines (27 US cents

per liter) indicate the cost per litero crude oil, which was $43/bbl atthat time. Te green lines are theUS prices (54 US cents per liter orgasoline and 57 US cents per liter ordiesel) representing product pricesdetermined in a competitive market

with an efficient refining industry;the US prices include about

10 US cents per liter o taxationthat was considered a reasonablededicated tax standard neededor road or general transportationinrastructure. Te yellow verticallines indicate Luxembourg productprices representing the approximate

Box 3.4 Retail fuel prices in Asia

Box figure Comparison of retail fuel prices in Asia (as of November 2004, US cents per liter)

Hong Kong, China

Korea, Rep. of

Papua New Guinea

Fiji Islands

Singapore

India

Cambodia

Bhutan

Nepal

Sri Lanka

Taipei,China

Tajikistan

Timor-Leste

Pakistan

Mongolia

Bangladesh

Lao Peoples Dem. Rep.

Thailand

Afghanistan

Kazakhstan

Philippines

Peoples Republic of China

Kyrgyz Republic

Viet Nam

Azerbaijan

Malaysia

Uzbekistan

Indonesia

Myanmar

Turkmenistan

0 50 100 150 200050100150200

5798 27 27 54 119

Diesel Super gasoline

Retail fuel prices of Luxembourg = approx. minimum entrance level for 10 European Union accession countries.

Retail fuel prices in the United States = average cost-covering retail prices including industry margin, VAT, and approximately10 US cents for the two road funds (federal and state). This fuel price, as it has no other specific fuel taxes, may be considered

the international minimum benchmark for a nonsubsidized road transport policy.

Crude oil prices on the world market (Brent at Rotterdam).

154

135

94

91

89

79

87

72

72

78

65

71

67

53

54

54

59

61

62

48

48

52

52

2

27

35

37

41

48

12

100

1

95

73

64

55

61

62

65

41

55

59

49

59

58

37

48

34

67

41

18

30

22

18

32

43

43

10

38

34

Source: GTZ. 2005. International Fuel Prices, available: http://www.gtz.de/fuelprices.
http://www.gtz.de/fuelpriceshttp://www.gtz.de/fuelpriceshttp://www.gtz.de/fuelprices


18/21


minimum level or the 10 EuropeanUnion accession countries. Only two

economiesHong Kong, China andKoreapriced around or above thisbenchmark.

For gasoline, the bulk o coun-tries (18) charged the green linebenchmark price or more while anadditional six charged the green lineprice excluding taxes o 10 US centsper liter. Whether at the margin thegreen line prices would representcost recovery in a country woulddepend on local circumstances,especially refining industry efficiency

and national distribution costs.Only a small raction o crude costsare covered in urkmenistan andMyanmar while in November 2004Indonesias gasoline prices might justhave covered crude costs. Uzbekistan,Malaysia, and Azerbaijan coveredcrude costs, but gasoline prices

at that time appear to have beensubstantially subsidized. For diesel,

five countries charged less than theindicative crude costurkmenistan,Myanmar, Indonesia, Malaysia, andAzerbaijan. As with gasoline, allcountries that did not recover crudecosts were oil producers. It is notablethat another 14 countries did notprice to the green line standard, i.e.,the bulk o countries provided sub-sidies or diesel.

Since November 2004, crudeprices have risen considerably, butmany o developing Asias govern-

ments have not ully passed thisthrough to retail prices. Tey haveextensively relied on increases indirect subsidies, cuts in petroleumtaxation, and losses by state-ownedpetroleum companies to avoid ullprice adjustment. Tus the verymixed picture or November in

cost recovery in transportation uel(including some minimal taxation)

has likely been heavily clouded sincethen.

Across Asia, the most heavilysubsidized oil products are nottransportation uels but productssuch as kerosene and LPG usedby the poor, mainly or cooking.(However, data on the structure andmagnitude o these subsidies are notreadily available.) Kerosene, in par-ticular, is heavily subsidized in manycountries, since it ofen accounts ora significant part o poor households

expenditure. For example, in Indiaand Nepal, kerosene absorbs about2% o total household spendingamong the poorest urban households(UNDP/ESMAP 2005). I keroseneprices increased significantly, poorhouseholds in these countries andelsewhere would be at risk.

Box 3.4 (continued)

smooth out the negative impacts on prices andpotential output, and are likely to ace more

difficult economic adjustments. In the absence osufficient oreign reserves or external financingopportunities, a deteriorating trade balance mustbe accommodated by reductions in domesticconsumption and investment. In such cases,a depreciating exchange rate will acilitateadjustments o domestic demand and will helpmove resources rom the nontraded to the tradedgoods sector. However, in poor countries withlarge external debts, additional external financingassistance on a grant basis or on highly conces-sionary terms may be needed as a temporarymeasure to help fill payments gaps.

Higher oil prices also pose challenges to net oilexporters. Much will depend on the distributiono income gains, and whether the non-oil sectoraces higher costs. In countries where oil revenuesare narrowly concentrated, the overall impact ohigher oil prices on aggregate demand may benegative, but where increased oil incomes spillover into the broader economy, private demandmay expand and generate inflationary pressures.

As a consequence o oreign exchange inflows, anappreciation o the real exchange rate is likely to

ollow, squeezing activity in the non-oil, tradedgoods sectorthe so-called Dutch disease.Sterilized intervention may slow the process

and help contain domestic inflation, but cannotstop it. I the increased oil income seems to betemporary, governments need to exercise cautionabout expanding expenditure programs. Eveni the gains look like being more permanent,the authorities need a plan to use them overa medium- to long-term horizon, integratethem within a broader expenditure planningramework, and ensure that spending decisionspass standard tests that guard against waste.

Oil-exporting countries may also considerthe benefits o making a precautionary reductiono their debts; o saving in oil stabilization undsheld in oreign currency assets (to finance uturedevelopment expenditures); and o targeting anon-oil fiscal deficit that would limit macro-economic strain. Tese are some o the issues thatthe net oil exporters in Central Asia and Pacific,or example, will continue to grapple with.


19/21


Long-term responses

Asia needs energy, particularly power, to develop.Te elimination o poverty and enhanced social

development will depend critically on securinguture supplies o energy and on ensuring that itsupports investments in agriculture, basic healthand sanitation, education, power, transport,and industry. For the oreseeable uture, oil willremain one oi not themajor source o energyor meeting these needs.

Over the long term, several actors willinfluence uture oil dependency and energyefficiency in developing Asia. An importantstarting point is or national energy policies tomake rational choices about the development o

an appropriate energy mix. Such a rameworkneeds to clearly set out strategies or ensuringefficiency in use and development, adequacy andreliability o supply, and measures to mitigateenvironmental impacts. Investor confidencein oil and other energy sectors benefits rom apredictable and transparent ramework to guidegovernment decisions over the long run. Animproved investor climate will in turn promotesupply and help stabilize prices.

Within the broader ramework o a nationalenergy policy, a number o specific measures are

likely to reduce susceptibility to the risks o highoil prices. However, on a broader view, promotingenergy efficiency and diversity transcends thenarrow boundaries o energy policy. Policieson competition and investment, transpor-tation, technology, and even finance all have animportant role to playas well as, o course,energy pricing policies.

Te development o competitive markets inoil and other energy products is also important.In many o the regions developing countries, theoil sector has long been dominated by inefficientstate-owned entities. Inviting the private sector toparticipate in the oil and energy sector is likely tobe beneficial but may require legal, institutional,and regulatory changes. Access to energy-efficienttechnologies and related know-how, which may beprotected under intellectual property saeguards,may not be possible without market opening andoreign participation. In the past, heavy regulationo ownership has aggravated supply and capacityproblems and has deterred investment.

ransport policy will play a major role ininfluencing uture oil dependency and energyefficiency, since vehicles are the largest source odemand or oil in developing Asia, and vehicle

ownership is set or explosive growth. Forexample, the PRC is set to become the worldssecond-largest automobile market within a decade.Decisions about investments in road and railinrastructure, urban transportation systems,

vehicle taxation, and user costs will all exert animportant structural influence on demand orand dependence on oil. Here, too, competitionshould have a key role to play in making choicesavailable to consumers and in helping ensure thatresources are used efficiently. Where competitionis not possible, the role o regulation should be to

help mimic competitive outcomes. An importantguiding principle should be that transport users,whatever the mode, should pay prices that ullyinternalize social costs.

Tere are o course many other areas wherepolicies will have a long-term effect on oildependency and energy efficiency. For example,ailure in rural credit markets may impedeinvestments by armers in uel-efficient methodsor generating power. In some cases, it may makesense to subsidize or provide tax incentives orclean energy alternatives that generate significant

external benefits in terms o health, time savings,or the reduced risks that ollow rom diversifiedenergy sources. Te case or carbon emission taxesis o course well understood and documented.

Another long-term response is illustratedin the buildup o strategic reserves o oil by thePRC and India. Strategic stocks are not intendedto guard against high prices; their main objectiveis to ensure oil availability in the event o aphysical disruption in supply. Early last year, theInternational Energy Agency (2004) est imatedthat the PRC had 35 days o crude oil reservesand that India had 15 days. Both countries havedeclared their intention to significantly increasestrategic reserves and are investing in storageacilities. Te Second ASEAN+3 MinistersMeeting on Energy, held on 13 July 2005, alsoaffirmed the importance o strategic oil reservesas an important element o an overall energysecurity package. At this time, the ASEAN+3Initiative aims to acquire oil stocks on a

voluntary and commercial basis.


20/21


Conclusions

Although there remains some uncertainty abouttheir uture path, higher oil prices could be here to

stay or some time. Te run-up in prices that hasoccurred since March 2005 would appear to havea significant permanent component. Supply aswell as demand pressures would now appear to befiguring more prominently in the market outlook.In this context, and with a view to its longer-runenergy security and efficiency, developing Asianeeds to reevaluate decisions that have been madein the belie that oil would remain cheap and thathigher prices would be temporary.

First, uel subsidies, artificially low prices,and low levels o taxation on oil products are

widespread in developing Asia. Te financialcosts o these subsidies have escalated sharplyand are now beginning to create fiscal strains.Tose countries that are yet to begin removingsubsidies may be able to draw useul lessonsrom the experiences o others, such as Tailand,that have moved quickly to dismantle them. Teidea that subsidies benefit the poor most doesnot always square with the acts on the ground.Although subsidies may provide short-term relierom the pain o higher oil prices, they do so athigh opportunity cost and at the risk o upsetting

macroeconomic stability.Second, ew countries in the region adequatelytax oil products. In most, excise taxes all ar belowinternational benchmarks. Given the likelihood oan exponential increase in the demand or energyin the coming decades, and Asias reliance on oil,taxes on oil products will have an important partto play in promoting sustainable energy use. Tepain rom higher taxation o oil products is morethan likely to be compensated by greater energyefficiency, a more diversified energy mix, and acleaner environment.

Tird, there is a wide body o evidenceto suggest that the right incentivesmarketincentiveswill generally work best in influ-encing choices about oil and energy consumption.Regulation, where used, should have a lighttouch and be used to emulate market outcomesrather than supplant them. Recourse to directadministrative controls, such as those now beingimplemented in some countries, should be usedwith care. Administrative controls are ofendifficult to implement, can be easily evaded, create

opportunities or rent seeking and corruption, andcreate significant efficiency losses. Besides, theyofen ail to curb consumption.

Fourth, or net oil importers, the appropriate

macroeconomic response to higher oil pricesis to fine-tune fiscal and monetary policy toaccommodate, not resist, needed adjustments inoutput and prices. Fiscal accommodation shouldbe largely automatic and should not attempt tocompensate or negative output effects that areunavoidable. Monetary policy should lean againstunderlying inflationary pressures. Favorableinitial conditions across much o developing Asiashould mean that most economies can bear theseadjustments without seriously jeopardizing growth.

Finally, or net oil exporters, higher prices will

provide resources that can be used to acceleratedevelopment. But a measured approach is neededin which the use o oil revenues is planned within amedium- to long-term ramework. o avoid the riskso developing a lopsided economic structure, caremust also be taken to avoid a rapid and excessiveappreciation o the real exchange rate that woulddivert resources out o non-oil, traded goods activity.

Selected references

Goldman Sachs. 2005. Reassessing long-term oil prices: Finding

a new equilibrium. Global Commodity Research. 17 August.International Energy Agency. 2004. Analysis o the Impact o

High Oil Prices on the Global Economy. Paris. Available:

http://www.iea.org/textbase/papers/2004/high_oil_prices.pd.

International Monetary Fund (IMF). 2000. Te Impact o

Higher Oil Prices on the Global Economy. Research Depart-

ment. Washington, DC. December. Available: http://www.im.

org/external/pubs/f/oil/2000/oilrep.PDF .

. 2005. Oil Market Developments and Issues. Policy

Development and Review Dept. Washington, DC. 1 March.

International rade Centre. 2005. International rade Statistics.

Available: http://www.intracen.org/tradstat/welcome.htm ,

downloaded 1 August.

United Nations Development Programme/World Bank Energy

Sector Management Assistance Programme (UNDP/ESMAP).

2005. Te Impact o Higher Oil Prices on Low Income

Countries and on the Poor. Washington, DC.

Verleger, P.K., Jr. 2005. Energy: A Gathering Storm? in Te

United States and the World Economy: Foreign Economic

Policy or the Next Decade, edited by C. Fred Bergsten

and the Institute or International Economics, pp. 209-246,

chapter 7, Washington, DC. Available: http://www.iie.com/

publications/chapters_preview/388/7iie3802.pd .
http://www.iea.org/textbase/papers/2004/high_oil_prices.pdfhttp://www.imf.org/external/pubs/ft/oil/2000/oilrep.PDFhttp://www.imf.org/external/pubs/ft/oil/2000/oilrep.PDFhttp://www.intracen.org/tradstat/welcome.htmhttp://wbln0018.worldbank.org/esmap/site.nsf/files/299-05_HigherOilPrices_Bacon.pdf/$FILE/299-05_HigherOilPrices_Bacon.pdfhttp://wbln0018.worldbank.org/esmap/site.nsf/files/299-05_HigherOilPrices_Bacon.pdf/$FILE/299-05_HigherOilPrices_Bacon.pdfhttp://www.iie.com/publications/chapters_preview/388/7iie3802.pdfhttp://www.iie.com/publications/chapters_preview/388/7iie3802.pdfhttp://www.iie.com/publications/chapters_preview/388/7iie3802.pdfhttp://www.iie.com/publications/chapters_preview/388/7iie3802.pdfhttp://wbln0018.worldbank.org/esmap/site.nsf/files/299-05_HigherOilPrices_Bacon.pdf/$FILE/299-05_HigherOilPrices_Bacon.pdfhttp://wbln0018.worldbank.org/esmap/site.nsf/files/299-05_HigherOilPrices_Bacon.pdf/$FILE/299-05_HigherOilPrices_Bacon.pdfhttp://www.intracen.org/tradstat/welcome.htmhttp://www.imf.org/external/pubs/ft/oil/2000/oilrep.PDFhttp://www.imf.org/external/pubs/ft/oil/2000/oilrep.PDFhttp://www.iea.org/textbase/papers/2004/high_oil_prices.pdf


21/21


Appendix table Share in final oil consumption, 2002, selected countries, by sector and by product (%)

Sector Industry Transport Agriculture Commerce andpublic services

Residential Othernonspecified

Bangladesh 7.7 49.0 20.4 0.0 22.8 0.0

China, Peoples Rep. of 31.5 41.6 9.8 9.5 7.6 0.1

Hong Kong, China 12.5 82.8 0.0 4.4 0.3 0.0

India 35.2 38.7 0.0 0.0 25.9 0.2

Indonesia 21.6 50.8 4.2 0.8 22.6 0.0

Kazakhstan 39.3 42.1 7.1 1.0 0.0 10.5

Korea, Rep. of 42.0 39.1 3.9 8.5 5.2 1.3

Kyrgyz Republic 0.0 57.1 0.0 0.0 0.0 42.9

Malaysia 25.3 66.8 0.5 4.0 3.3 0.0

Myanmar 10.6 79.0 0.1 0.0 8.9 1.5

Nepal 4.3 40.8 8.9 9.2 36.9 0.0

Pakistan 14.5 76.9 1.8 1.8 5.1 0.0

Philippines 13.3 65.0 2.2 11.4 8.0 0.0

Singapore 38.8 45.4 0.0 0.0 0.0 15.8

Sri Lanka 13.5 70.9 0.3 1.8 4.2 9.2Taipei,China 48.1 42.6 2.5 2.0 3.7 1.0

Tajikistan 0.0 89.8 0.0 0.0 0.0 10.2

Thailand 22.7 62.2 10.2 0.0 4.9 0.0

Turkmenistan 0.0 26.0 0.0 0.0 0.0 74.0

Uzbekistan 9.5 60.1 23.6 0.0 0.6 6.3

Viet Nam 2

asian development bank- the challenge of higher oil prices

Documents