Appendix 4: Commodity price series data

(1)

Appendix 1: Share of 3 leading commodities in total exports

Country Principal exports Share of exports in 1990

1 2 3 1 2 3 54

Afghanistan Hides Wool Cotton 13 5 2 21

Algeria Petroleum 70 70

Argentina Petroleum Soybean meal Wheat 8 7 7 53

Bahrain Petroleum Aluminium 55 15 70

Bangladesh Shrimp Jute Tea 10 7 2 19

Barbados Sugar 22 23

Benin Cotton Petroleum Bananas 73 9 4 88

Bolivia Zinc Silver Hardwood sawn 16 6 4 42

Botswana Beef Copper 4 2 6*

Brazil Iron ore Soybean meal Aluminium 8 5 4 31

Burkina Faso Cotton Hides 57 6 63

Burundi Coffee Cotton Tea 25 4 2 33

Cambodia Soybeans Jute 8 5 13

Cameroon Petroleum Coffee Cocoa beans 50 8 7 88

Cape Verde Petroleum Bananas Shrimp 65 7 2 74

Chad Cotton Hides 50 1 52

Chile Copper Gold Iron ore 47 3 2 55

Colombia Petroleum Coffee Coal 29 21 8 68

Congo Petroleum Hardwood logs Hardwood sawn 92 1 1 95

Costa Rica Coffee Beef Sugar 17 3 2 24

Côte d’Ivoire Cocoa beans Coffee Hardwood sawn 36 8 5 59

Cuba Sugar Copper Shrimp 77 4 3 86

Dominican Rep. Sugar Cocoa beans Coffee 7 2 1 15

DR Congo Copper Coffee Petroleum 39 15 13 73

Ecuador Petroleum Bananas Shrimp 52 17 13 90

Egypt Petroleum Cotton 29 8 40

El Salvador Coffee Sugar Shrimp 44 4 3 51

Ethiopia (incl. Eritrea) Coffee Hides Sugar 45 21 6 71

Gabon Petroleum Manganese Hardwood logs 89 6 3 99

Gambia Groundnuts Groundnut oil Cotton 20 14 6 41

Ghana Gold Cocoa beans Hardwood sawn 43 23 6 84

Guatemala Coffee Sugar Bananas 28 11 6 60

Guinea Coffee Gold Cotton 6 2 1 10

Guinea-Bissau Bananas Shrimp Petroleum 51 23 5 83

Haiti Coffee Cocoa beans 10 1 12

Honduras Bananas Coffee Shrimp 32 26 6 75

India Tea Iron ore Cotton 3 3 2 18

Indonesia Petroleum Plywood Rubber 29 10 3 54

Iran Petroleum Hides 88 1 90

(2)

1 2 3 1 2 3 54

Iraq Petroleum 90 90

Jamaica Aluminium Sugar Bananas 64 7 3 75

Jordan Phosphates Potassium 20 13 33

Kenya Tea Coffee Sisal 20 14 1 39

Kuwait Petroleum 92 93

Lao PDR Hardwood logs Coffee 34 18 52

Lesotho Wool Wheat 9 3 12

Liberia Iron ore Rubber Hardwood logs 55 28 8 97

Madagascar Shrimp Coffee Sugar 13 13 7 39

Malawi Tobacco Tea Sugar 68 11 7 91

Malaysia Petroleum Hardwood logs Rubber 15 4 4 37

Mali Cotton Groundnuts Cottonseed oil 62 2 1 65

Mauritania Shrimp Iron ore Petroleum 43 39 3 85

Mauritius Sugar 29 30

Mexico Petroleum Copper Coffee 36 2 1 43

Mongolia Copper Hides Wool 5 2 57

Morocco Phosphates Superphos. Shrimp 10 9 5 28

Mozambique Shrimp Sugar Cotton 42 7 6 58

Nicaragua Coffee Beef Sugar 22 20 12 85

Niger Tobacco Cotton Hides 2 2 1 5*

Nigeria Petroleum Cocoa beans 97 1 98

Oman Petroleum Copper 92 1 93

Pakistan Cotton Rice Shrimp 8 4 1 16

Panama Bananas Shrimp Sugar 26 15 11 62

Paraguay Cotton Soybeans Beef 34 28 14 87

Peru Copper Petroleum Zinc 23 10 10 55

Philippines Copper Coconut oil Shrimp 6 4 3 19

Qatar Petroleum 79 79

Rwanda Coffee Tea Tin 61 7 2 73

Sao Tome & Principe Cocoa beans 83 83

Saudi Arabia Petroleum 86 87

Senegal Groundnut oil Phosphates Groundnuts 18 8 2 36

Sierra Leone Aluminium Coffee Cocoa beans 19 6 4 29

Somalia Lamb Bananas Sugar 46 23 2 74

South Africa Coal Copper Iron ore 7 2 1 13

Sri Lanka Tea Rubber 26 4 32

St Lucia Bananas Coconut oil 58 2 61

Sudan Cotton Lamb Groundnuts 43 5 4 56

Swaziland Sugar Beef Cotton 34 1 1 36

Syria Petroleum Cotton 45 4 50

Tanzania Coffee Cotton Sugar 19 18 13 62

Thailand Rice Shrimp Rubber 5 5 4 18

Togo Phosphates Cotton Coffee 44 21 7 85

(3)

1 2 3 1 2 3 54

Trinidad & Tobago Petroleum Sugar 68 1 70

Tunisia Petroleum Superphos. Shrimp 17 3 2 24

Turkey Tobacco Petroleum Cotton 3 2 1 10

Uganda Coffee Maize Hides 76 6 2 89

United Arab Emirates Petroleum 70 71

Uruguay Beef Wool Rice 13 7 6 32

Venezuela Petroleum Aluminium Iron ore 80 4 2 87

Vietnam Rice Coffee Rubber 13 4 3 25

Yemen Petroleum Shrimp 90 6 98

Zambia Copper Sugar Tobacco 86 1 1 89

Zimbabwe Tobacco Maize Cotton 23 7 6 50

* the major export product of these countries (diamonds for Botswana, uranium for Niger) are not included in this research due to unavailability of data.

Column 54 indicates the total share of all 54 commodities included in this research in total exports in 1990.

Source: Author’s own calculations

(4)

Appendix 2: Countries and countr y groupings

The following countries 93 were included in this research. The countries are classified into groups based on geographic location and their predominant exports.

A country is labelled as an exporter of a particular type of resource if this resource constitutes at least 50 percent of the total commodity exports. If no single type of resource accounts for 50 percent of exports, the country is labelled a ‘mixed’ exporter.

Small islands are reported in italics, landlocked countries in bold.

Source: Author Type of

exporter Sub-Saharan Africa Middle East and

North Africa Asia-Pacific Latin America Caribbean

Agricultural food

Agricultural non-food

Minerals

Oil

Mixed

Botswana Burundi Cote d’Ivoire Ethiopia (incl.

Eritrea) Gambia Guinea Guinea-Bissau Kenya Madagascar Mauritius Mozambique Rwanda Sao Tome &

Principe Senegal Somalia Swaziland Tanzania Uganda

Benin Burkina Faso Lesotho Malawi Mali Niger Zimbabwe

DR Congo Ghana Liberia Sierra Leone South Africa Togo Zambia

Cameroon Cape Verde Congo Gabon Nigeria

Mauritania

Chad Sudan

Jordan Morocco

Algeria Egypt Iraq Oman Saudi-Arabia Tunisia Yemen

Turkey

Bahrain Iran Kuwait Qatar Syria UAE

Bangladesh Cambodia India Philippines Sri Lanka Thailand Vietnam

Afghanistan Lao PDR

Mongolia

Indonesia

Malaysia Pakistan

Argentina Brazil Costa Rica El Salvador Guatemala Honduras Nicaragua Panama Paraguay Uruguay

Bolivia Chile Peru

Ecuador Mexico Venezuela

Colombia

Barbados Cuba

Dominican Republic Haiti

St Lucia

Jamaica

Trinidad & Tobago

(5)

Appendix 3: Commodities included in country indices

Category Name SITC code* Price data source Notes

Agricultural – food Bananas 0573 UNCTAD

Beef 0111 IFS

Butter 023 IFS

Cocoa beans 0721 IFS & UNCTAD 2 series

Coconut oil 4243 IFS

Coffee 0711 IFS 3 series

Cottonseed oil 4233 UNCTAD

Groundnut oil 4234 IFS

Groundnuts 2221 IFS

Lamb 0112 IFS

Maize 044 IFS

Palm kernel oil 4244 UNCTAD

Palm oil 4242 IFS

Pepper 0751 UNCTAD

Rice 042 IFS 2 series

Shrimp 036 IFS

Sorghum 04592 IFS

Soybean meal 08131 IFS

Soybean oil 4232 IFS

Soybeans 2222 IFS

Sugar 061 IFS 3 series

Sunflower seed oil 4236 IFS

Tea 0741 IFS

Wheat 041 + 046 IFS 2 series

Agricultural – non-food Copra 2231 IFS

Cotton 2631 IFS

Fish meal 03501 IFS

Hardwood logs 2472 IFS

Hardwood sawn 2483 IFS

Hides 21 IFS

Jute 264 IFS

Linseed oil 4241 IFS

Newsprint 6411 IFS & FAO

Plywood 6342 IFS

Rubber (natural) 232 IFS 2 series

(6)

Category Name SITC code* Price data source Notes

Agricultural – non-food Sisal 2654 IFS

contd. Tobacco 121 IFS

Wool 2681 + 2682 IFS

Minerals Aluminium 2873 + 6841 IFS

Coal 32 IFS

Copper 28711 + 28712 + 6821 IFS

Gold 97 IFS

Iron ore 2815 + 2816 IFS

Lead 2874 IFS

Manganese 2877 UNCTAD

Nickel 28721 IFS

Phosphate rock 2713 IFS

Potash 2714 IFS

Silver 6811 IFS

Superphosphate 56222 IFS

Tin 2876 IFS 3 series

Tungsten 28792 UNCTAD

Zinc 2875 IFS 2 series

Oil Petroleum 333 IFS

* Based on Standard International Trade Classification (SITC) Revision 2

(7)

Appendix 4: Commodity price series data

Aluminum: London Metal Exchange, standard grade, spot price, minimum purity 99.5 percent, c.i.f. U.K. ports (Wall Street Journal, New York, and Metals Week, New York).3 Prior to 1979, U.K. producer price, minimum purity 99 percent (Metal Bulletin, London). (15676drz)

Bananas: Latin America: Central America and Ecuador, first class quality tropical pack, Chiquita, Dole and Del Monte, U.S.

importer’s price f.o.b. U.S. ports (Sopisco News, Guayaquil).1 (24876u.z)

Beef: Australia/NZ (U.S. Ports): frozen boneless, 85 percent visible lean cow meat, U.S. import price, f.o.b. U.S. port of entry (The Yellow Sheet, The National Provisioner Daily Market and News Service, Chicago, Illinois).3 Prior to December 1975, 90 percent visible. (19376kbz)

Butter: New Zealand (London). (19676flz)

Coal: Australia: Thermal coal. 12000 btu/pound, less than 1% sulfur, 14% ash, f.o.b. piers, Newcastle/Port Kembla (World Bank). (19374VRZ)

Cocoa Beans: Average of the daily prices of the nearest three active future trading months on the London Terminal Market and on the New York Coffee, Sugar and Cocoa Exchange at time of the London close (Article 26 of the International Cocoa Agreement, 1993). The average price expressed in US dollars is converted into SDR equivalent at the appropriate daily official US dollar/SDR exchange rate published by the IMF. Estimates for the years 1971 to 1985 are calculated using the monthly averages of the US dollar/SDR exchange rate. (International Cocoa Organization, London). (65276r.zM44)

Coconut Oil: Philippines/Indonesia (New York)*: Philippines (DataStream). (56676aiz)

Coffee: Other milds: Arithmetic average of El Salvador Central Standard, Guatemala prime washed, Mexico prime washed, prompt shipment, ex-dock, New York.3 (38676ebz)

Brazil (New York): Unwashed arabica, Santos No. 4, ex-dock, New York.3 (22376ebz).

Uganda (New York)*: Robusta: New York cash price. Côte d’Ivoire Grade II, and Uganda Standard. Prompt shipment, ex-dock, New York. Prior to July 1982, arithmetic average of Angolan Ambriz and 2AA and Ugandan Native Standard (Patton’s Complete Coffee Coverage, New York).3 (79976ecz)

Copper: United Kingdom: London Metal Exchange, grade A cathodes, spot price, c.i.f. European ports (Wall Street Journal, New York, and Metal Bulletin, London).3 Prior to July 1986, higher grade, wire bars or cathodes. (11276c.z)

Copra: Philippines: Phil/Indo, c.i.f. Northwest European ports (Oil World, Hamburg).1 (56676agz)

Cotton: Liverpool Index: Midd. 13/32 inches, Liverpool Index ‘A’, average of the cheapest fourteen of ten styles; c.i.f. Liverpool (Cotton Outlook, Liverpool from January 1968 to May 1981 strict middling, SM 11/16 inches; prior to 1968, Mexican SM 11/16.1 (11176f.zM40)

Cottonseed oil: United States, Prime Bleachable Summer Yellow (P.B.S.Y.), FOB Gulf. Prior to October 1994, United States, P.B.S.Y., CIF Rotterdam. (Oil World, Hamburg).

Fish meal: Any origin, 64/65% protein, cost and freight, Hamburg (calculated from the Hamburg wholesale price).(Oil World, Hamburg).

Gold: United Kingdom: 99.5 percent fine. London, afternoon fixing. 3 (11276krz)

Groundnuts: Any origin: U.S. Runner, 40–50 percent shelled basis, c.i.f. Rotterdam. (69476bhz) Groundnut Oil: Any Origin: c.i.f. Rotterdam (Oil World, Hamburg).

Hardwood Logs: Malaysia, meranti, Sarawak best quality, sale price charged by importers, Japan (World Bank, Washington, D.C.). From January 1988 to February 1993, average of Sabah and Sarawak in Tokyo weighted by their respective import volumes in Japan. From February 1993 to present, Sarawak only.2 (54876vxz)

Hardwood Sawnwood: Malaysian sawnwood, dark red meranti, select and better quality, standard density, c.i.f. U.K. Port Tropical Timbers, Surrey, England).2 (54876rmz)

Hides: United States: Wholesale dealer’s price, packer’s heavy native steers, over 53 lbs. (formerly over 58 lbs.), Chicago, f.o.b. shipping point (Wall Street Journal, New York).3 Prior to November 1985, U.S. Bureau of Labor Statistics, Washington, D.C. (11176p.z)

(8)

Iron Ore: Brazil: Carajas fines, 67.55 percent FE (iron) content, contract price to Europe, f.o.b. Ponta da Madeira (Companhia Vale de Rio Doce, Rio de Janeiro, Brazil).1 (22376gaz)

Jute: Raw Bangladesh BWD, f.o.b. Chittagong/Chalna (World Bank, Washington, D.C. Beginning 1977, UNCTAD source.)2 (51376x.z)

Lamb: New Zealand (London)*: PL, frozen, wholesale price at Smithfield Market, London (National Business Review, Auckland, New Zealand). (19676pfz)

Lead: United Kingdom: London Metal Exchange, 99.97 percent pure, spot, c.i.f. European ports (Wall Street Journal, New York, and Metals Week, New York).3 (11276v.z) United States: Common grade domestic pigs in New York. (00176v.z)

Linseed Oil: Any origin, ex-tank Rotterdam (Oil World, Hamburg). (11176niz)

Maize: United States (U.S. Gulf Ports): U.S. No. 2 yellow, prompt shipment, f.o.b. Gulf of Mexico ports (USDA Grain and Feed Market News, Washington, D.C.).3 (11176j.zM17)

Manganese ore: 48/50% Mn, max 0.1%P, metallurgical FOB. (Metal Bulletin, London).

Newsprint: Finland (unit value). (17274ulz).

Nickel: United Kingdom: London Metal Exchange, melting grade, spot, c.i.f. North European ports (Wall Street Journal, New York, and Metals Week, New York). Prior to 1980, INCO price, c.i.f. Far East and American ports (Metal Bulletin, London).3 (15676ptz)

Palm kernel oil: Malaysia, CIF Rotterdam: Prior to September 1980: Dutch, FOB ex-mill. (Oil World, Hamburg).

Palm Oil: Palm Oil Futures (first contract forward) 4–5 percent FFA (Bursa Malaysian Derivatives Berhad).1 Prior to 1974, UNCTAD. 2 (54876dgz)

Malaysia (unit value). (54874dgz)

Pepper: White Sarawak/Muntok, fair average quality, spot London. (The Public Ledger, London, United Kingdom). Prior to January 2003: white Sarawak, Singapore market, fair average quality, main European ports. (Market News Service, ITC, UNCTAD/WTO, Geneva).

Petroleum: Average Crude Price: U.K. Brent (light), Dubai (medium), and West Texas Intermediate, equally weighted.

† Prior to 1983, Alaska North Slope (heavy) was used in the composition of this line instead of West Texas Intermediate. Annual data through 1994 for this earlier version are available in the 1995 IFS yearbook, with quarterly and monthly data available in the monthly issues through the January 1996 issue. (00176aaz)

Phosphate Rock: Morocco (Casablanca): 70 percent BPL, contract, f.a.s. (British Sulphur Monthly Newsletter). Prior to 1981, 72 percent BPL, f.a.s.1 (68676awz)

Potash: Canada: Muriate of potash, f.o.b., Vancouver. Average of daily quotations. (15676qrz)

Plywood: Philippines: Lauan, 3-ply, extra, 91 cm x 182 cm x 4 mm, wholesale price, spot, Tokyo. (56676wxz)

Rice: Thailand: White milled 5 percent broken, nominal price quotes, f.o.b. Bangkok (USDA Rice Market News, Little Rock, Arkansas).1 (57876n.zM81)

Thailand (unit value). (57874n.z)

Rubber: Malaysia: No. 1 R.S.S., prompt shipment, f.o.b. Malaysian/Singapore ports (The Financial Times, London).3 (548761.z)

Thailand (unit value). (57874l.z)

Shrimp: United States: Mexican, west coast, white, No. 1 shell on, headless, 26 to 30 count per pound, wholesale price at New York (World Bank). (11176blz)

Silver: United States: 99.9 percent grade refined, Handy and Harman, New York. Average of daily quotations. (11176y.z) Sisal: East African, ungraded, c.i.f. European ports (UNCTAD).2 (63976mlz)

Sorghum: United States: No. 2 yellow, prompt shipment, f.o.b. Gulf of Mexico ports (USDA Grain and Feed Market News, Washington, D.C.) (11176trz)

Soybeans: Soybean futures contract (first contract forward) No. 2 yellow and par: (Chicago Board of Trade). (11176jfz) Soybean Meal: Soybean Meal Futures (first contract forward) Minimum 48 percent protein (Chicago Board of Trade). (11176jjz) Soybean Oil: Soybean Oil Futures (first contract forward) exchange approved grades (Chicago Board of Trade). Prior to April 1973, Dutch crude oil, ex-mill. (11176jiz)

(9)

Sugar: EU Import Price: Unpacked sugar, c.i.f. European ports. Negotiated export price for sugar from ACP countries to EU under the Sugar Protocol (Lomé Convention). (EU office, Washington, D.C.).3 (11276i.z)

Free Market: CSCE contract No. 11, nearest future position (Coffee, Sugar and Cocoa Exchange, New York Board of Trade).

(00176iaz)

U.S. Import Price: CSCE contract No. 14, nearest future position (Coffee, Sugar and Cocoa Exchange, New York Board of Trade). (11176iazM02)

Sunflower Seed Oil: Sunflower Oil, US export price from Gulf of Mexico (DataStream). (11276soz)

Superphosphate: United States (U.S. Gulf Ports): Triple-superphosphate, bulk, spot, f.o.b. Gulf of Mexico Ports (Fertilizer Week, CRU International Ltd., London).3 (11176asz)

Tea: Average Auction (London): Mombasa auction price for best PF1, Kenyan Tea. Replaces London auction price beginning July 1998.3 (11276s.z)

Tin: Any Origin (London): London Metal Exchange, standard grade, spot, c.i.f. European ports (Wall Street Journal, New York).

From December 1985 to June 1989, Malaysian Straits, minimum 99.85 percent purity, Kuala Lumpur Tin Market settlement price. Prior to November 1985, London Metal Exchange (Wall Street Journal, New York and Metals Week, New York).3 (11276q.z)

Bolivia (unit value). (21874q.z)

Malaysia (unit value): Primary tin. (54874q.z)

Tobacco: U.S. Import Unit Value of general unmanufactured tobacco. (USDA, Foreign Agricultural Service). (11176m.z) Tungsten ore: Wolframite and scheelite, minimum content of WO3: 65%, CIF Europe. Prior to April 1992: wolfram. (Metal Bulletin, London).

Wheat: United States: No. 1, hard red winter, ordinary protein, prompt shipment, f.o.b. Gulf of Mexico ports (USDA Grain and Feed Market News, Washington, D.C.)3 (11176d.z)

Argentina (unit value). (21374d.z)

Wool: Australia-New Zealand 48’s: Coarse wool, 23 micron (AWEX, Australian Wool Exchange) Sydney, Australia. (11276hdz) Zinc: United Kingdom: London Metal Exchange, high grade cash, c.i.f. U.K. ports, 98 percent pure, spot (Wall Street Journal, New York, and Metals Week, New York).3 Prior to January 1987, standard grade. (11276t.z)

Bolivia (unit value). (21874t.z)

1 Average of weekly quotations.

2 Monthly quotations.

3 Average of daily quotations.

Source: IFS database, International Monetary Fund; Handbook of Statistics 2005, UNCTAD

(10)

Appendix 5: Commodity price index construction

Quarterly data on commodity prices was obtained from the International Monetary Fund’s IFS database and UNCTAD’s Handbook of Statistics 2005. The set of nominal price data consists of 65 price series on 54 different commodities. A number of commodities thus have multiple price series in the data set, which are used to more accurately approximate the actual export prices paid for the resource to a certain country. For instance, there are 3 series for tin prices, of which the ‘Tin Malaysia’

series is used to value the tin exports of Malaysia, the ‘Tin Bolivia’ series for Bolivia, and the ‘any origin’ series for all other tin-exporting countries.

Unfortunately, some of the price series were not complete. In several cases, there were missing observations at the beginning of the sample. Since identical sample length is important when measuring volatility, in cases where too many observations were missing, a trade-off had to be made between dropping the commodity from the data set and shortening the time series for all commodities.

Here, I used the following rule: only if the missing data constituted less than 5 years, the commodity was retained in the index and the missing values were generated. The method used for this is described below.

A small number of commodities were initially selected but then dropped from the index due to lack of data. These are barley, natural gas, olive oil, poultry, wood pulp, softwood logs, softwood sawn, uranium, and urea. The absence of natural gas and uranium in the indices are the most flagrant ones, as is the omission of raw diamonds for which no price series is available. These missing commodities potentially have a large impact on the price indices for some countries, as mentioned earlier.

Fortunately, the available price series are rather complete. In the cases of coal and tobacco, the first 2 resp. 3 years were missing; these were replaced by the first available observation. In the case of newsprint, data was missing at the end of the sample. Here, the omissions were filled using data obtained from the FAO’s FAOSTAT database. Price series in the UNCTAD handbook only run until 2004, hence for the 7 series from this source the missing values for 2005 are held constant at the 2004 values. The total number of missing quarterly observations is 60 out of a total sample of 2665¹, thus amounting to about 2 percent of the observations.

In order to obtain real prices, the nominal price series were deflated using the MUV index. This is an index of the unit value of manufactured exports from developed countries (1990=100) which is obtainable from the World Bank. It is commonly used as a deflator for commodity price series; see for instance Grilli and Yang (1988), Dehn (2000) and Cashin, Liang and McDermott (2000).

The export data used to construct the weights are annual fob in 1990 US dollars. They are derived from the UN’s COMTRADE database, which was accessed through the World Integrated Trade Solution (WITS). This data was complemented by trade statistics from the UNCTAD Commodity

1164 quarters times 65 commodity price series

(11)

Yearbook 1994, its Handbook of International Trade and Development Statistics, and the 1993 and 1994 issues of the International Trade Statistics Yearbook, published by the UN Statistical Office.

All country’s trade statistics for the base years were checked manually to identify any important omissions or odd-looking figures. For the UAE, there was an unusually low value for crude petroleum exports, which was corrected using OPEC trade statistics. Trade data for Rwanda revealed that this country exported US$ 26 million worth of gold in 1990, and Burundi reported exports of gold worth US$ 96 million. However, both countries had no significant gold mining operations at the time. It is widely believed that the gold was illegally obtained from the DR Congo, and brought across the border by smugglers or armed forces². Although revenues from these exports did in fact accrue to Rwanda and Burundi, they have not been consistent over the past decades; therefore these entries were removed from the data.

Contrary to Dehn (2000), I chose to leave some countries out of the sample because their 1990 export data would clearly not be representative for the whole period under consideration. For instance, Equatorial Guinea was omitted because it did not yet export petroleum in 1990, whereas that commodity currently accounts for 90 percent of the country’s export earnings. So-called ‘refinery countries’ that export petroleum only because they import crude oil for refinement were also deleted.

These include Singapore, South Korea, Aruba and the Netherlands Antilles. Another small number of countries were removed in a later stage, either because their resource exports constituted less than 10 percent of their total exports, or because no data was available on GDP per capita or control variables.

For some countries, it was not possible to find export weights for 1990. In these cases a different base year was chosen, preferably as close to 1990 as possible. The countries with different base years are:

1988: Liberia, Sudan

1988-1989 average: Afghanistan, Chad, Cuba, DR Congo, Iraq, Iran, Mauritania, Somalia

1991: Nigeria, Yemen

1992: Ghana

1993: Benin, Burundi, Congo, Dominican Rep., Gabon, South Africa, Zambia

1994: Bahrain, Mozambique, Uganda

1995: Guinea, Guinea-Bissau

1996: Mongolia

Price data and export weights are used to construct the quarterly country-specific price indices according to the Deaton-Miller price index as outlined by equation [1]. Then, using EViews these indices are separated into trend and volatility around the trend by the Hodrick-Prescott filter. The quarterly data is then first aggregated into annual data. Annual trend growth is taken as the percentage change in the trend HP-filtered variable over a given year. Annual volatility is calculated as

2For more, see Olsson and Congdon Fors (2004).

(12)

the average standard deviation of the quarterly volatility values in a given year; hence, it is a measure of intra-year volatility. The dataset containing annual observations is used for some of the descriptive statistics in this paper, for instance the tables in appendices 7 and 8. For the regressions, an even higher degree of aggregation was used in order to capture long-term effects; therefore, a 5-year dataset was constructed. For the price trend variable, I take the growth rate in the trend between the last quarter of the previous 5-year period and the last quarter of the current period. For the volatility variable, I take the average of the volatility around the trend in all quarters in the 5-year period as the average volatility for that period.³

3The last period contains six years.

(13)

Appendix 6: Commodity price indices per country

Left axis depicts volatility (green line), right axis indicates price index (blue line) and trend (red line). Graphs are drafted in EViews v5.0 on the basis of quarterly data. Please note: Y-axis differs between graphs.

Source: Author’s own calculations

-60 -40 -20 0 20 40 60

40 60 80 100 120 140 160

1965 1970 1975 1980 1985 1990 1995 2000 2005 AFGHANISTAN Trend Cycle Hodrick-Prescott Filter (lambda=1600)

-80 -40 0 40 80 120

0 50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

ALGERIA Trend Cycle

Hodrick-Prescott Filter (lambda=1600)

-40 -20 0 20 40 60 80

40 80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

ARGENTINA Trend Cycle

-80 -40 0 40 80

40 80 120 160 200 240 280

1965 1970 1975 1980 1985 1990 1995 2000 2005

BAHRAIN Trend Cycle

-40 -20 0 20 40 60

40 80 120 160 200

1965 1970 1975 1980 1985 1990 1995 2000 2005

BANGLADESH Trend Cycle

-100 0 100 200 300 400 500

0 200 400 600 800

1965 1970 1975 1980 1985 1990 1995 2000 2005

BARBADOS Trend Cycle

-50 0 50 100

50 100 150 200 250

1965 1970 1975 1980 1985 1990 1995 2000 2005

BENIN Trend Cycle

-30 -20 -10 0 10 20 30

40 80 120 160 200

1965 1970 1975 1980 1985 1990 1995 2000 2005 BOLIVIA Trend Cycle Hodrick-Prescott Filter (lambda=1600)

-80 -40 0 40 80

40 80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

BOTSWANA Trend Cycle

-40 -20 0 20 40 60

80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

BRAZIL Trend Cycle

-80 -40 0 40 80 120

50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

BURKINAFASO Trend Cycle

-200 -100 0 100 200 300

0 200 400 600 800

1965 1970 1975 1980 1985 1990 1995 2000 2005

BURUNDI Trend Cycle

(14)

-40 0 40 80 120

50 100 150 200 250 300 350

1965 1970 1975 1980 1985 1990 1995 2000 2005

CAMBODIA Trend Cycle

-80 -40 0 40 80

50 100 150 200 250

1965 1970 1975 1980 1985 1990 1995 2000 2005

CAMEROON Trend Cycle

-80 -40 0 40 80

0 50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

CAPEVERDE Trend Cycle

-80 -40 0 40 80 120

50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

CHAD Trend Cycle

-50 0 50 100

50 100 150 200 250

1965 1970 1975 1980 1985 1990 1995 2000 2005

CHILE Trend Cycle

-60 -40 -20 0 20 40 60

40 80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

COLOMBIA Trend Cycle

-80 -40 0 40 80 120

0 50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

CONGO Trend Cycle

-100 -50 0 50 100 150

0 100 200 300 400

1965 1970 1975 1980 1985 1990 1995 2000 2005

COSTARICA Trend Cycle

-100 -50 0 50 100 150

0 100 200 300 400 500

1965 1970 1975 1980 1985 1990 1995 2000 2005 COTEDIVOIRE Trend Cycle Hodrick-Prescott Filter (lambda=1600)

-100 0 100 200 300 400

0 100 200 300 400 500 600

1965 1970 1975 1980 1985 1990 1995 2000 2005

CUBA Trend Cycle

-50 0 50 100 150

0 100 200 300 400

1965 1970 1975 1980 1985 1990 1995 2000 2005

DOMINICANREP Trend Cycle

-40 0 40 80

40 80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

DRCONGO Trend Cycle

-40 -20 0 20 40 60

40 80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

ECUADOR Trend Cycle

-80 -40 0 40 80

40 80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

EGYPT Trend Cycle

-200 -100 0 100 200 300

0 100 200 300 400 500

1965 1970 1975 1980 1985 1990 1995 2000 2005

ELSALVADOR Trend Cycle

(15)

-100 -50 0 50 100 150

0 100 200 300 400

1965 1970 1975 1980 1985 1990 1995 2000 2005

ETHIOPIA Trend Cycle

-80 -40 0 40 80

0 50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

GABON Trend Cycle

-50 0 50 100

40 80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

GAMBIA Trend Cycle

-40 -20 0 20 40 60

40 80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

GHANA Trend Cycle

-80 -40 0 40 80

50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

GUATEMALA Trend Cycle

-100 -50 0 50 100 150

0 100 200 300 400

1965 1970 1975 1980 1985 1990 1995 2000 2005

GUINEA Trend Cycle

-30 -20 -10 0 10 20 30

60 80 100 120 140

1965 1970 1975 1980 1985 1990 1995 2000 2005 GUINEABISSAU Trend Cycle Hodrick-Prescott Filter (lambda=1600)

-200 -100 0 100 200 300

0 100 200 300 400 500 600

1965 1970 1975 1980 1985 1990 1995 2000 2005

HAITI Trend Cycle

-40 -20 0 20 40 60

40 80 120 160 200

1965 1970 1975 1980 1985 1990 1995 2000 2005

HONDURAS Trend Cycle

-40 -20 0 20 40

80 100 120 140 160 180 200

1965 1970 1975 1980 1985 1990 1995 2000 2005

INDIA Trend Cycle

-40 -20 0 20 40 60

40 80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005 INDONESIA Trend Cycle Hodrick-Prescott Filter (lambda=1600)

-80 -40 0 40 80 120

0 50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

IRAN Trend Cycle

-80 -40 0 40 80 120

0 50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

IRAQ Trend Cycle

-40 -20 0 20 40 60 80

40 80 120 160 200

1965 1970 1975 1980 1985 1990 1995 2000 2005

JAMAICA Trend Cycle

-50 0 50 100

50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

JORDAN Trend Cycle

(16)

-100 -50 0 50 100 150 200

0 100 200 300 400 500

1965 1970 1975 1980 1985 1990 1995 2000 2005

KENYA Trend Cycle

-80 -40 0 40 80 120

0 50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

KUWAIT Trend Cycle

-50 0 50 100

80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

LAOS Trend Cycle

-50 0 50 100

50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

LESOTHO Trend Cycle

-40 -20 0 20 40 60

80 100 120 140 160 180

1965 1970 1975 1980 1985 1990 1995 2000 2005 LIBERIA Trend Cycle Hodrick-Prescott Filter (lambda=1600)

-80 -40 0 40 80

50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

MADAGASCAR Trend Cycle

-20 -10 0 10 20 30

60 80 100 120 140 160 180

1965 1970 1975 1980 1985 1990 1995 2000 2005

MALAWI Trend Cycle

-40 -20 0 20 40 60

40 80 120 160 200 240

1965 1970 1975 1980 1985 1990 1995 2000 2005

MALAYSIA Trend Cycle

-80 -40 0 40 80 120

50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

MALI Trend Cycle

-40 -20 0 20 40

80 100 120 140 160

1965 1970 1975 1980 1985 1990 1995 2000 2005

MAURITANIA Trend Cycle

-40 -20 0 20 40 60

60 80 100 120 140 160 180

1965 1970 1975 1980 1985 1990 1995 2000 2005 MAURITIUS Trend Cycle Hodrick-Prescott Filter (lambda=1600)

-80 -40 0 40 80

0 50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

MEXICO Trend Cycle

-80 -40 0 40 80

50 100 150 200 250 300

1965 1970 1975 1980 1985 1990 1995 2000 2005

MONGOLIA Trend Cycle

-80 -40 0 40 80 120

80 120 160 200 240 280 320

1965 1970 1975 1980 1985 1990 1995 2000 2005

MOROCCO Trend Cycle

-40 -20 0 20 40

60 80 100 120 140 160 180

1965 1970 1975 1980 1985 1990 1995 2000 2005

MOZAMBIQUE Trend Cycle