RDP 2010-08: Sources of Chinese Demand for Resource Commodities Appendix E: Data Used in Regressions

Ivan Roberts and Anthony Rush

November 2010

Table E1: Additional Data Descriptions and Sources
Variable Description and source
Value of bilateral non-oil resource imports The sum of annual bilateral imports of ‘metal ores’ (SITC code 28) and ‘coke and coal’ (SITC code 32) in US dollars. As both ‘metal ores’ and ‘coke and coal’ SITC codes are consistent across SITC revision 2 and SITC revision 3 classifications, the dataset comprises all bilateral import observations reported under either SITC revision over the period 1980–2008.
Source: United Nations COMTRADE database
Resource prices Source: IMF Primary Commodity Prices database
Volume of non-oil resource imports The value of bilateral non-oil resource imports by country i from country j in year t divided by resource prices. Metal ore sub-components are deflated by a comparable IMF commodity price index where one is available, with 2005 chosen as a base year. For metal ores that lack a corresponding price index (such as manganese), we deflate their nominal trade value by the IMF metal price index. Coke and coal imports are deflated by the IMF coal price index. The subcomponents are then weighted together to create ‘real’ volumes of bilateral resource imports.
Real GDP Annual real GDP in 1990 prices.
Source: United Nations National Accounts Main Aggregates Database, ‘GDP and its Breakdown at Constant 1990 Prices in US Dollars, All countries for all years – sorted alphabetically’ excel file. Available at <http://unstats.un.org/unsd/snaama/dnltransfer.asp?fID=6>. October 2009 vintage data accessed.
The UN real GDP data are reported differently to the COMTRADE trade data for a few economies. Specifically, UN COMTRADE aggregates trade data for Belgium and Luxembourg into the one region over 1980–1998 (but record trade data for the two countries separately thereafter), whereas GDP data are recorded separately for the two countries in all years. Similarly, trade data for Botswana, Lesotho, Namibia, South Africa and Swaziland are aggregated into the South African Customs Union region over 1980–1999, but real GDP data are reported for each country individually. From 2000, both trade and GDP data are reported separately for each of the five countries. As Belgium is a sizeable importer and South Africa a major exporter of resources, in years where the trade data has been aggregated we also aggregate GDP across the two regions. We then identify the two regions as distinct economies in all regressions.
While trade data are reported for the former USSR, former Yugoslavia and former Czechoslovakia as aggregated regions over 1980–1991, comparable real GDP data for these regions could not be obtained. Similarly, trade data are reported for East and West Germany separately in the years 1980–1990, but consistent real GDP data were not available. Thus, trade data for these economies have been omitted from the regressions.
Population Adjustments were made for Belgium and Luxembourg, and the South African Customs Union to ensure consistency with the real GDP data.
Source: Angus Maddison's ‘Statistics on World Population, GDP and Per Capita GDP, 1–2008 AD’. Data available at <http://www.ggdc.net/maddison/>. Site accessed December 2009.
Real exports Real exports in 1990 prices, with adjustments made for Belgium and Luxembourg, and the South African Customs Union to ensure consistency with the real GDP data.
Source: see Real GDP
Real imports Real imports in 1990 prices, with adjustments made for Belgium and Luxembourg, and the South African Customs Union to ensure consistency with the real GDP data.
Source: see Real GDP
Real investment Real investment in 1990 prices, with adjustments made for Belgium and Luxembourg, and the South African Customs Union to ensure consistency with the real GDP data.
Source: see Real GDP
Table E2: List of Countries(a)
Afghanistan(b) Central African Republic Ghana Liberia
Albania Chad(b) Greece Libya
Algeria Chile Grenada Lithuania
Angola(b) China Guatemala Luxembourg(c)
Antigua and Barbuda Colombia Guinea Macedonia
Argentina Comoros(b) Guinea-Bissau(b) Madagascar
Armenia Congo (Brazzaville) Guyana Malawi
Australia Costa Rica Haiti(b) Malaysia
Austria Côte d'Ivoire Honduras Maldives
Azerbaijan Croatia Hong Kong Mali
Bahamas, The Cuba Hungary Malta
Bahrain Cyprus Iceland Mauritania
Bangladesh Czech Republic India Mauritius
Barbados Denmark Indonesia Mexico
Belarus Djibouti Iran Moldova
Belgium(c) Dominica Iraq(b) Mongolia
Belgium and Luxembourg(c) Dominican Republic Ireland Montenegro(b),(h)
Belize Ecuador Israel Morocco
Benin Egypt Italy Mozambique
Bermuda El Salvador Jamaica Namibia(g)
Bolivia Equatorial Guinea(b) Japan Nepal
Bosnia and Herzegovina Eritrea(d) Jordan Netherlands
Brazil Estonia Kazakhstan New Zealand
Brunei Ethiopia(d) Kenya Nicaragua
Bulgaria Fiji Kiribati Niger
Burkina Faso Finland Kuwait Nigeria
Burundi France Kyrgyzstan Norway
Cambodia Gabon Laos(b) Oman
Cameroon Gambia, The Latvia Pakistan
Canada Georgia Lebanon Palau(b),(i)
Cape Verde Germany(e) Lesotho(f),(g) Panama
Papua New Guinea Sao Tome and Principe Spain Turkey
Paraguay Saudi Arabia Sri Lanka Turkmenistan
Peru Senegal Sudan Uganda
Philippines Serbia(h) Swaziland(g) Ukraine
Poland Seychelles Sweden United Arab Emirates
Portugal Sierra Leone(b) Switzerland United Kingdom
Qatar Singapore Syria United States
Romania Slovakia Tajikistan(b) Uruguay
Russia Slovenia Tanzania Uzbekistan(b)
Rwanda Solomon Islands Thailand Venezuela
Saint Kitts and Nevis(j) Somalia(b) Togo Vietnam
Saint Lucia South Africa(g) Tonga Yemen(e)
Saint Vincent and the Grenadines South African Customs Union(g) Trinidad and Tobago Zambia
Samoa South Korea Tunisia Zimbabwe
Notes:
(a) Data span 1980–2008. While data were available for Bhutan, Botswana, Suriname, Timor-Leste, and Vanuatu, because residuals for these countries were particularly large they were treated as outliers and therefore excluded from the regressions. However, the results for all equations remain robust to their inclusion.
(b) Resource importer only.
(c) Data for ‘Belgium’ and ‘Luxembourg’ span 1999–2008, while ‘Belgium and Luxembourg’ span 1980–1998.
(d) Data span 1993–2008.
(e) Data span 1991–2008.
(f) Resource exporter only.
(g) ‘South African Customs Union’ data span 1980–1999. Data for Lesotho, Namibia, South Africa and Swaziland span 2000–2008.
(h) Data span 2006–2008.
(i) Data span 1992–2008.
(j) Data span 1981–2008.