Stanford Oil Wealth Management Dataset:

Transcription

1 : Benchmarking oil states success in developing, managing, and disbursing hydrocarbon revenues CODEBOOK AND DATA Christine Jojarth, Ph.D., Stanford University

2 Contents Acknowledgements and data structure Country selection Variable profiles and general conversions Individual variable data sheets Stage 1: Revenue generation Government hydrocarbon revenues per hydrocarbon output Competition in oil industry Operational efficiency of national oil company (NOC) Domestic gasoline price relative to US gasoline price Reserve to production ratios Resource to production ratios Stringency of environmental regulation Gas flared per oil output Stage 2: Revenue management Real growth in non-oil GDP Real growth in non-oil exports Consumer price inflation Volatility of official exchange rate Central government non-oil fiscal balance Central government fiscal balance Applied tariffs Global competitiveness Ease of doing business Economic freedom Stage 3: Revenue disbursement Human development index Education index Infant mortality Mainland phone line density Access to water Electric power transmission and distribution losses Road network density Infrastructure quality Logistics performance index Control of corruption Government effectiveness Management performance Stage of development relative to wealth References Data Page 2 of 42

3 Acknowledgements The Stanford Oil Wealth Management dataset is the result of a collaborative, interdisciplinary research project launched in October Over the course of these years, this project has benefited from a great number of people and organizations. First of all, this dataset would not have been possible without the generous financial support of the Revenue Watch Institute and of the Microsoft Corporation. Thomas Heller, Michael McFaul, Mark Thurber, and David Victor proved indispensable thought partners throughout the design stage. The participants of the conference Turning Oil Wealth into Development in May 2007 organized by Stanford s Center on Democracy, Development, and the Rule of Law (CDDRL) and of the February 2008 workshop of Stanford s Program on Energy and Sustainable Development (PESD) provided critical guidance in the difficult balancing act between selecting indicators that best capture the underlying conceptual ideas and choosing indicators with sufficient data coverage. Lindsay Rickey and Rebecca Schindel provided excellent research assistance. Page 3 of 42

4 1 and data structure The eightfold surge in oil prices in as many years has propelled oil into the center stage of politics, business, and ordinary citizens daily life around the world but nowhere are discussions more heated and the implications felt more directly than in oil exporting states. On the one hand, these states find themselves in an enviable position. They are awash with cash. OPEC s net oil export revenues are projected to exceed $1 trillion in 2008 almost twice as much as the record earnings of the year before (EIA 2008). Petro states are keen to use this financial momentum for a great leap forward, for a fundamental diversification of their economy that offers more high-value jobs in less volatile and more sustainable sectors. On the other hand, they are also well aware of the negative economic 1 and political 2 spillover effects heavy reliance on oil revenues can bring about and are eager to learn from historic experience and from their peers how to mitigate these problems. The Stanford Oil Wealth Management (SOWM) dataset provides an empirical foundation in support of oil states efforts to manage their oil wealth wisely, to maximize the social and economic development outcomes they achieve from the monetization of their hydrocarbon resources. For this purpose, the SOWM dataset distinguishes three stages in the oil well-to-hospital value chain where each stage presents its own particular governance challenges. The first stage, revenue generation, analyzes how the petroleum sector is organized. Does a country s regulatory and fiscal framework facilitate efficient and sustainable oil and gas production and assign the greatest possible share of so-generated rents to the government? The second stage, revenue management, examines the extent to which oil states manage to mitigate negative spill-over effects of the hydrocarbon sector on the non-oil economy. The third and final stage, revenue disbursement, sheds light on governments ability to maximize the social value of public spending. In each of the three stages, the Stanford Oil Wealth Management dataset distinguishes between outcome-related and policy-related indicators. The former highlight how much of the desired outcome a country manages to realize. How much money does the state get for every barrel of oil and gas that is being extracted (Stage 1)? How fast does the non-oil sector grow (Stage 2)? How developed is that country s physical infrastructure and human capital (Stage 3)? There are obviously a large number of factors that influence these outcome indicators, and only some of them are within a government s immediate control. For instance, the revenues a country collects from its hydrocarbon sector may be relatively low solely because its production costs are unusually high for geological and geophysical reasons. Its non-oil sector may be held back by an economic crisis rooted in its main trading partner. Developing infrastructure may be particularly cost-intensive in one country because of adverse climatic conditions or a vast, sparsely populated territory. In an attempt to control for such exogenous factors, the Stanford Oil Wealth Management dataset complements the outcome indicators with a series of policy indicators. An oil state can be considered to be a strong performer only if good out- 1 e.g. Dutch disease (Corden 1984, ), debt overhang (Manzano and Rigobon 2001) 2 e.g. violent conflict (Collier and Hoeffler 2005, ), non-democratic government (Ross 2001, ) Page 4 of 42

5 comes are backed by good policies, whereas the coexistence of strong outcome data and poor policies is a reason to look more closely into the possibility of outcomes having been boosted by unusually favorable exogenous factors. Revenue generation Revenue management Revenue disbursement GOAL Maximize oil revenues over full industry cycle Create conditions for non-oil sector competitiveness Maximize social return of public spending INDICATORS Outcome Policy Size of government take Non-oil sector growth Level of human capital and infrastructure development Efficiency of oil sector Sustainability of oil production Macroeconomic stability Business climate Quality of governance 2 Country selection The Stanford Oil Wealth Management dataset presents data for a total of 31 countries from around the world. These countries have in common that they are all oil and gas producers for whom the hydrocarbon sector constitutes as central pillar of their economy and/or government finances. By selecting countries based on their relative dependence on the hydrocarbon sector rather than on that sectors absolute magnitude (e.g. measured in terms of reserve size or production volumes) the Stanford Oil Wealth Management dataset follows mainstream research practice in resource economics (Sachs and Warner 1995; Ross 1999). The rationale for this selection method is that many of the economic and political challenges posed by the hydrocarbon sector are conditioned upon its size relative to the rest of the system. For instance, the upward pressure exerted on a country s currency from a massive expansion of oil exports can easily be offset when oil exports account for only a small fraction of that country s total trade flows. Similarly, the power of oil money to buy political loyalty depends in part on how easily people can raise money from other sources. Specifically, we examined three different dependency metrics, namely the share of hydrocarbon exports in total exports, the value added by the hydrocarbon sector relative to the total gross domestic product (GDP), and the share of government hydrocarbon revenues in total government revenues. We included every state in our sample that passed a metric-specific threshold in at least one year between 1995 and 2007 with respect to at least one of these three dependency metrics. We determined the threshold for each metric by examining the frequency distribution of dependency values among oil and gas reserve states. The Epanechnikov kernel density function for each of the three dependency metrics turned out to be clearly bimodal, thus separating the universe of oil and gas reserve states into two distinct categories: one group of countries with distinctly low and one with high dependency values. We took the local mini- Page 5 of 42

6 mum as our threshold value for each of the three dependency metrics (40 percent for export dependency 3, 38 percent for GDP dependency, and 48 percent for government revenue dependency). Based on the export dependency metric we identified a 30 countries as hydrocarbon dependent. The only additional country identified as hydrocarbon dependent by the two other dependency metrics is Timor-Leste for which no export data was available, bringing the total of hydrocarbon dependent states to 31 (see table below). This codebook focuses on a subset of 25 hydrocarbon dependent states. Six countries namely Brunei, Chad, Equatorial Guinea, Iraq, Timor-Leste, and Turkmenistan are dropped because of poor data coverage. Annual data related to the full set of 31 hydrocarbon dependent countries is reported in the Stanford Oil Wealth Management Excel dataset. 3 The selection based on export dependency is robust. A threshold value of only 30 percent identifies the same set of countries as dependent on hydrocarbon exports. Page 6 of 42

7 Table 1: Hydrocarbon dependent states Country name Hydrocarbon exports (in % of total exports) Average dependency ( ) on Value added by hydrocarbon sector (in % of total GDP) Hydrocarbon revenues (in % of total government revenues) Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Countries not included because of poor data quality 5 Brunei Chad Equatorial Guinea Iraq Timor-Leste n/a Turkmenistan Bolivia is included in the dataset because it surpassed the export dependency threshold in the years 2006 and 2007 even though its average dependency throughout remained below the threshold. 5 In contrast to the 25 countries listed above, these countries lack data for a significant number of the years, thereby making the comparison of averages problematic. 6 Chad qualifies as hydrocarbon dependent because it surpassed the export dependency threshold in , the revenue dependency threshold in 2006, and the GDP threshold in , though its average dependency between was below the threshold due to a late onset of its oil production. Page 7 of 42

8 3 Variable profiles and general conversions This section provides a complete codebook of the indicators underlying our assessment of oil states development performance. The indicators are listed thematically according to their relative position in the three-stage value chain of revenue generation, management, and disbursement. Whenever available, data was collected for the years 1995 to For each indicator we provide the following information: The underlying rational for including indicator in our assessment. The variable description. The variable code. Years with comprehensive coverage. The units in which the variable is measured. The data sources. The methodology used to produce the variable, including any additional processing of the data beyond that of the data providers. The observed mean, standard deviation, minimum (min) and maximum (max) values for all 25 oil or gas dependent countries with comprehensive data coverage. Monetary values are generally reported in constant 2000 US$ terms. Data that was only available in nominal local currency units was first converted into US$ using the official exchange rate provided by the World Bank s World Development Indicators (WDI) 7 and then deflated using the deflator provided by the U.S. Department of Commerce 8. When reporting hydrocarbon volumes (on reserves or production) we combined oil volumes and gas volumes whereby gas volumes were first converted into barrels of oil equivalents (boe), using the conversion factors provided in the BP Statistical Review of World Energy For a few cases, countries had multiple currencies at the same time, and different sources refer to different currencies. These other exchange rates were taken from Page 8 of 42

9 4 Individual variable data sheets Page 9 of 42

10 4.1 Stage 1: Revenue generation Government hydrocarbon revenues per hydrocarbon output STATA name revhc_prodhc Stage 1 Category Outcome Number 1.1 Years Revenues (incl. taxes, signature bonuses, transit fees, etc) government collects from the hydrocarbon sector relative to hydrocarbon output of same year. As the steward of a nation s hydrocarbon wealth governments should strive to maximize the revenues they collect from the hydrocarbon industry over the full production cycle. US$, constant 2000/ boe Hydrocarbon revenue data (in order of precedence): IMF Regional Economic Outlook 2008 for Middle East and Central Asia ( IMF reports on individual countries (Article IV Consultations, Statistical Appendices, Selected Issues), dataset underlying IMF The Role of Fiscal Institutions in Managing the Oil Revenue Boom ( Oil and gas production data (in order of precedence): BP Statistical Review of World Energy 2008; EIA World Production of Crude Oil, NGPL, Other Liquids, and Refinery Processing Gain, EIA World Dry Natural Gas Production, Most Recent Annual Estimates, Methodology Divided hydrocarbon revenues by hydrocarbon output and calculated for each country the average over revhc_prodhc country_name revhc_~c 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 10 of 42

11 4.1.2 Competition in oil industry STATA name comp_oilindustry Stage 1 Category Policy Number Subjective evaluation of licensing terms and liberalization trends which determine the extent to which oil companies can compete freely and under predictable rules Assuming governments still get a large share of oil revenues, a competitive environment should lead to more efficient companies and higher government revenues per production. Score, max. score of 100 indicates regulatory environment in which oil companies can compete freely Business Monitory International Oil & Gas Reports for individual countries 9 Years 2008 Methodology Country score of 2008 comp_oilin~y country_name year comp_o~y Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep comp_oilindustry corresponds to BMI s Industry Risk variable Page 11 of 42

12 4.1.3 Operational efficiency of national oil company (NOC) STATA name noceff Stage 1 Category Policy Number Years 2004 Methodology Operational efficiency of National Oil Companies measured as hydrocarbon output volumes per labor input. Many National Oil Companies are overstaffed, thereby resulting in foregone profits for the NOC and reducing the revenues the government can collect from the NOC. Thousands of barrels of gas and oil and gas produced daily by NOC per number of NOC employees Energy Intelligence Top 100 Rankings Divided daily hydrocarbon output of NOC by number of NOC employees. if a country had more than one NOC, we combined the numbers for output and employees of each NOC in that country. noceff country_name year noceff Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 12 of 42

13 4.1.4 Domestic gasoline price relative to US gasoline price STATA name subsid_fuel_nd_l_perc Stage 1 Category Policy Number Measures the difference between domestic price for gasoline and gasoline price in the United States in % of US prices. Negative values indicate that gasoline is cheaper in country X than in the United States. Many oil states dictate the domestic price for hydrocarbon products to be well below international prices. This results in foregone profits for oil companies and foregone revenues for the government. % (negative number indicates the gasoline price in foreign country is less than in the US) Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) International Fuel Prices Years Methodology Calculated difference between price of domestic gasoline in country X and price of gasoline in the US in the same year as a percentage of the US price. Calculated for each country the simple average over subsid_fue~c country_name subsid~c 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 13 of 42

14 4.1.5 Reserve to production ratios STATA name reshc_prodhc Stage 1 Category Policy Number Number of years that country can sustain current hydrocarbon production levels given the size of its proved hydrocarbon reserves. The sustainability of government hydrocarbon revenues largely depends on maintaining production volumes for the foreseeable future. Ideally, the R-to-P ratio remains close to constant thanks to ongoing investments in successful exploration that expands the reserve base at the same rate as existing reserves are being depleted. Years Years Methodology Oil and gas reserve and production data (in order of precedence): BP Statistical Review of World Energy 2008; EIA World Production of Crude Oil, NGPL, Other Liquids, and Refinery Processing Gain, EIA Oil Proved Reserves, All Countries EIA World Dry Natural Gas Production, Most Countries, Most Recent Annual Estimates, EIA Natural Gas Proved Reserves, All Countries Divided volume of proved hydrocarbon reserves (mboe) by volume of annual hydrocarbon production (mboe/y). Calculated for each country the simple average over reshc_prodhc country_name reshc_~c 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 14 of 42

15 4.1.6 Resource to production ratios STATA name resourcesoil_prodoil_2006_bgr Stage 1 Category Policy Number Number of years that country can sustain current oil production levels given the size of its undiscovered and economically and/or technically unrecoverable oil resources Measures the sustainability of current oil production, similar to the reserve-to-production variable above. Main difference is that this indicator is only about oil and is theoretically independent of the success of exploration. Years Years 2006 Methodology Bundesanstalt fuer Geowissenschaften und Rohstoffe Kurzstudie Reserven, Ressourcen und Verfügbarkeit von Energierohstoffen. Kurzf 2006,tem plateid=raw,property=publicationfile.pdf/energiestudie_kurzf_2006.pdf Divided volume of oil resources (mt) by volume of annual oil production (mt/y). resourceso~r country_name year re~6_bgr Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 15 of 42

16 4.1.7 Stringency of environmental regulation STATA name envirreg_stringent Stage 1 Category Policy Number Subjective evaluation of stringency of environmental regulations based on Executive Opinion Survey 2006, Stringent environmental regulations increase production costs and thereby lower potential government hydrocarbon revenues. However, they also make oil and gas production more sustainable by reducing the damage caused by the industry. Score; min. score of 1 indicates lax environmental regulations compared to most countries, max score of 7 indicates environmental regulations that are among the world s most stringent. Years 2007 Methodology Value of 2007 World Economic Forum The Global Competitiveness Report New York: Palgrave. envirreg_s~t country_name year envirr~t Angola United Arab Emirates Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Algeria Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago Venezuela, RB Yemen, Rep Page 16 of 42

17 4.1.8 Gas flared per oil output STATA name gas_flared_prodoil Stage 1 Category Policy Number Measures the share of associated gas that is being flared. In addition to presenting a serious environmental hazard, gas flaring is a wasteful practice that results in foregone profits and thus non-realized government revenues. An optimal industrial organization provides the oil industry sufficient incentives to develop and market associated gas for either the domestic market or for export. mboe/mb Years Methodology Data on gas flaring: World Bank A Twelve Year Record of National and Global Gas Flaring Volumes Estimated Using Satellite Data. except for Bahrain: Carbon Dioxide Information Analysis Center (CDIAC). Data on oil production (in order of precedence): BP Statistical Review of World Energy 2008; EIA World Production of Crude Oil, NGPL, Other Liquids, and Refinery Processing Gain, Divided volume of gas flared (mboe) by volume of annual oil production (mb) for each year. Calculated for each country the simple average over gas_flared~l country_name gas_fl~l 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 17 of 42

18 4.2 Stage 2: Revenue management Real growth in non-oil GDP STATA name gdpnhc_cagr_9806 Stage 2 Category Outcome Number CAGR (%) Years Methodology Compound annual growth rate of real non-hydrocarbon GDP For oil states long-term economic success it is indispensable that they succeed in growing their nonhydrocarbon sector alongside the hydrocarbon sector as the former is non-depleteable, creates more employment opportunities, and can buffer the huge swings typical for the hydrocarbon sector. In order of precedence: IMF Regional Economic Outlook 2008 Middle East and Central Asia ( IMF Regional Economic Outlook 2008 Sub-Saharan Africa ( IMF reports on individual countries (Article IV Consultations, Statistical Appendices, Selected Issues), Banco Central de Venezuela ( Using our data on non-hydrocarbon GDP in constant 2000 USD, we calculated the compound annual growth rate over gdpnhc_~ country_name gdp~ Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 18 of 42

19 4.2.2 Real growth in non-oil exports STATA name expnhc_cagr_9706 Stage 2 Category Outcome Number CAGR (%) Years Methodology Compound annual growth rate of real value of non-hydrocarbon exports Growth in non-hydrocarbon exports sheds light on the sustainability of growth in the non-hydrocarbon sector. A state can be considered particularly successful in revenue management when it succeeds in translating strong growth in the domestic non-oil sector into a comparably strong expansion of non-oil exports as this indicates the international competitiveness of its non-oil sector and diversifies the sources of foreign currency. In order of precedence: IMF reports on individual countries (Article IV Consultations, Statistical Appendices, Selected Issues), United Nations Commodity Trade Statistics Database (UNCOMTRADE): Standard International Trade Classification (SITC) Rev. 1, commodity code 33 and 34. Using our data on non-hydrocarbon exports in constant 2000 USD, we calculated the compound annual growth rate over expnhc_~ country_name exp~ Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 19 of 42

20 4.2.3 Consumer price inflation STATA name Inflation Stage 2 Category Policy Number Year average in % Years Annual change in the average price of consumer goods and services purchased by representative household. Income generated by the hydrocarbon sector often expands more rapidly than other economic sectors thereby creating supply bottlenecks and an upward pressure on the price of factor inputs and non-oil sector goods and services. In order of precedence: World Development Indicators, IMF Regional Economic Outlook 2008 Middle East and Central Asia (:// IMF Regional Economic Outlook 2008 Sub-Saharan Africa ( IMF reports on individual countries (Article IV Consultations, Statistical Appendices, Selected Issues). Methodology Calculated for each country the simple average over inflation country_name inflat~n 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 20 of 42

21 4.2.4 Volatility of official exchange rate STATA name exchange_lcu_usd Stage 2 Category Policy Number Years Volatility of the official exchange rate A common explanation for the oil curse is the so-called Dutch disease which stipulates that the nonhydrocarbon sector can suffer from a rapid expansion of the hydrocarbon sector through the upward pressure the latter exerts on the local currency, thereby hampering the former s international competitiveness. In order to protect the non-hydrocarbon sector from this risk, governments of oil states need to mitigate the currency impact of the oil sector through sterilization, stabilization, and/or savings measures. Low volatility of the exchange rate is an indicator of the government s success in creating a stable macroeconomic environment. Local currency unit per US$ World Development Indicators (WDI) Methodology Calculated for each country the standard deviation of its exchange rate between 1995 and zexchange_~d country_name zexcha~d 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 21 of 42

22 4.2.5 Central government non-oil fiscal balance STATA name fiscbalnhc_gdpnhc Stage 2 Category Policy Number Central government overall fiscal balance excluding oil revenues. Given the high volatility, unpredictability and ultimate depleteability of hydrocarbon revenues a government is well advised to limit the extent to which it relies on this source of revenues. The non-oil fiscal balance highlights a country s fiscal vulnerability as high levels of expenditures may not be sustained when oil prices or production levels drop. In % of non-oil GDP Years Methodology In order of precedence: IMF Regional Economic Outlook 2008 Middle East and Central Asia (:// IMF Regional Economic Outlook 2008 Sub- Saharan Africa ( IMF reports on individual countries (Article IV Consultations, Statistical Appendices, Selected Issues). Data post-2003 is ready-made; values for earlier years were obtained by subtracting oil revenues and expenditures from total revenues and then dividing by non-oil GDP. Calculated for each country the average over fiscbalnhc~c country_name fiscba~c 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 22 of 42

23 4.2.6 Central government fiscal balance STATA name fiscbal_gdp Stage 2 Category Policy Number Difference between total central government revenues and expenditures. Fiscal balance measures the sustainability of government revenues and expenditures. In % of GDP Years Methodology In order of precedence: IMF Regional Economic Outlook 2008 Middle East and Central Asia ( IMF Regional Economic Outlook 2008 Sub-Saharan Africa ( IMF reports on individual countries (Article IV Consultations, Statistical Appendices, Selected Issues). Data post-2003 is ready-made; values for earlier years were obtained by subtracting expenditures from total revenues and then dividing by GDP. Calculated for each country the average over fiscbal_gdp country_name fiscba~p 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 23 of 42

24 4.2.7 Applied tariffs STATA name tariff_applied Stage 2 Category Policy Number % Years Methodology The weighted average of a country s applied tariff rates for all goods (weighted by trade import values). Provides an indication of the extent to which non-oil sector growth can be considered to be internationally competitive and thus sustainable. High non-oil sector growth rates need to be discounted if they occurred in a country that shields its domestic economy from international competition through high tariffs. World Bank Trade Indicators: WB gives averages for , , and Used the averages as the values for each year (for 2006, used the averages). Calculated for each country the average over tariff_app~d country_name tariff~d 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 24 of 42

25 4.2.8 Global competitiveness STATA name gci_score Stage 2 Category Policy Number The Global Competitiveness Index assesses annually the global competitiveness of economies around the world based on 12 pillars of competitiveness. The pillars include: Institutions, Infrastructure, Macroeconomic Stability, Health and Primary Education, Higher Education and Training, Goods Market Efficiency, Labor Market Efficiency, Financial Market Sophistication, Technological Readiness, Market Size, Business Sophistication and Innovation. The resource curse literature suggests that governments of oil dependent states do not have an incentive (or even a disincentive) to promote non-oil sector growth through deregulation, liberalization, and privatization. The Global Competitiveness Index is one of three publicly available rankings included in the SOWM dataset to cast light on the overall business climate. Score; maximum score of 7 indicates most competitive economy World Economic Forum; Years 2006 & 2007 Methodology Calculated simple average for each country over the years 2006 and gci_score country_name gci_sc~e 1. Algeria Angola. 3. Azerbaijan Bahrain Bolivia Cameroon Congo, Rep.. 8. Ecuador Gabon. 10. Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan. 21. Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep.. Page 25 of 42

26 4.2.9 Ease of doing business STATA name business_ease Stage 2 Category Policy Number Years The World Bank compiles the Ease of Doing Business index from 10 sub-indices: starting a business, dealing with construction permits, employing workers, registering property, getting credit, protecting investors, paying taxes, trading across borders, enforcing contracts, and closing a business. The rankings for each sub-index are averaged together (each one given equal weight) and the Ease of Doing Business ranking is based on this score. The theoretical resource curse literature suggests that governments of oil dependent states do not have an incentive (or even a disincentive) to promote non-oil sector growth through deregulation, liberalization, and privatization. The Ease of Doing Business index is one of three publicly available rankings included in the SOWM dataset to cast light on the overall business climate. Ranking; 1 = most business-friendly regulations World Bank: Methodology Calculated simple average for each country over the years business_e~e country_name busine~e 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya. 14. Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 26 of 42

27 Economic freedom STATA name econfree_score Stage 2 Category Policy Number Years The Heritage Foundation measures the freedom of an economy based on the freedom in ten different spheres: business freedom, trade freedom, fiscal freedom, government size, monetary freedom, investment freedom, financial freedom, property rights, freedom from corruption, and labor freedom. The theoretical resource curse literature suggests that governments of oil dependent states do not have an incentive (or even a disincentive) to promote non-oil sector growth through deregulation, liberalization, and privatization. The Economic Freedom index is one of three publicly available rankings included in the SOWM dataset to cast light on the overall business climate. Score; maximum score of 100 indicates the freest economy Heritage Foundation: Methodology Calculated simple average for each country over econfree_s~e country_name econfr~e 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 27 of 42

28 4.3 Stage 3: Revenue disbursement Human development index STATA name hdi_value Stage 3 Category Outcome Number Years Composite development indicator that considers life expectancy, educational attainment, and income. The resource curse literature suggests that governments of oil dependent states tend to have a relatively weak incentive to promote human capital development because they do not rely on the productivity of their population for revenues. The Human Development Index is one is out of three human capital indicators included in the Stanford Oil Wealth Management dataset. Score, maximum score of 100 indicates highest level of human development. United Nations Development Program (UNDP): Human Development Report: Methodology Calculated simple average for each country over the years hdi_value country_name hdi_va~e 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 28 of 42

29 4.3.2 Education index STATA name edu_index Stage 3 Category Outcome Number Years The Education Index is comprised of adult literacy rates and the combined gross enrollment ratio for primary, secondary and tertiary schooling, weighted to give literacy more significance in the statistic. The resource curse literature suggests that governments of oil dependent states tend to have a relatively weak incentive to promote human capital development because they do not rely on the productivity of their population for revenues. The Education Index is one is out of three human capital indicators included in the Stanford Oil Wealth Management dataset. Composite score; maximum score of 100 indicates highest level of education United Nations Development Program (UNDP): Human Development Report: Methodology Calculated simple average for each country over the years edu_index country_name edu_in~x 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 29 of 42

30 4.3.3 Infant mortality STATA name mortinf Stage 3 Category Outcome Number Years Number of children dying under one year of age The resource curse literature suggests that governments of oil dependent states tend to have a relatively weak incentive to promote human capital development because they do not rely on the productivity of their population for revenues. Infant mortality is one is out of three human capital indicators included in the Stanford Oil Wealth Management dataset. Number of infant deaths per 1,000 live births World Development Indicators (WDI) Methodology Calculated simple average for each country over the years mortinf country_name mortinf Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 30 of 42

31 4.3.4 Mainland phone line density STATA name phone_density Stage 3 Category Outcome Number Years Methodology Number of mainland phone lines per 10,000 people. High quality infrastructure is indispensable for sustaining non-oil sector growth. The density of phone lines is one out of six infrastructure indicators included in the Stanford Oil Wealth Management dataset, whereby higher values indicate better infrastructure quality. Phone lines per 10,000 people World Development Indicators (WDI) Divided total phone lines by (population/10,000). Calculated simple average for each country over the years phone_dens~y country_name phone_~y 1. Algeria Angola Azerbaijan Bahrain Bolivia Cameroon Congo, Rep Ecuador Gabon Iran, Islamic Rep Kazakhstan Kuwait Libya Nigeria Norway Oman Qatar Russian Federation Saudi Arabia Sudan Syrian Arab Republic Trinidad and Tobago United Arab Emirates Venezuela, RB Yemen, Rep Page 31 of 42