Transparenting Initial Empirics and Policy Applications Daniel Kaufmann and Ana Bellver World Bank Institute http://worldbank.org/wbi/governance Presentation at the Pre-Conference on Institutional Change for Growth and Poverty Reduction in Low-Income Countries at the International Monetary Fund, Washington, DC, July 6-7, 2005 1
Main Messages of Transparenting Major contributions exist on right to know/access to information and transparency, but little progress on: -- measurement and unbundling of transparency -- empirical analysis -- policy analysis and strategic/policy focus Propose a Index (as in UCM/Governance work): Economic/Institutional vs. Political Finding of large variance worldwide in terms of transparency: i) cross-country: incl. within same region & transparency present in some emerging economies; and ii) within country: economic vs. political transparency associated with better socio-economic indicators, less corruption, & improved competitiveness Results are supported at the micro level: in-country diagnostic complement: Bolivia and Chile illustrations Policy implications: transparency core in 2 nd- generation institutional reforms, role of IFIs, & concrete strategies 2
Towards an Aggregate Index of : Basic Unbundling into Two Dimensions Ec/Institutional transparency: content, accessibility and usefulness of economic and institutional information provided by public institutions Economic and Financial Information Implementation of Freedom of Information laws in the budget and its process of economic policy-making of the public administration Political transparency: response to citizenry s demand for open debate and for accountability in political institutions: in political funding Openness and competitiveness of political participation Independence of the media Disclosure of Politician s Assets, Voting records, etc 3
Sources of Indicators Cross-Country Surveys of Firms: Global Competitiveness Survey, World Competitiveness Yearbook Expert Assessments from NGOs, Think Tanks: Reporters Without Borders, Freedom House, Amnesty International, Fundar & other Latin American NGOs, Brown University s Center for Public Policy, International, Polity IV Project, International Budget Project, Center for Public Integrity, Freedominfo Expert Assessments from Multilateral Agencies: Islam Economic Index, United Nations, World Bank CPIA, African Development Bank CPIA, UNECA 4
Sources for Indicators, 2004 African Development Bank Country Policy & Institutional Assessment poll Brown University s Centre for Public Policy Center for Public Integrity Freedom House Freedom House Freedominfo/Banisar Fundar/International Budget Project International Budget Project Global E-Government 2004 Public Integrity Index Freedom in the World Nations in Transit Global Survey 2004 Latin American Index of Budget Opening Budgets to Understanding and Debate, 2004 objective assessment Poll poll poll objective assessment poll objective assessment Institute for Management Development OECD/WB Polity IV Project World Competitiveness Yearbook Budget Practices and Procedures survey Political Regime Characteristics and Transitions, 1800-2002 survey objective assessment poll 5
Sources for Indicators, 2004 Cont d Reporters Without Borders Reporters Without Borders poll State Department/ Amnesty International International UN Economic Commission for Africa CIRI Human Rights dataset Global Corruption Report 2004 African Governance Indicators, 2005 poll objective assessment poll United Nations World Bank World Bank World Economic Forum World Public Sector Report 2003 Islam Index 2004 Country Policy & Institutional Assessment Global Competitiveness Report objective assessment objective assessment poll survey 6
Perceptions -based data and Objective-data 1.Examples of Perceptions -based data Firms are usually informed clearly and transparently by the Government on changes in policies affecting their industry? (IT) Is freedom of the media guaranteed? (PT) In practice, is the right of access to information effective? (IT) 2.Examples of Objective data Is the deposit interest rate published quarterly? (IT) Is the budget documentation placed on the internet? (IT) Is there a Freedom of information Law? (IT) Is disclosure of contributions to political parties mandatory? (PT) 7
Why Aggregate Indicators? Basic Premise: individual data sources provide a noisy signal of broader concept of governance, e.g.: Publication of economic variables EC/INST TRANSPARENCY Disclosure of political funding POLITICAL TRANSPARENCY of policy EC/INST TRANSPARENCY Freedom of the press POLITICAL TRANSPARENCY Benefits of Aggregation -- through the U.C. Method aggregate indicators are more informative about broad concepts of transparency broader country coverage than individual indicator generate explicit margins of error for country scores 8
Building Aggregate Indicators Use Unobserved Components Model (UCM) to construct composite transparency indicators, and margins of error for each country Estimate of transparency: weighted average of observed scores for each country, re-scaled to common units Weights are proportional to precision of underlying data sources Precision depends on how strongly individual sources are correlated with each other Margins of error reflect (a) number of sources in which a country appears, and (b) the precision of those sources 9
Unobserved Components Model Observed indicator k of governance in country j, y(j,k), is noisy indicator of true governance in country j, g(j): ( ) yjk (, ) = α(k) + β(k) gj () + ε(, jk) Variance in measurement errors is same across countries for each source, but different across sources: E [ ] 2 2 ε( j,k) = (k) σ ε Identifying assumption: Measurement errors are uncorrelated across sources highly correlated sources measure governance with more precision 10
Estimates of Governance from UCM UCM allows us to infer the distribution of governance in a country conditional on the observed data for that country K( j) E[g(j) y(j,1),...,y(j,k(j))] Best estimate of governance is the mean of this conditional distribution: 2 σ ε ( k ) w ( k ) = K ( j ) 2 1 + σ ε ( k ) k = 1 So estimate of governance is weighted average of rescaled scores, with weights proportional to precision of each source: = w(k) y( j,k) α(k) k = 1 β(k) 11
Precision of Estimates from UCM Reliability or precision of estimate of transparency for each country is the standard deviation of this conditional distribution: SD[g( j) y(j,1),..., y(j,k(j))] K( j) = 1+ σ k= 1 ε (k) 2 1 2 These standard errors are smaller for countries that (a) appear in more sources, and/or (b) appear in more reliable sources 12
Estimating Unobserved Components Model Distinguish between representative and nonrepresentative sources For representative sources, estimate parameters α(k), β(k), and σ ε (k) using maximum likelihood Construct initial estimate of governance using representative sources only For non-representative sources, estimate parameters by regressing each source on initial estimate of governance Construct final estimate of transparency using all sources 13
Levels of Worldwide, 2004 Estimates of transparency for 194 countries Assess the precision of these estimates, using standard errors generated by UCM Simple rule of thumb: cross-country differences in governance significant if 90% confidence regions don t overlap Precision of governance indicators has improved over time with more, and better, data sources e.g. MCA eligibility requires countries above median on Control of Corruption 2004 KK data: 33% of countries significantly above (at 90% confidence level)--much improved Single source: less than 10% of countries are significantly above median (at 90% confidence) 14
Institutional, Bottom Quartile (Initial, 2004)* 2.5 High Level Margin of Error 0 Low -2.5 LIBERIA KOREA, NORTH AFGHANISTAN SOMALIA TURKMENISTAN LAOS LIBYA MICRONESIA SYRIA ERITREA KIRIBATI CONGO DJIBOUTI IRAQ IVORY COAST NIGER UZBEKISTAN AZERBAIJAN SIERRA LEONE CAMEROON MALAWI GABON ANGOLA Source: Transparenting by BK (2005). N. of countries: 194. * Preliminary data- Not for quotation or to establish rankings. 15
Institutional, Middle-Low Quartile (Initial, 2004)* 2.5 High Level Margin of Error 0 Low -2.5 GAMBIA ZIMBABWE MYANMAR BURUNDI ETHIOPIA MAURITANIA BENIN MALI KENYA NIGERIA ZAMBIA MOZAMBIQUE MADAGASCAR EGYPT TANZANIA HONDURAS VIETNAM IRAN BURKINA FASO NAMIBIA GEORGIA LEBANON UGANDA FIJI Source: Transparenting by BK (2005). N. of countries: 194. * Preliminary data- Not for quotation or to establish rankings. 16
Institutional, Middle-High Quartile (Initial, 2004)* 2.5 High Level Margin of Error 0 Low -2.5 GHANA GUATEMALA TUNISIA MOROCCO BOTSWANA BELARUS MACEDONIA ALBANIA PARAGUAY NEPAL RUSSIA URUGUAY ARMENIA CHINA INDONESIA COSTA RICA SOUTH AFRICA JORDAN ROMANIA BOLIVIA MALAYSIA CROATIA COLOMBIA PANAMA SLOVAK REPUBLIC Source: Transparenting by BK (2005). N. of countries: 194. * Preliminary data- Not for quotation or to establish rankings. 17
Institutional, Top Quartile (Initial, 2004) * 3.3 0.8-1.7 BULGARIA INDIA THAILAND PERU UKRAINE ARGENTINA HUNGARY ICELAND AUSTRIA BRAZIL SPAIN POLAND SLOVENIA ITALY NORWAY GERMANY JAPAN MEXICO FINLAND NETHERLANDS FRANCE DENMARK SWEDEN AUSTRALIA 18 UNITED KINGDOM CHILE CANADA UNITED STATES High Level Margin of Error Low Source: Transparenting by BK (2005). N. of countries: 194. * Preliminary data- Not for quotation or to establish rankings.
Institutional, Selected Countries (Initial, 2004)* 3.5 High Level Margin of Error 0.5 Low -2.5 LIBERIA KOREA, NORTH SYRIA MALAWI MYANMAR BURUNDI ZAMBIA SENEGAL VENEZUELA BANGLADESH GEORGIA Source for data: Transparenting by BK (2005).Colors are assigned according to the following criteria: Dark Red, bottom10 th percentile rank; Light Red between 10 th and 25 th Orange, between 25 th and 50 th ; Yellow, between 50 th and 75 th ; Light Green between 75 th and 90 th ; Dark Green above 90 th. FIJI GUATEMALA BOTSWANA PARAGUAY RUSSIA CHINA INDIA UKRAINE BRAZIL SLOVENIA JAPAN FRANCE SWEDEN 19 CHILE UNITED STATES
Political, Bottom Quartile (Initial, 2004)* 2.5 0-2.5 KOREA, NORTH CUBA LIBYA TURKMENISTAN ZIMBABWE CHINA SUDAN SAUDI ARABIA BELARUS SYRIA SOMALIA RWANDA EGYPT BHUTAN 20 KYRGYZ REPUBLIC LIBERIA YEMEN AZERBAIJAN CHAD CAMEROON UKRAINE MALAYSIA RUSSIA High Level Margin of Error Low Source: Transparenting by BK (2005). N. of countries: 194. * Preliminary data- Not for quotation or to establish rankings.
Political, Middle-Low Quartile (Initial, 2004)* 2.5 High Level Margin of Error 0 Low -2.5 ANGOLA 21 BANGLADESH MALDIVES ETHIOPIA ALGERIA JORDAN PAKISTAN VENEZUELA CAMBODIA ZAMBIA COLOMBIA NIGERIA CONGO GUATEMALA KENYA TURKEY INDONESIA UGANDA HONDURAS PARAGUAY MOZAMBIQUE MADAGASCAR Source: Transparenting by BK (2005). N. of countries: 194. * Preliminary data- Not for quotation or to establish rankings.
Political, Middle-High Quartile (Initial, 2004)* 2.5 High Level Margin of Error 0 Low -2.5 CROATIA INDIA MEXICO BULGARIA BOLIVIA NAMIBIA ARGENTINA BRAZIL BOTSWANA KOREA, SOUTH ITALY ISRAEL HONG KONG GREECE GHANA CHILE SOUTH AFRICA HUNGARY URUGUAY 22 SLOVAK REPUBLIC AUSTRIA CYPRUS BELIZE Source: Transparenting by BK (2005). N. of countries: 194. * Preliminary data- Not for quotation or to establish rankings.
Political, Top Quartile (Initial, 2004)* 2.5 0-2.5 SWEDEN POLAND COSTA RICA SPAIN FRANCE JAPAN LATVIA UNITED KINGDOM ESTONIA GERMANY CANADA PORTUGAL UNITED STATES 23 MARSHALL ISLANDS LUXEMBOURG NEW ZEALAND SWITZERLAND FINLAND NORWAY ICELAND DENMARK High Level Margin of Error Low Source: Transparenting by BK (2005). N. of countries: 194. * Preliminary data- Not for quotation or to establish rankings.
Political, Selected Countries (Initial, 2004)* 2.5 High Level Margin of Error 0 Low -2.5 KOREA, NORTH CUBA VIETNAM CHINA LIBERIA AZERBAIJAN UKRAINE RUSSIA BANGLADESH JORDAN MALAWI Source for data: Transparenting by BK (2005).Colors are assigned according to the following criteria: Dark Red, bottom10 th percentile rank; Light Red between 10 24 th and 25 th Orange, between 25 th and 50 th ; Yellow, between 50 th and 75 th ; Light Green between 75 th and 90 th ; Dark Green above 90 th. * Preliminary data- Not for quotation or to establish rankings. HONDURAS GEORGIA PANAMA ROMANIA NAMIBIA ARGENTINA PHILIPPINES BOTSWANA ITALY GHANA CHILE SPAIN PORTUGAL BELGIUM DENMARK
Overall, Selected Countries (Initial, 2004)* 2.5 High Level Margin of Error 0 Low -2.5 KOREA, NORTH CUBA SOMALIA SUDAN YEMEN IVORY COAST GAMBIA CHINA ZAMBIA VENEZUELA NIGERIA GEORGIA JORDAN Source for data: Transparenting by BK (2005).Colors are assigned according to the following criteria: Dark Red, bottom10 th percentile rank; Light Red between 10 th and 25 th Orange, between 25 th and 50 th ; Yellow, between 50 th and 75 th ; Light Green between 75 th and 90 th ; Dark Green above 90 th. * Preliminary data- Not for quotation or to establish rankings. MADAGASCAR SENEGAL PARAGUAY UKRAINE TURKEY PANAMA BOLIVIA EL SALVADOR GREECE BRAZIL SLOVAK REPUBLIC MEXICO ITALY CHILE DENMARK 25 CANADA UNITED STATES
High Control of Corruption vs. Institutional (Initial, 2004) 3 Control of Corruption Low 2 1 0-1 -2 r = 0.71 FIN ISL SGP NZL DNK LUX CHE SWE AUT NOR NLD AUS GBR CAN DEU LIE HKG IRL BEL ES P FRA BHS ARE MLT PRT JPN SVN ATG BRB BWA CYP OMN KWT BHR CRI ISR ES T BTN HUN TWNITA QAT GRD URY GRC ZAF SUR KNA VCT TUN MUS SVK LTU CPV JOR LCA MYS CZE BRN DMA LVA NAM SAU MDV POL KOR SMR SYC SAM HRV NRUKIR PCIMRT TTO LSO MAR MCO BLZ PAN BGR MDG FJI EGY GHA LKA COL BRA TUR FSM TMP ROM THA MEX BENRWA BFA NIC PER SEN LBN GUY IND SLV YUG VUT MLI DZA ARG MNGMKD JAM GAB WBGTZA BIH ARMCHN DOM PHL ERI IRN STP GMBCUB NPL TON ZMB VNM UGA TUV SYRGNB HND RUS MOZ GTM LBR GIN CMR ECU ALB NERSLE YEM MWI BOL ETH PNG MHL LBY MDA PAK DJI TGO KEN VEN UKR SWZ KGZ GEO IDN COG CIV ZWE KHM BLR AZE PRY LAO COM TJK AGO BDI NGABGD KAZ SLB UZB AFGTKM SDN ZAR CAF PRK HTI IRQ MMR SOM GNQ -1.5 0.5 2.5 Low Institutional High 26 Sources: Governance Matters IV by KKM (2005) and Transparenting transparency by BK (2005). N.countries: 190 CHL
High Control of Corruption vs. Political (Initial, 2004) 3 Control of Corruption Low 2 1 0-1 -2 r = 0.76 SGP ARE LUXCHE AUT SWE GBR NLDNOR CAN AUS DEU USA LIE IRL BEL ESP FRA BHS JPN MLT PRT SVN ATG BWA OMN BHR ISR CYP CRI BRB EST BTN KWT ITA HUN TWN QAT GRC ZAF URY GRD JOR SVK SUR TUN MYS CPV MUS CZE KNA LTU VCT LCA BRN DMA SAU NAM LVA MDV KOR POL HRV MRT SYC SAM MAR TTO SMR PAN BGR LSO KIRNRU MCOPCI BLZ LKA MDG FJI EGY COL BRA GHA TUR ROM THA MEX TMP RWA YUG SEN PER NIC FSM LBN BFA IND SLV BEN GUY CHN DZA MNG VUT WBG NPL TZA MKD DOM ARG PHL MLI JAM IRN GABARM BIH ERI GMB VNM ZMB UGA TON STP SYR RUS GNB GTM HND ALB CMR YEM MOZ ECU TUV LBR GIN SWZ TGO UKR ETHPAK MDASLENER MWI BOL MHL LBY BLR KGZ DJI VEN KEN IDN GEO PNG ZWE CIV KHM AZE COG PRY LAO KAZ BDI TJK AGO BGD NGA COM UZB SLB TKM SDN ZAR AFG CAF HTI IRQ SOM GNQ 27 Sources: Governance Matters IV by KKM (2005) and Transparenting transparency by BK (2005). N.countries: 190 HKG CHL FIN NZL ISL DNK -2.0 0.0 2.0 Low Political High
Global Competitiveness vs. Institutional (Initial, 2004) High 6 FIN Global Competitiveness Index 4 r = 0.79 ARE BHR MYS LUX GRC THALTU HUN TUNCYP ZAF JOR LVA SVK BWA CHN MAR MUS TTO URY CRI SLV NAM PAN IND EGY ROM HRVBGR GHA JAM COLPER TUR DZA RUS IDN DOM GMB KEN VNM LKA ARG ZMB TZA UGABIH VEN GTM MKD UKR MWI MLI NGA ECU YUG PAK MDG MOZ HNDGEO BOL ZWE PRY NIC BGD AGO ETH TCD ISL AUT HKG BEL ES P MLT CZE TWN SVN DNK SWE NOR SGP CHEJPN NLD NZL ISR ES T PRT KOR IRL FRA ITA MEX BRAPOL PHL GBR CAN CHL Low 2-1.5 0.5 2.5 Low Institutional High Sources: EOS 2004 and Transparenting transparency by BK (2005). N. of countries: 104. 28
Regression Results: Institutional Dep Variable: Control of Corruption Dep Variable: Global Competitiveness Index 1 2 3 4 5 UCM Institutional 0.61 0.25 0.37 0.34 0.66 8.59*** 3.93*** 1.14 4.75*** 2.02** Political Rights 0.14 0.11 0.16-0.01 0.04 5.14*** 5.43*** 3.79*** 0.37 0.85 Political Rights * Institutional 0.04 0.01 0.85 0.25 GDP per capita (PPP) 1.05 10.36*** 1.12 7.61*** Observations 188 188 188 103 103 Adjusted R-squared 0.58 0.73 0.59 0.77 0.61 29 T-ratios in italics. A constant term was used in all regressions but not shown here
High Global Competitiveness Index Low Global Competitiveness vs. Political (Initial, 2004) 6 4 2 FIN USA TWN SWE DNK SGP NOR JPN CHE ISL GBR NLD AUT CAN NZL ARE ISR HKG EST CHL ESP PRT BEL BHRMYS KOR IRL FRA LUX MLT SVN THA GRC HUN CZE LTU TUN ZAFCYP JOR SVK LVA BWAITA MEX MAR SLV MUS TTO URYCRI PAN IND NAM BRA POL EGY ROM HRVBGR COL TUR PER GHA JAM RUSDZA IDN DOM GMB KEN LKA ARG PHL ZMB TZA UGA BIH UKR VEN GTM MKD PAK NGA MWI YUG ECU MLI HND GEO MOZ MDG BOL PRY NIC BGD AGO ETH TCD -1.5 0.5 2.5 Low Political r = 0.60 High Sources: EOS 2004 and Transparenting transparency by BK (2005). N. of countries: 104. 30
Regression Results: Political Dep Variable: Control of Corruption 1 2 3 4 UCM Political 0.61 0.44-0.37 0.05 3.80*** 3.63*** 2.01** 0.34 Political Rights 0.03-0.04 0.12 0.01 0.41 0.68 2.22** 0.20 Political Rights * Political 0.20 0.08 6.58*** 3.10*** GDP per capita (PPP) 1.26 1.15 12.88*** 9.96*** Observations 188 188 188 188 Adjusted R-squared 0.42 0.72 0.52 0.74 31 T-ratios in italics. A constant term was used in all regressions but not shown here
Regression Results: Political Dep Variable: Control of Corruption Dep Variable: Global Competitiveness Index 1 2 3 4 5 6 UCM Political 0.61 0.44-0.37 0.05-1.28-0.24 3.80*** 3.63*** 2.01** 0.34 2.87*** 0.69 Political Rights 0.03-0.04 0.12 0.01 0.19-0.04 0.41 0.68 2.22** 0.20 2.43** 0.59 Political Rights * Political 0.20 0.08 0.31 0.11 6.58*** 3.10*** 5.16*** 2.14** GDP per capita (PPP) 1.26 1.15 1.23 12.88*** 9.96*** 7.24*** Observations 188 188 188 188 103 103 Adjusted R-squared 0.42 0.72 0.52 0.74 0.56 0.77 T-ratios in italics. A constant term was used in all regressions but not shown here 32
related to more competitiveness.. Overall Index (Initial, 2004) and Global Competitiveness Index (GCI) 6.5 6.0 5.5 5.0 4.5 Global Competitiveness In 4.0 3.5 3.0 2.5 2.0 R 2 =0.59 1.5-1.5-1 -0.5 0 0.5 1 1.5 2 2.5 Overall Index Source: GCI: World Economic Forum (2004), Overall Index, 2005 33
Ec/Institutional (Initial, 2004) and Cost to register a business R 2 = -0.47 120 Cost to register a business (%of GNI per capita) 100 80 60 40 20 0 Low transparency Medium transparency High transparency Source: Cost of business: World Development Indicators, 2002; Ec/Institutional Index 2005. 34
Spread of Freedom of Information laws Has the potential of changing the way citizens relate to their governments Twenty years ago only ten OECD nations had laws which specifically guaranteed the rights of citizens to access information from public institutions. Today, over fifty countries have adopted comprehensive FOI laws and over thirty have pending efforts all over the world 35
Source: www.freedominfo.org 36
associated with better human development indicators even after controlling per income Life Expectancy Fem. Literacy Immunization GNI per cap 0.42*** 0.62*** 0.31*** (0.08) (0.18) (0.09) 0.38*** 0.41*** 0.27*** (0.10) (0.16) (0.14) Constant -0.02 0.21 0.10 (0.06) (0.10) (0.06) # obs 164 76 163 R 2 0.43 0.36 0.24 ** indicates significance at.05 level and *** at.01 level. Source: Life expectancy, Female Literacy and Immunization: World Development Indicators, 2003 37
The media can play a key role in increasing the demand for transparency Radios per 1000 hab. vs. Overall Index (Initial, 2004) General 0.8 0.6 0.4 0.2 0-0.2-0.4-0.6 0-200 201-500 501-3300 Radios per 1000 hab Sources: Radios per 1000 people: World Development Indicators 2002, Overall Index 2005, # countries 49 38
FOI laws- necessary but not sufficient Freedom of Information (FOI) Law vs. Extent of Information Actually Provided by Government to Enterprise Sector, 2004 7 Government 4 1 No FOI Processing FOI Freedom of Information Law FOI adopted Sources: Government transparency: Executive Opinion Survey (EOS), 2004 & FOI: Global Survey, 39 2004. # of countries per x-axis category: No FOI (26), Processing FOI (26), FOI Adopted (51).
Freedom of Information Laws and Government, 2004 NO FOI Processing FOI FOI Adopted Good Informati on from Governm ent (EOS) Bahrain, China, Gambia, Hong Kong, Luxembourg Malaysia, Mali, Mauritius Morocco, Singapore Switzerland, Taiwan Tunisia, United Arab Emirates Chile, Germany, Ghana, Namibia Zambia Australia, Canada, Denmark, Finland Iceland, India, Ireland, Japan Netherlands, New Zealand, Norway, Sweden, South Africa United Kingdom, United States Medium Informati on from Governm ent (EOS) Algeria, Costa Rica Cyprus, Jordan Macedonia, Madagascar Malta, Vietnam Botswana, Brazil El Salvador, Indonesia Kenya, Malawi Mozambique, Nigeria Sri Lanka, Uganda Austria, Belgium Colombia, Estonia France, Greece, Israel, Jamaica South Korea, Lithuania, Mexico, Philippines, Portugal, Slovak Republic Slovenia, Spain, Thailand, Trinidad And Tobago Poor Informati on from Governm ent (EOS) Chad, Egypt Venezuela Zimbabwe Argentina, Bangladesh Bolivia, Ethiopia Guatemala, Honduras Nicaragua, Paraguay Russia, Tanzania Uruguay Angola, Bosnia-Herzegovina Bulgaria,Croatia, Czech Republic, Dominican Republic, Ecuador, Georgia, Hungary, Italy, Latvia, Pakistan, Panama, Peru, Poland, Romania, Serbia, Turkey, Ukraine Source: Freedominfo (2004) and EOS (2004). 40
Freedom of Information Law (FOIL) vs Fiscal (IMF) NO FOIL Processing FOIL FOIL Adopted High (IMF rating >.5) BRAZIL CHILE GERMANY PAPUA NEW GUINEA ALBANIA,ARMENIA,BULGARIA,CANADA, CZECH REPUBLIC, ESTONIA,FRANCE, GREECE, HUNGARY, INDIA, ISRAEL, ITALY, JAPAN,LATVIA, ITHUANIA,MEXICO, POLAND PHILIPPINES, PORTUGAL, ROMANIA,SLOVAK REPUBLIC SLOVENIA, SWEDEN UKRAINE, UNITED STATES Low (IMF rating =<.5) AZERBAIJAN,BENIN, BURKINA FASO,CAMEROON IRAN,KAZAKHSTAN; KYRGYZ REPUBLIC, MALI,MAURITANIA,MONGOLIA, RWANDA, TUNISIA BANGLADESH GHANA,HONDURAS MALAWI,MOZAMBIQUE NICARAGUA,SRI LANKATANZANIA, UGANDA, URUGUAY COLOMBIA GEORGIA KOREA, SOUTH PAKISTAN PERU TURKEY Declined AFGHANISTAN, ALGERIA, ANDORRA, ANTIGUA AND BARBUDA,BAHAMAS, BAHRAIN, BARBADOS, BHUTAN, BRUNEI, BURUNDI, CAMBODIA, CAPE VERDE, CENTRAL AFRICAN REPUBLIC, CHAD, CHINA, COMOROS, CONGO, Congo, Dem. Rep. (Zaire), COSTA RICA, CUBA, CYPRUS, DJIBOUTI, DOMINICA, EGYPT, EQUATORIAL GUINEA, ERITREA, FIJI, GABON, GAMBIA, GRENADA, GUINEA, GUINEA-BISSAU, GUYANA, HONG KONG, IVORY COAST, JORDAN, KIRIBATI, KOREA, NORTH, KUWAIT LAOS, LEBANON, LIBERIA, LIBYA, LUXEMBOURG, MACEDONIA, MADAGASCAR, MALAYSIA, MALDIVES, MALTA, MARSHALL ISLANDS, MAURITIUS, MICRONESIA, MONACO, MOROCCO, MYANMAR, NAURU, NIGER, OMAN, PALAU, QATAR, SAMOA, SAN MARINO, SAO TOME AND PRINCIPE, SAUDI ARABIA, SENEGAL, SEYCHELLES, SIERRA LEONE, SINGAPORE, SOLOMON ISLANDS, SOMALIA, ST. KITTS AND NEVIS, ST. LUCIA, ST. VINCENT AND THE GRENADINES, SUDAN, SURINAME, SWAZILAND, SWITZERLAND, SYRIA, TAIWAN, TIMOR, EAST, TOGO, TONGA, TURKMENISTAN, TUVALU, UNITED ARAB EMIRATES, VANUATU, VENEZUELA, VIETNAM, WEST BANK GAZA, YEMEN, ZIMBABWE ARGENTINA, BELARUS BOLIVIA, BOTSWANA EL SALVADOR ETHIOPIA, GUATEMALA, HAITI INDONESIA, KENYA LESOTHO, NAMIBIA NEPAL, NIGERIA PARAGUAY, RUSSIA ZAMBIA ANGOLA, AUSTRALIA, AUSTRIA, BELGIUM, BELIZE, BOSNIA- HERZEGOVINA, CROATIA, DENMARK, DOMINICAN REPUBLIC ECUADOR, FINLAND,ICELAND,IRAQ IRELAND,JAMAICA,LIECHTENSTEIN MOLDOVA,NETHERLANDS NEW ZEALAND, NORWAY,PANAMA SERBIA,SOUTH AFRICA SPAIN,TAJIKISTAN THAILAND TRINIDAD AND TOBAGO UNITED KINGDOM UZBEKISTAN 41
FOI Laws & Fiscal (IMF) in Latin America NO FOI Processing FOI FOI Adopted High (IMF rating >.5) BRAZIL CHILE MEXICO HONDURAS COLOMBIA Low (IMF rating =<.5) NICARAGUA URUGUAY PERU ANTIGUA AND BARBUDA ARGENTINA BELIZE Declined BAHAMAS BARBADOS COSTA RICA CUBA DOMINICA GRENADA BOLIVIA EL SALVADOR GUATEMALA HAITI PARAGUAY DOMINICAN REPUBLIC JAMAICA PANAMA TRINIDAD AND TOBAGO ST. KITTS AND NEVIS ST. LUCIA ST. VINCENT AND THE GRENADINES 42
Case study Bolivia diagnostic Aggregate Cross-Country empirical analysis is insufficient: i) rough tool for policy applications at a country level; ii) challenges of causality direction Thus, complementing with micro-data within one country One such survey: In-depth survey of 1250 officials in 90 public institutions in Bolivia, accounts for endogeneity = % of cases where the actions of public officials & decision-making process are transparent Main findings service delivery performance depends negatively on the level of corruption and positively on external voice of users and transparency -related determinants seem to be more relevant in explaining performance of public agencies that other variables such as wage satisfaction or internal organizational rules 43
6a: Public Service Delivery and in Bolivia's Public Institutions High Service Delivery Performance 100 80 60 40 r =.58 PoliceSantaCruz CustomsSantaCruz TaxSantaCruz CustomsCochabamba JudicialCouncil HighCourt-SantaCruz NatPolice NatCustoms HighCourt-LaPaz Ombudsman NatComptroller ConstitutionalTribune CentralBank Low 20 30 40 50 60 70 80 90 Low High Source: Kaufmann, Mastruzzi and Zavaleta (2003), based on 90 national, departmental and municipal agencies covered in the Public Officials Survey 44
of Public Institutions and Purchase of public positions, Bolivia. Source: Kaufmann, Mastruzzi and Zavaleta (2003), based on 90 national, departmental and municipal agencies covered in the Public Officials Survey 45
Economic/Institutional & Political Index (Initial, 2004) High 2.5 Ec/Institutional Index Political Index 0 Indices OECD East Asia NICs East Asia dev. South Asia Sub-saharan Africa M. East/N. Afr Former Soviet Union Eastern Europe Latin America Caribbean -2.5 CHILE Low 46 Index ranges from -2,5 (worst) to 2.5 (best).
The case of Chile: unbundling transparency Since 1996, Chile has made considerable advances in governance indicators compare to the Latin American average, and today it is highly rated in governance In terms of economic and institutional transparency, Chile is in 8 th position (eg Central Bank of Chile among the best rated in effectiveness and transparency in the world) Yet Chile faces challenges on political transparency -- the gap : Ec/Institutional = 2.38 vs. Political = 0.82 Weak areas requiring specific progress: Implementation of Freedom of Information law with effective mechanism to access the information Financial disclosure for public officials, legislators, judges, etc. Further transparency in ( sole sourced ) procurement Disclosure of political funding/expenditures (& lobby law) 47
Policy applications- reforms as second generation institutional change Despite potential benefits, transparency reforms rarely integrated into institutional reform programs reforms can be substitutes to (over)- regulation, to creation of additional public institutions (eg. A-C agencies) and incessant legal drafting -- which have higher costs and less benefits reforms often require political capital: but if present there are significant technocratic areas where the IFIs can have a much larger and key role reforms have low financial cost, and high benefits: net savers of resources (eg. E*procurement) reforms well-suited to be entry points catalyzing further institutional change, since: i) effective in changing incentives of political leaders to serve broad social groups, and ii) politically more feasible than industry wide-regulation 48
Basic Scorecard: 10 Reform Components 1. Public Disclosure of Assets and Incomes (and outside earnings) of Candidates, Public Officials, Politicians, Legislators - & dependents 2. Public Disclosure of Political Campaign contributions by individuals and firms, and of campaign expenditures 3. Public Disclosure of Parliamentary Votes, w/out exceptions 4. Effective Implementation of Conflict of Interest Laws, separating business, politics, legislation, & government 5. Publicly blacklisting firms bribing in public procurement 6. Effective Implementation of Freedom of Information Law, with easy access to all to government information 7. Fiscal/Financial transparency: central/local budgets; EITI 8. E*procurement: transparency (web) and competition 9. Adoption and implementation of Lobby Law 10. Judiciary transparency and disclosure on funding, assets of judges, and on full disclosure of judicial decisions 49
Should IFIs Scale up work on? IFI Adopting and Mainstreaming Strategies, based on Diagnostics that unbundle and are comprehensive (not just narrow economic transparency) Integrating to our analysis as Regulatory-relieving and Institutionally-relieving: with an exception supporting more IFAIs? IFI transparency reforms internally within our own institutions Back to basics : as the topic of December conference? 50
Data for Analysis and Informing Policy Advice, Not for Precise Rankings Any data on Governance, Institutions, and Investment Climate is subject to a margin of error. It is not intended for precise country rankings, but to highlight relative strengths and weaknesses and draw analytical and policy lessons. The data presented here and in the Report do not necessarily reflect official views on rankings by the World Bank or its Board of Directors. Errors are responsibility of the authors. http://worldbank.org/wbi/governance 51