Transparenting Transparency Some Empirics and Policy Applications. Daniel Kaufmann and Ana Bellver World Bank Institute

Transcription

1 Transparenting Some Empirics and Policy Applications Daniel Kaufmann and Ana Bellver World Bank Institute Presentation at the Conference on and Governance, Centre on Asia and Globalization, Lee Kwan Yew School of Public Policy, National University of Singapore, May 10 th,

2 Main Messages of Transparenting Major contributions have taken place on right to know & access to information, & on transparency in general New Book by Ann Florini: great contribution Generally, in the field, less progress on: -- measurement and unbundling of transparency -- empirical analysis -- policy analysis and strategy/policy focus We propose a Index (related to Governance Indicators): Economic/Institutional & Political dimensions 2

3 Main Messages, Cont. Finding: large variance worldwide in terms of transparency: i) cross-country: incl. within same region; transparency is present in some emerging economies; and, ii) within country: economic vs. political transparency the deficit associated with better socioeconomic indicators, less corruption, & improved competitiveness: i) institutional transparency in linear fashion; political in non-linear fashion Results are supported at the micro level: incountry diagnostic complement: (Bolivia & Chile) Policy implications: transparency is a core for 2 ndgeneration institutional reforms -- role of IFIs in 3 supporting reforms, & in our internal reforms

4 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 4 Disclosure of Politician s Assets, Voting records, etc

5 Sources of Data 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 5

6 African Development Bank Sources for Indicators 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 Institute for Management Development OECD/WB 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 World Competitiveness Yearbook Budget Practices and Procedures survey poll poll poll poll objective assessment poll survey survey objective assessment Polity IV Project Political Regime Characteristics and Transitions, poll 6

7 Data Sources for 2005, Cont d Reporters Without Borders State Department/ Amnesty International International UN Economic Commission for Africa Reporters Without Borders CIRI Human Rights dataset Global Corruption Report 2004 African Governance Indicators, 2005 poll poll objective assessment poll United Nations World Public Sector Report 2003 poll World Bank World Bank World Economic Forum Islam Index 2004 Country Policy & Institutional Assessment Global Competitiveness Report objective assessment poll survey 7

8 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) 8

9 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 9

10 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 10

11 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 11

12 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 ) 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) 12

13 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) These standard errors are smaller for countries that (a) appear in more sources, and/or (b) appear in more reliable sources 13

14 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 14

15 Levels of Worldwide 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) 15

16 Institutional, Bottom Quartile 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 High Level Margin of Error Low 16 Source: Transparenting by BK (2005). N. of countries: 194.

17 Institutional, Middle-Low Quartile 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 High Level Margin of Error Low Source: Transparenting by BK (2005). N. of countries:

18 Institutional, Middle-High Quartile 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 High Level Margin of Error Low Source: Transparenting by BK (2005). N. of countries:

19 Institutional, Top Quartile HUNGARY BRAZIL HONG KONG SPAIN POLAND SLOVENIA TAIWAN ITALY KOREA, SOUTH GERMANY JAPAN MEXICO FRANCE ESTONIA SINGAPORE NEW ZEALAND DENMARK SWEDEN AUSTRALIA UNITED KINGDOM CHILE CANADA UNITED STATES High Level Margin of Error Low Source: Transparenting by BK (2005). N. of countries: 194.

20 Institutional, Selected Countries High Level Margin of Error 0.5 Low -2.5 LIBERIA KOREA, NORTH SYRIA MALAWI MYANMAR BURUNDI ZAMBIA SENEGAL VENEZUELA BANGLADESH 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. GEORGIA FIJI GUATEMALA BOTSWANA PARAGUAY RUSSIA CHINA INDIA UKRAINE BRAZIL JAPAN ESTONIA SINGAPORE UNITED KINGDOM 20 CHILE UNITED STATES

21 Political, Bottom Quartile KOREA, NORTH CUBA LIBYA TURKMENISTAN ZIMBABWE CHINA SUDAN SAUDI ARABIA BELARUS SYRIA SOMALIA RWANDA EGYPT BHUTAN 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.

22 Political, Middle-Low Quartile ANGOLA BANGLADESH MALDIVES ETHIOPIA ALGERIA JORDAN PAKISTAN VENEZUELA CAMBODIA ZAMBIA COLOMBIA NIGERIA CONGO GUATEMALA KENYA TURKEY INDONESIA UGANDA HONDURAS PARAGUAY MOZAMBIQUE MADAGASCAR High Level Margin of Error Low Source: Transparenting by BK (2005). N. of countries: 194.

23 Political, Middle-High Quartile CROATIA INDIA MEXICO BULGARIA BOLIVIA NAMIBIA ARGENTINA BRAZIL BOTSWANA KOREA, SOUTH ITALY ISRAEL HONG KONG GREECE GHANA CHILE SOUTH AFRICA HUNGARY URUGUAY 23 SLOVAK REPUBLIC AUSTRIA CYPRUS BELIZE High Level Margin of Error Low Source: Transparenting by BK (2005). N. of countries: 194.

24 Political, Top Quartile NORWAY POLAND COSTA RICA SPAIN FRANCE JAPAN LATVIA UNITED KINGDOM ESTONIA GERMANY CANADA PORTUGAL UNITED STATES MARSHALL ISLANDS LUXEMBOURG NEW ZEALAND SWITZERLAND FINLAND ICELAND SWEDEN DENMARK High Level Margin of Error Low Source: Transparenting by BK (2005). N. of countries: 194.

25 Political, Selected Countries High Level Margin of Error 0 Low -2.5 KOREA, NORTH CUBA VIETNAM CHINA LIBERIA CHAD CAMEROON SINGAPORE 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 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. HONDURAS GEORGIA PANAMA ROMANIA NAMIBIA ARGENTINA PHILIPPINES BOTSWANA ITALY GHANA CHILE SPAIN PORTUGAL 25 BELGIUM DENMARK

26 Overall, Selected Countries KOREA, NORTH CUBA SOMALIA SUDAN YEMEN IVORY COAST GAMBIA CHINA ZAMBIA VENEZUELA NIGERIA GEORGIA JORDAN MADAGASCAR SENEGAL PARAGUAY UKRAINE TURKEY PANAMA BOLIVIA EL SALVADOR SINGAPORE BRAZIL 26 SLOVAK REPUBLIC MEXICO ITALY CHILE DENMARK CANADA UNITED STATES High Level Margin of Error Low 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.

27 Economic/Institutional & Political Index, preliminary High Indices -2.5 OECD East Asia NICs East Asia dev. South Asia Sub-saharan Africa M. East/N. Afr Former Soviet Union Eastern Europe Latin America Caribbean CHILE SINGAPORE Low Index ranges from -2,5 (worst) to 2.5 (best). 27

28 Control of Corruption vs. Institutional High Control of Corruption Low r = 0.71 FIN ISL LUX CHE SWE AUT NOR NLD AUS GBR CAN DEU LIE HKG IRL BEL ESP FRA BHS ARE MLT PRT JPN SVN ATG BRB BWA CYP OMN KWT BHR CRI ISR EST 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 NERYEM SLE MWI BOL ETH PNG MHL LBY MDA PAK DJI TGO KEN VEN UKR SWZ KGZ GEO IDN COGCIV ZWE KHM BLR AZE PRY LAO COM TJK AGO BDI NGABGD KAZ SLB UZB AFGTKM SDN ZAR CAF PRK HTI IRQ MMR SOM GNQ SGP NZL DNK Low Institutional High 28 Sources: Governance Matters IV by KKM (2005) and Transparenting transparency by BK (2005). N. of countries: 190 CHL

29 Global Competitiveness vs. Institutional 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 Low Institutional High Sources: EOS 2004 and Transparenting transparency by BK (2005). N. of countries:

30 Regression Results: Institutional Dep Variable: Control of Corruption Dep Variable: Global Competitiveness Index UCM Institutional *** 3.93*** *** 2.02** Political Rights *** 5.43*** 3.79*** Political Rights * Institutional GDP per capita (PPP) *** *** Observations Adjusted R-squared T-ratios in italics. A constant term was used in all regressions but not shown here

31 High Control of Corruption Low Control of Corruption vs. Political r = 0.76 SGP ARE FIN NZL ISL DNK LUXCHE AUT SWE GBR NLDNOR CAN AUS DEU USA LIE IRL BEL ESP FRA BHS JPN MLT PRT SVN ATG BWA OMN BHR KWT ITA ISR CYP CRI BRB EST BTN HUN TWN QAT GRC ZAF URY GRD SVK SUR TUN MYS JOR MUS KNA LTU CPV CZE VCT LCA BRN DMA SAU NAM LVA MDV KOR POL HRV MRT SYC SAM MAR TTO PAN BGR LSO KIRNRU SMR MCO PCI BLZ LKA MDG FJI EGY COL BRA TUR GHA ROM THA MEX TMP RWA YUG SEN PER NIC FSM LBN BFA IND SLV BEN GUY CHN DZA MKD DOM MNG ARG VUT WBGIRN TZA PHL MLI NPL GAB ARM BIH JAM ERI GMB VNM SYR RUS GNB ZMB UGA HND TON STP GTM ALB CMR MOZ ECU TUV YEM TGO UKR PAK MDASLENER MWI LBY LBR GIN BOL BLR ETH VEN ZWE SWZ PNG MHL KGZ KEN DJI IDN GEO CIV KHM AZE COG PRY LAO KAZ TJK NGA COM UZB BDI AGO BGD SLB TKM SDN ZAR AFG CAF HTI IRQ SOM GNQ Low Political HKG CHL High Sources: Governance Matters IV by KKM (2005) and Transparenting transparency by BK (2005). N. of countries:

32 Global Competitiveness vs. Political High Global Competitiveness Index Low 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 Low Political r = 0.60 High Sources: EOS 2004 and Transparenting transparency by BK (2005). N. of countries:

33 Regression Results: Political Dep Variable: Control of Corruption UCM Political *** 3.63*** 2.01** 0.34 Political Rights ** 0.20 Political Rights * Political *** 3.10*** GDP per capita (PPP) *** 9.96*** Observations Adjusted R-squared T-ratios in italics. A constant term was used in all regressions but not shown here

34 Regression Results: Political Dep Variable: Control of Corruption Dep Variable: Global Competitiveness Index UCM Political *** 3.63*** 2.01** *** 0.69 Political Rights ** ** 0.59 Political Rights * Political *** 3.10*** 5.16*** 2.14** GDP per capita (PPP) *** 9.96*** 7.24*** Observations Adjusted R-squared T-ratios in italics. A constant term was used in all regressions but not shown here 34

35 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. Nowadays, well over fifty countries have adopted comprehensive FOI laws and over thirty have pending efforts all over the world 35

36 36

37 Table 1. 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). 37

38 FOI laws- necessary but not sufficient Freedom of Information (FOI) Law vs. Extent of Information Actually Provided by Government to Enterprise Sector, Government 4 1 No FOI Processing FOI Freedom of Information Law FOI adopted Sources: Government transparency: Executive Opinion Survey (EOS), 2004 & FOI: Global Survey, # of countries per x-axis category: No FOI (26), Processing FOI (26), FOI Adopted (51). 38

39 Freedom to Information Law (FOIL) vs Fiscal (IMF) IMF Fiscal FOIL: NO FOIL Processing FOIL FOIL Adopted High (IMF rating >.5) BRAZIL CHILE GERMANY PAPUA NEW GUINEA ALBANIA,ARMENIA,BULGARIA,C ANADA, CZECH REPUBLIC ESTONIA,FRANCE, GREECE, HUNGARY, INDIA, ISRAEL, ITALY, JAPAN,LATVIA, LITHUANIA,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 LANKA TANZANIA,UGANDA URUGUAY COLOMBIA GEORGIA KOREA, SOUTH PAKISTAN PERU TURKEY Declined IMF 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 KINGDOM39 UZBEKISTAN

40 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 40 organizational rules

41 6a: Public Service Delivery and in Bolivia's Public Institutions High Service Delivery Performance r =.58 PoliceSantaCruz CustomsSantaCruz TaxSantaCruz CustomsCochabamba JudicialCouncil HighCourt-SantaCruz NatPolice NatCustoms HighCourt-LaPaz Ombudsman NatComptroller ConstitutionalTribune CentralBank Low Low High Source: Kaufmann, Mastruzzi and Zavaleta (2003), based on 90 national, departmental and municipal agencies covered in the Public Officials Survey 41

42 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 42

43 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 43 Disclosure of political funding/expenditures (& lobby law)

44 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 Ann Florini s Regulation by Revelation Indeed! 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 (e.g. e*procurement) reforms well-suited to be entry points catalyzing further institutional change 44

45 Basic Scorecard: 10 Reform Components 1. Public Disclosure of Assets and Incomes (and outside earnings) of Candidates, Public Officials, Politicians, Legislators & Judiciary 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 8. E*procurement: transparency (web) and competition 9. Freedom of the Press / Media 10. Power of Data : Encouraging international benchmarking and incountry surveys on transparency, to monitor progress 45

46 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 and integrity reforms internally within our own institutions example starting from home 46

47 Power of Data, and Citizen Oversight Tracking Education spending in Uganda equiv. US$ per student Public info campaign Intended grant Actual grant received by primary school (means) Source: Uganda Public Expenditure Tracking Surveys 47

48 On, the Power of Data, Governance and Development There has never been a famine in any country that has been a democracy with a relatively free press... I know of no exception. Amartya Sen, Nobel Laureate It is now generally recognized that better, more timely, information results in better, more efficient resource allocations. Joseph Stiglitz, Nobel Laureate If you cannot measure a problem, you cannot address it. Lord Kelvin 48

49 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. 49