55637 BETTER REGULATION FOR GROWTH GOVERNANCE FRAMEWORKS AND TOOLS FOR EFFECTIVE REGULATORY REFORM REGULATORY QUALITY INDICATORS INVESTMENT CLIMATE ADVISORY SERVICES WORLD BANK GROUP WB204_RQI.indd i 5/17/10 3:49:03 PM ©2010 The World Bank Group 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org All rights reserved Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The World Bank encourages dissemination of its work and will normally grant permission to repro- duce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of The Publisher, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. About the Investment Climate Advisory Services of the World Bank Group The Investment Climate Advisory Services (IC) of the World Bank Group helps governments implement reforms to improve their business environment, and encourage and retain investment, thus fostering competitive markets, growth and job creation. Funding is provided by the World Bank Group (the International Finance Corporation­IFC, the Multilateral Investment Guarantee Agency­MIGA, and the World Bank) and over 15 donor partners working through the multi-donor FIAS platform. The findings, interpretations and conclusions included in this note are those of the author and do not necessarily reflect the view of the executive directors of the World Bank Group or the governments they represent. Better Regulation for Growth Program The Better Regulation for Growth (BRG) Program was launched in 2007 by the Dutch Ministry of Foreign Affairs, the UK Depart- ment for International Development (DFID) and the IC of the World Bank Group. The objective of the BRG Program is to review and synthesize experiences with regulatory governance initiatives in developing countries, and to develop and disseminate practical tools and guidance that will help developing countries design and implement effective regulatory reform programs. Reports and other documentation developed under the BRG Program are available at: www .ifc.org/brg WB204_RQI.indd ii 5/17/10 3:49:06 PM Table of Contents Acknowledgements.............................................................................. v Executive Summary ............................................................................ vi Introduction ......................................................................................... 1 Conceptual Issues................................................................................. 3 Data Sources and Database on Regulatory Quality and Governance .... 5 OECD Product Market Regulation .........................................................................5 World Bank: Indicators of governance and institutional quality ...............................10 World Bank: Doing Business indicators ................................................................11 OECD: Indicators of government capacity to produce high-quality regulations ..........13 Surveys of independent regulatory authorities and regulators .................................13 Surveys of business ...........................................................................................14 Country fact-sheets on regulatory reform, Better Regulation and RIA ........................14 Datasets on RIAs and methodology ....................................................................15 An overview ....................................................................................................15 How governments use existing indicators ............................................................21 Use of Indicators ................................................................................ 21 How governments use existing indicators ............................................................21 Set of indicators ................................................................................................22 Scorecards .......................................................................................................22 Regulatory process standards and checklists .........................................................22 Review by auditing bodies, internal watchdogs, and external evaluations ................23 Surveys ............................................................................................................24 Regulatory budgets, administrative burdens, and agendas .....................................25 How international organizations use indicators .....................................................27 How academics use indicators............................................................................29 Conclusions, Critical Remarks and Suggestions ................................... 31 Bibliography ...................................................................................... 34 Annex 1: Better Regulation for Growth Program ................................ 39 iii WB204_RQI.indd iii 5/17/10 3:49:06 PM List of Tables and Figure Tables Table 1: Mapping Databases on Regulatory Quality and Governance .......................6 Table 2: Advantages and Disadvantages of Different Types of Measures ..................16 Table 3: Purpose, Aim and Types of Measurement of Databases .............................18 Figure Figure 1: Correlation Between "Government Effectiveness" and Other WBI Governance Indicators.......................................................................................11 iv WB204_RQI.indd iv 5/17/10 3:49:06 PM ACKNOWLEDGEMENTS This report has been prepared by Claudio Esra Yilmaz, Eva Buitenkamp, Fabrizio de Radaelli and Fabrizio de Francesco, University of Francesco, Greg Bounds, Jeroen Nijland, Jose- Exeter. phine Kanyi, Ksenija Vidulic, Lars Grava, Margo Thomas, Mustafizur Rahman, Nick Godfrey, An early draft of this document was discussed at Richard Sandall, Stephen Rimmer, Scott Jacobs the Technical Meeting of the Advisory Panel of and Ulrich Ernst. the Better Regulation for Growth Program held in The Hague on November 13­14, 2008. Furthermore, advanced drafts of the report ben- Comments to this draft were provided by par- efited from valuable comments from Adrian ticipants to this meeting: Andre Nijsen, Andreja Nembhard, Delia Rodrigo, Lars Grava, Masrur Marusic, Andrea Renda, Claudio Radaelli, Colin Reaz, Peter Ladegaard, Richard Sandall, Stephen Kirkpatrick, Delia Rodrigo, Edward Donelan, Rimmer, and Ulrich Ernst. v WB204_RQI.indd v 5/17/10 3:49:09 PM EXECUTIVE SUMMARY1 This working paper is part of a set of two papers Systems of indicators have been developed commissioned by IC to critically discuss the state ad hoc, by organizations pursuing different of the art in regulatory indicators and make sug- objectives over time and across the world. gestions for project-level indicators. Conceptual underpinnings, modalities of data-gathering, and types of usage reflect In this first working paper we present and compare different operationalizations of regulatory systems of indicators of regulatory quality, analyz- quality. ing their conceptual underpinnings, technical properties, and usage by governments, stakehold- In the construction of regulatory indicator ers and academics. After having discussed the data- datasets, the first essential step is to gauge the sets and the types of data available, we consider a quality of data, in order to construct insight- set of critical aspects related to the design, data- ful and meaningful measurements. Especially gathering and utilization of indicators, providing in developing countries, the consideration of suggestions for improvement. the quality of data is preliminary to any tech- nical and statistical discussion on how to Our major findings and proposals are the treat data. following: Western governments tend to use single mea- 1 This paper was written by Fabrizio De Francesco and Clau- sures with the specific purpose of improving dio Radaelli, University of Exeter. We are grateful to Peter methods of regulatory analysis (through stud- Ladegaard and his team at IC for having assisted us with ies that analyze the conduct of regulators). comments and perceptive insights throughout the produc- tion of this paper and the companion paper on Project level Little attention is paid to indicators of the indicators. Our research on regulatory indicators is also economic outcomes of regulatory reform pro- connected to a project on Regulatory impact assessment in comparative perspective funded by the Economic and Social grams and to improving their accuracy. This Research Council of the UK (Feb 2005­Jan 2009). is also due to the complexity in disentangling vi WB204_RQI.indd vi 5/17/10 3:49:10 PM the mechanism that links an improvement of We have not found systematic attempts to regulatory outcome to economic performance. measure how regulatory reform is changing the Thus, the usefulness of reform programs and attitudes towards regulation of regulators and regulation itself is poorly understood. This policy officers who develop legislation. Surveys may be even truer in developing countries of regulators are scarce and not-coordinated. where the quality of micro and macroeco- Yet if one fundamental aspect of regulatory nomic indicators is lower than in developed reform is to change attitudes and approaches countries. to regulation, this dimension cannot be ignored. The experience with business surveys Most datasets contain simple measures rather is more robust. than composite indexes. The necessary inte- gration of regulatory quality measures with Apart from the fundamental issue of data- other components of structural reform and gathering, it is important, particularly in governance is generally low. Exceptions are developing countries, to link regulatory indi- the regulatory quality indicators of the World cators with processes of knowledge utilization Bank ­ integrated with other dimensions of and decision making. Knowledge utilization good governance ­ and the integration of is more important than knowledge produc- administrative burdens indicators and targets tion in this field. Very few governments have with the European Union (EU) Lisbon taken credible commitments about which agenda for growth and jobs. Statistical analy- indicators they are going to use, when, and sis shows that regulatory reform is correlated how. Examples are targets, benchmarking with other broader dimensions of governance. exercises in international organizations, peer reviews, and regulatory budgets. Although the indicators currently available have been created for different purposes, the Turning to usage, indicators can be used for a systems are not mutually exclusive. Indeed, variety of purposes, such as learning, over- there is scope for cumulating different sources sight, monitoring compliance with regulatory or combining insights from different mea- policy, naming and shaming, and comparative sures. However, this potential has not as yet benchmarking. There is a full range of possible been exploited. usages, in a continuum from "hard" to "soft" system of indicators. In one extreme, one can Different indicators track different dimen- find indicators that are used for overseeing sions of quality and measure the performance regulators and monitoring compliance with of different tools. Per se, this is not a limita- government's regulatory policy. In the middle tion. Governments need an integrated, com- ground there are comparative benchmarks. prehensive approach to regulatory reform. Finally, indicators are also learning tools. Different systems provide different insights on dimensions that can be usefully com- Consultants and academics have presented bined. However, the predominant mode of several ways in which regulatory indicators usage at the level of governments is still a should be aggregated. In this working paper, rather eclectic "pick and choose" from the however, we make the point that aggregation wide range of measures available. Accord- performed directly by policymakers is the best ingly, in the second part of the paper we argue option ­ obviously subject to technical con- that a system of measurement should balance trols on the consistency of the aggregation. the different indicators in relation to their Aggregation of simple measures is one major purpose, data source, and methodology. way in which preferences about regulatory vii WB204_RQI.indd vii 5/17/10 3:49:11 PM quality of political actors are revealed. When a propose also a system of measurement that government transparently chooses an indica- relies exclusively on the judgment of profes- tor instead of another, or aggregates by put- sional independent evaluation in order to ting more weight on a measure rather than review, correct and improve (through evalua- another in order to create an index, it makes a tion standards) the overall better regulation choice about how to value different dimen- policy. sions of regulatory quality. It also sets the basis for possible usages of regulatory measures, by Aggregation should reflect basic technical clarifying why an index reflects reform priori- rules, but more importantly, it should be per- ties better than another. Thus, processes of formed in a way that conveys messages to aggregation that involve those who are directly those who are supposed to make use of the responsible for setting regulatory policy are composite measures. To illustrate, if the goals the first step in the direction of utilization. of regulatory reform are institutionalization, economic growth, accountability, interaction To clarify this point and introduce account- with the stakeholders and communication, ability, we envisage a division of labor aggregation should be performed along these between independent experts, on the one four dimensions, not in relation to abstract hand, and senior civil servants and elected technical properties. politicians, on the other. Experts should be the guardians of the technical quality of Finally, utilization should also involve regula- menus of indicators suggested for adoption tors. At the moment, regulatory datasets are to policymakers. The selection and aggrega- not a resource for a desk officer developing tion of measures taken from these menus regulations on, for instance, energy efficiency should be usefully left to a process of delib- in country A. This officer will not have access eration and choice led by policymakers. They to the methods and data used by other offi- should use selection and aggregation to reveal cers in countries B, C and D to prepare their their definition of regulatory quality and sort impact assessments or other evidence-based out their preferences transparently. This does analyses on this topic. Since the number of not mean that anything goes, and that senior key regulations is not infinite, and processes bureaucrats and elected politicians should of convergence, standardization, and regula- operate within unaccountable processes. tory harmonization have ushered in some They should be accountable to their citizens forms of regulatory rapprochement, there is a about why and how they choose a specific demand for datasets that would improve reg- measurement system and what they want to ulatory analysis and methodology. We have achieve with it. Moreover, as the paper on found a prototype of this approach to indica- Project Level Indicators shows in detail,2 we tors, Diadem (Section 2), a product still under construction but with considerable 2 Project Level Indicators of the Better Regulation for Growth potential. (BRG) Program. viii WB204_RQI.indd viii 5/17/10 3:49:11 PM INTRODUCTION With the emergence of the regulatory state3 and about the incidence of anticipated phenomena of the consolidation of the new public manage- change . . . [and are] in relation to the goals that ment movement, scholars, international organi- are meant to be achieved, they are couched in zations and governments have paid attention to terms of what is expected to happen" (Whitmore the concepts and measures of extent and quality et al., 2006: 345). of regulation. Indeed, indicators of regulatory quality, performance measures of regulatory gov- And what is expected to happen is an improve- ernance and targets for the reduction of paper- ment of the regulatory environment. Since quality work and administrative burdens have become refers to properties that constitute or characterize prominent items on the reform agenda of gov- something (the features) as well as appraisal and ernments and international organizations. judgment of the features, it is necessary to set a clear association between indicators (the measure, In this paper we review the literature and the expe- the yardstick) as a component of broader evalua- rience of governments with indicators of regula- tion processes and the phenomena to observe and tory quality, identifying their main advantages and measure. In doing so, it is possible to clarify the limitations. This choice is informed by the delicate definition of (regulatory) quality, the latter being balance between positive and normative consider- the concern and claim of evaluation studies (Stake ations at work in this field. Indeed, indicators are a and Schwandt, 2006: 404). common language in evaluative processes through which people translate diversity of observations The next essential step is to recognize that regula- into manageable information "bits" (Whitmore tory quality is not univocal and depends, among et al., 2006: 345). They "only provide information other things, on the institutional and administra- tive context in which the term is used (Radaelli and 3 See "What is Regulatory Governance," the overview paper of the BRG program, for further background on the emer- De Francesco, 2007: chapter 2). Accordingly, there gence of the regulatory state. are several methodologies, methods and techniques 1 WB204_RQI.indd 1 5/17/10 3:49:12 PM to discern and represent quality ­ via external stan- World Bank and the European Union (EU). dards, participants and stakeholders' judgment and With two tables we also draw attention to the perception, and arguably even intuition, and lived relationship between indicators and dimensions experience (Stake and Schwandt, 2006). Unless we of evaluation. We relate indicators to better regu- enter context, "[q]uality is multi-faceted, contested, lation policy tools, what type of data are used and never fully representable" (Stake and Schwandt, (objective or subjective measures), what is mea- 2006: 405). In other words, political institutions sured (the quality of regulatory environment or and administrative contexts reduce complexity, the broader quality of governance), and finally since they provide established rules of the game for whether and how they are integrated in an over- the democratic aggregation of preferences and dis- all evaluation process. cipline of the access and participation of stakehold- ers in policy formulation processes (Radaelli and It is important to note that regulatory indicators De Francesco, 2007: chapter 2). can operate as an information device. Information is an important resource on which the relationship Given this set of analytical challenges, this paper among politicians, regulators and stakeholders is is structured as follows. The next section exam- based. By providing new additional information ines the conceptual basis of our analysis of indi- on the quality of regulatory process and environ- cators of regulatory quality. We then provide a ment, indicators are capable of modifying the rela- map (in Sections 2 and 3) of the properties and tionship among the interest parties (De Francesco features of indicators currently in use, to com- and Radaelli, 2007). Thus, we argue for a trans- pare and identify gaps and flaws in evaluation parent and accountable process of adoption and systems used by governments and international selection. Section 4 concludes by formulating sev- organizations such as the Organization for Eco- eral suggestions on how to improve the measure- nomic Cooperation and Development (OECD), ment systems for regulatory quality. 2 WB204_RQI.indd 2 5/17/10 3:49:13 PM CONCEPTUAL ISSUES Before we embark upon an examination of mea- that regulations survive scrutiny in cabinet and sures, we need to look at the concepts around parliament, and increase the popularity of the which indicators are developed. Most of the indi- incumbent. For a social scientist, good regulation cators developed so far are linked to a notion of means either "efficient regulation" or "legitimate regulatory reform captured by the labels "better regulation," depending on how economic and regulation" or "capacity of governments to pro- socio-political considerations are balanced within duce high quality regulation." But exactly what is the preference function of the social scientist. regulatory quality? For a civil servant educated in the continental The concept of regulatory quality is particularly European legal tradition, more often then not difficult to define. It is contingent on different procedures and administrative acts provide the regulatory reform priorities such as deregulation, most visible clue to good regulation. Hence a public protection from hazards at a reasonable civil servant would normally feel that good regu- cost, or the creation of different political oppor- lations are the ones developed by following stan- tunity structures for pressure groups and bureau- dard operating procedures and administrative crats, without having a final regulatory outcome procedures. Thus, there are different legitimate in mind. claims about what is high-quality regulation. Most countries have provided list of principles Moreover, different stakeholders have different and regulatory priorities documents to make the preferences about regulatory quality. For devel- notion of regulatory quality more concrete. opment agents such as the World Bank, the pur- pose of good regulation is to support sustainable The great advantage of committing a government and equitable economic growth, poverty reduc- to a set of indicators is that by selecting an indica- tion and better governance. For a politician in tor or a set of measures a government makes con- office, however, regulatory quality often means cepts and principles operational. It provides a 3 WB204_RQI.indd 3 5/17/10 3:49:14 PM practical way to sort out the different claims or succeed not just because of how they were about quality. A commitment to a set of indica- appraised in the first place. More decisive are tors also reveals a government's preferences about variables such as the culture of compliance of the the purpose of regulation, and about regulatory firms, the strategies pursued by inspectors, the quality. administrative culture and the implementation of contingent and complimentary reforms, the There is a difference between quality of the regu- role of courts in enforcing regulations, and more latory tools employed by a government (such as generally the behaviour of the actors involved in regulatory impact assessment, consultation, sim- regulatory implementation. Regulation is an plification plans, codification exercises, plans to incomplete contract when it is designed. Indica- reduce administrative burdens, regulatory cost- tors of regulatory outcomes and impact have to ceilings for departments and agencies), quality of be designed taking into account the problems of the overall process to produce regulations through the causal chain that goes from regulatory tools different administrative requirements (adminis- to real-world regulations and their economic trative procedure acts, freedom of information impact. Further, even though the causal complex- acts, notice-and-comment, and obligation to ity has been tackled, there is still the issue of the respond to the received comments), and quality quality of data, an important aspect to consider of the regulations as experienced by firms and especially among the developing countries. With citizens. Most of the indicators available track these conceptual caveats in mind, we can turn to down the first dimension. To capture the other the discussion of the main datasets for regulatory dimensions is more difficult, since regulations fail quality. 4 WB204_RQI.indd 4 5/17/10 3:49:14 PM DATA SOURCES AND DATABASE ON REGULATORY QUALITY AND GOVERNANCE The aim of this section is to provide a schematic or inhibit competition. These economy-wide review of the existing databases on regulatory gov- indicators summarize a wide array of regulatory ernance and, more specifically, on better regula- provisions across the OECD countries (including tion tools, including regulatory impact assessment state control, barriers to entrepreneurship, inter- (RIA). We believe that this exercise is a necessary national trade, and investment). They have been precondition to discuss and propose comprehen- calculated for 1998 and 2003. Relying on empir- sive typologies of index/indicators of regulatory ical findings that a reduction of regulation is quality. Table 1 summarizes the purpose, the con- closely intertwined with an enhancement of com- tents, and the methodology of each database. In petition and productivity in product markets the remainder, we focus on the most relevant (Conway et al., 2003: 4), the aim of those indica- indicators of regulatory quality developed by the tors is to assess the relative friendliness of regula- World Bank and the OECD. We will group the tions to market mechanisms across countries and other datasets according to their purpose. Present- to show the interrelations between certain types ing a synthetic table (Table 3), the final part of of indicators and clusters of countries, covering this section presents the main findings according several regulatory areas (Conway et al., 2005). to various dimensions of measurement system. It There is an explicit orientation towards policy- gives also a comprehensive overview of available relevant recommendations and specific sugges- database (generated also by academic research) on tions for policy reform. regulatory governance and reform (Table 2). This is a database composed of objective indica- tors, covering both economic and administra- tive types of regulation. All these indicators can OECD Product Market Regulation be broken down into their constituent elements Product Market Regulation indicators measure (e.g., state control over business enterprises; the extent to which regulatory settings promote barriers to entrepreneurship; and barriers to 5 WB204_RQI.indd 5 5/17/10 3:49:15 PM WB204_RQI.indd 6 6 Table 1: Mapping Databases on Regulatory Quality and Governance Purpose and Level Dataset of Evaluation Contents Countries Years of Observation Methodology Webpage World Bank: Indicators Ranking Incidence of market- 209 countries 1996, 1998, 2000, Statistical analysis of http://info.worldbank. of governance and unfriendly policy and 2002, 2004 and business perception of org/governance/wgi/ Quality of regulation as institutional quality perception of regulatory 2006 regulatory quality index.asp perceived by stakehold- (regulatory quality) burdens ers World Bank: Doing Ranking Regulatory costs of 181 world wide Annual since 2004 The methodology varies www.doingbusiness.org Business indicators carrying out business countries according to the Quality of regulatory activities regulatory sector outcome under analysis OECD Product Market Country comparison Extent of regulation in OECD countries 1998, 2003 Weighted aggregation www.oecd.org/ Regulation and policy recommen- manufacturing and of self-assessment document/ dations several non- information about 36/0,2340,en_2825_ manufacturing sectors regulatory provisions 495698_35790244_ Extent of regulation 1_1_1_1,00.html OECD: Indicators of Descriptive ­ country Regulatory quality OECD countries and 1998, 2002, 2005 Self-assessed question- No website? regulatory management comparison (benchmark- management as defined selected non-members naire www.oecd.org/regref systems (2007) or ing) by the 2005 OECD indicators of government Principles for Regulatory Quality of administrative capacity to produce Quality and Perfor- procedure high-quality regulations mance. It includes (1998, 2000) performance/outcome indicators. World Bank: Regulatory Data and concept Independent regulatory 39 responses and six 2004 Survey of regulators and http://www.nera.com/ transparency (NERA creation authorities' transparency in-depth case-studies in-depth case studies PracticeArea.asp? Consulting) PA_ID=100&more= Quality of administrative ClientExp&c_ID=450 procedure Telecommunications and Data creation Telecommunication and 45 countries for 2003 Survey of regulators http://www.aei- electricity regulation electricity regulatory 178 variables brookings.org/ Quality of administrative database (AEI Brooking) authorities and their (telecommunication) 20 publications/abstract procedure scope, configuration, Countries for 374 .php?pid=724 organization, procedure variable (electricity) and decision-making process 5/17/10 3:49:16 PM WB204_RQI.indd 7 Canadian surveys of Evaluation of RIA Extent of impact and Canada 1999 Mixed (Survey of On file by the authors regulators ­ Regulatory quality of Canadian regulators and case- Consulting Group and government's regulatory studies) Delphi Group policy The Evaluation Evaluation of RIA Extent of institutionaliza- EU Commission 2007 Mixed (survey of http://ec.europa.eu/ Partnership tion, quality of analysis, regulators and case- governance/impact/ and balance of studies) key_en.htm economic, social, and environmental dimensions OECD Implementation Survey design Business perception OECD countries 1997 Postal questionnaire sent On file by the authors Guidelines of the recommendation and survey about regulatory to businesses in the Multi-Country Business data creation areas (employment, OECD countries. Survey: benchmarking environmental, taxation) Quality of regulation regulatory and is composed of the administrative business following sections: environment (1998) ­ Quality of regulations ­ Quality of administra- tion (enforcing regulations) ­ Administrative compliance costs of regulations. OECD: Regulatory Descriptive ­ country General context of the OECD countries 2002 Self-assessed question- http://www.oecd.org/ Performance: Ex Post comparison (benchmark- strategies/policies (to naire sent to the formal document/27/0,3343, Evaluation of Regulation ing) evaluate regulatory tools group of directors and en_2649_34141_ Policies (2003) and institutions) and experts for better 34238491_1_1_1 Quality of better details about evaluation regulation and RIA _1,00.html regulation and RIA programs OECD RIA Inventory Data collection Elements of RIA systems OECD Member States 2003 Desk research followed http://www.oecd.org/ (fact sheets) in OECD countries such up by a government dataoecd/22/9/ as: review 35258430.pdf Quality of RIA ­ Type of analysis ­ Scope of coverage ­ Public disclosure ­ Quality control ­ Cost-benefit analysis ­ Social discount rate ­ Risk assessment ­ Effects on competition and market openness 7 ­ Ex-post monitoring. (Continued ) 5/17/10 3:49:16 PM WB204_RQI.indd 8 8 Purpose and Level Dataset of Evaluation Contents Countries Years of Observation Methodology Webpage Report of the Greek Cross-country compari- The extent of institutional 13 old Member States. 2002 EU directors of better Presidency on the son and policy design and implementa- regulation self- implementation progress formulation tion of the EU MS better assessment questionnaire of the Action Plan of the regulation tools. Quality of RIA `Mandelkern Report on The items are around Better Regulation' ­ 30. Spring 2002 European Commission: Descriptive + scorecard State of the art of better 25 EU Member States Compilation of national http://ec.europa.eu/ Report on the implemen- on business start up & regulation policy. plus Norway reports enterprise/enterprise_ tation of the European RIA policy/charter/index_ Charter for Small en.htm Quality of RIA Enterprises Indicators of regulatory Descriptive overview of Overview of the extent, 20 EU member states 2004, 2006, 2007 Survey of EU directors of quality BR & RIA coherence and and Norway better regulation integration of the Quality of RIA different components of better regulation as a public policy EVIA Comparative analysis of Evaluation study on the 27 EU member states 2007 Mixed (desk research, http://web.fu-berlin.de/ RIA impact of RIA on and RIAs case studies survey, evaluative ffu/evia decision making (EU, UK, NL, Den, Pol) measurement Quality of RIA containing normative recommendations ENBR/DIADEM Data creation Cross-country compari- 25 countries and 2007 Comparative measure- www.enbr.org son of RIA systems 4 non-EU countries ments Quality of RIA through three levels of analysis: macro- institutional, RIA program level, and individual RIA level 5/17/10 3:49:16 PM WB204_RQI.indd 9 OECD Red Tape Cross-country compari- Country performance 11 OECD countries 2005­2007 Revised version of the http://www.oecd.org/ Assessment project son (benchmarking), and practice on Dutch Standard Cost document/3/0,3343, measurement methodol- selected administrative Model en_2649_34141_ ogy test (SCM), and procedures, permits, 34061123 policy recommendations and licensing (both in _1_1_1_1,00.html general and on specific Quality of simplification sectors ­ road freight), and concerning SME business activities The cost effectiveness Collection of data and List of publications Not applicable Literature review https://research. analysis register (Tufts standardization of containing cost-utility conducted according tufts-nemc.org/cear/ Medical Center, USA) cost-effectiveness analyses and the to a specific coding. default.aspx methodology population target. Quality of RIA The list includes also the ration $/QALY expressed in 2002$ and an assessment of the quality of the publication (value from 1 to 7). Commonwealth Business http://www.cbcglobal Council's Business .org/cbc_pages/ Envrionment Survey publications.aspx 9 5/17/10 3:49:17 PM international trade and investment Conway tape. Obviously, the data-collection relies on et al., 2005: 8). questionnaires that contain a measure of sub- jective discretion. The data sources are a self-assessment and multiple- choice questionnaire sent to OECD member states' governments and, in minimal part, external sources, World Bank: Indicators of governance i.e. documents (published and unpublished) by the OECD and other organizations. Where it was nec- and institutional quality essary, a follow-up validation of data has been con- This database is a set of indicators covering the ducted through conversation between the OECD quality of regulatory outcomes within an overall and national officials. governance framework. Based on numerous indexes of regulatory quality as perceived by To attain an overall indicator, all regulatory pro- stakeholders, the World Bank's indicators of gov- visions have been grouped according to specific ernance and institutional quality provide coun- areas. Survey-based and qualitative information tries with a tool for comparison and benchmark- has been transformed into a numerical format by ing. There are two broad types of institutional creating a scale from 0 (little market regulation) measures: performance measures that provide to 6 (a highly regulated market). A subjective assessment of the quality of governance and pro- weight has been assigned to each variable. The cess measures that describe the institutional sum of values of each regulatory provision of a input. Kaufmann et al. (2008; 2005; 2003) iden- specific regulatory area provides the first level of tify six dimensions of governance, i.e. voice and indicators. The system of indicators is designed accountability, policy stability, government effec- as "a pyramid with 16 low-level indicators at the tiveness, rule of law, control of corruption, and base and one overall indicator of product market regulatory quality. Correlations across the differ- regulation at the top" (Conway et al., 2005: 7). ent dimensions over the years are quite high. This methodology of data collection involves some subjective judgments in replying and inter- Figure 1 shows the correlation between govern- preting the responses, scoring qualitative informa- ment effectiveness (GE) and the other dimen- tion into ordinal indicators, and finally compiling sions (QR = regulatory quality, RL = rule of law; the data' (Nicoletti et al., 2000: 16­17). The anal- CC = control of corruption; PS = political stabil- ysis of product market regulation can be used to ity; VA = voice and accountability) for every year reveal the presence of different clusters or coun- of observation since 1996. tries, such as common law countries, continental European countries, and so on. The correlation between regulatory quality and government effectiveness has increased from an Reckoning that this dataset focuses on the initial p = .8 (in 1996) to p = .95 in 2005. Spe- extent of product market regulation as a narrow cifically on regulatory quality, this dimension of manifestation of quality (derived by the under- governance "includes measures of the incidence lying assumption that less regulation increase of market-unfriendly policies such as price con- the level of productivity and that in turn pro- trol or inadequate bank supervision, as well as duce more economic growth), this dataset can perception of the burdens imposed by excessive also be used to connect some measures to the regulation in areas such as foreign trade and busi- tools of better regulation. An example is the ness development" (Kaufmann et al., 2003: 3). indicator "barriers to entrepreneurship." This can be related to the results of simplification Kaufmann et al. (2003: 39) acknowledge that initiatives or to the programs to remove red their system of subjective indicators contains 10 WB204_RQI.indd 10 5/17/10 3:49:17 PM Figure 1: Correlation Between "Government Effectiveness" and Other WBI Governance Indicators 1 0.95 0.9 0.85 0.8 0.75 0.7 0.65 0.6 0.55 0.5 1996 1998 2000 2002 2003 2004 2005 GE*VA GE*PS GE*RQ GE*RL GE*CC Source: Damonte A. `Le sfide della governance europea', presentation at the URGE Conference on Governance multilivello e politica europea, Collegio Carlo Alberto, Moncalieri, Italy, 29 November 2006. Notes: n = 151 cases (cases with high number of missing observations were excluded) QR = regulatory quality, RL = rule of law; CC = control of corruption; PS = political stability; VA = voice and accountability) substantial margins of error, which should be In addition, ideological biases in the perceptions carefully taken into account in comparing coun- of experts have been addressed, investigated and tries. However, they argue that the same margin mostly discounted by Kaufmann et al. as irrele- of error exists in objective measures as well when vant. Standard errors of the different dimensions they are used to portray broader concepts, such as tend to decrease over time, showing that the regulatory quality or the efficiency of governance judgments on which the indicators are based tend (Kaufmann et al., 2003: 28). Their system of to become more homogeneous. The quality here indicators makes explicit that (and takes into is subjective and represented through improve- account) this margin of error is due to an unob- ments in the table of countries: the more a coun- served component. This is essentially done try has raised its ranking, the more its regulatory through the aggregation process, the so-called environment has been perceived as improved. unobserved components model that follows spe- cific steps - common to other indicators. The first step is to rescale the individual indicators from World Bank: Doing Business each underlying source in order to make them indicators comparable across data sources. The second step is to construct a weighted average of each of these The Doing Business (DB) database provides rescaled data sources for obtaining an aggregate objective measures of business regulations and indicator of governance. Preferring subjective their enforcement, comparable across 181 econo- measures that are more correlated with each other mies. The dataset is popular for the ranking of in order to depict a common but unobserved countries according to the number of procedures, phenomenon, the statistical model chosen by time and costs borne by a typical firm to carry out Kaufmann and his associates assigns the weights activities. These are other measures of the strict- on the basis of estimates of the precision of each ness of regulations assumed not associated to safe- data source. guarding consumers' choice but to policymakers' 11 WB204_RQI.indd 11 5/17/10 3:49:17 PM activities in creating and increasing the rents for There is a correlation between the measures of their constituents. regulation of labor and measures of entry. Coun- tries that tend to regulate labor markets strictly The accuracy of data collected through govern- also tend to regulate procedures for setting a new ment publication and online information is tested firm up and judicial proceedings relatively more by contacting government agencies and law firms. heavily than other countries (Botero et al., 2003: Time and official costs are usually aggregated into 26­9). a summary "full cost" measure, which in case of business start up "adds up the official expenses Although they are often praised for their capacity and an estimate of the value of the entrepreneur's to reach a high-political impact on policymakers time, valuing his time at the country's per capita and stakeholders, there have been criticisms of income per working day" (Djankov et al., 2001: the theoretical and methodological approach of 11). In employment protection law, the method- DB indicators. Indeed, the Independent Evalua- ology of aggregation is substantially different. The tion Group (IEG) at the World Bank has recently analysis covered three areas of regulation ­ published an evaluation report (IEG, 2008). The employment, collective bargaining and social report concludes that "the indicators have been security laws ­ in 85 countries. Each of these three highly effective in drawing attention to the bur- areas is divided into more specific fields. dens of business regulation, but cannot by them- selves capture other key dimensions of a country's Aggregation is multi-staged and similar to the business climate, the benefits of regulation, or OECD's indicators of Product Market Regula- key related aspects of development effectiveness. tion. First, all variables of a regulatory area are Thus, the Bank Group and stakeholders need to codified. Variables are assigned a value of one consider the DB indicators in a country context when regulation constrained the negotiation and interpret them accordingly" (IEG, 2008: xv). between employer and employee, and zero other- From a theoretical perspective the report stresses wise. For example, fixed-term contracts are allowed that the association between business regulatory in Venezuela only for temporary tasks, while in environment and macroeconomic outcomes has Vietnam they are allowed for any task. On this not been fully explored and explained because of component of the employment law index, Vene- the "[m]any other factors affecting macroeco- zuela scores 1, Vietnam 0. Second, the variables nomic outcomes, and the direction of causality comprising a regulatory area were aggregated into between regulation and economic outcome is sub-indexes. Finally, the three overall measures, very difficult to isolate" (IEG, 2008: xv). namely employment law, collective bargaining and social security law indexes, were derived from Moreover, the theoretical assumption behind the an arithmetic sum of the sub-indexes. This is an regulatory benchmark is that less regulation is example of a simple but effective aggregation in desirable. This assumption has essentially three which the weights are given by the presence of flaws. Firstly, other variables such as infrastruc- sub-components of comprehensive indexes. This ture, labor and skill impact on firms' investment is our recommended aggregation method if gov- decisions. Secondly, regulatory reforms aimed at ernment wishes to get more aggregate indicators reducing the quantity of regulations overlook the of regulatory quality.4 associated regulatory benefits. Thirdly, the actual regulatory costs do not derive from what is writ- ten on a piece of legislation but on how the legis- lation is observed, complied and enforced. We 4 For further information see the paper Project Level Indicators prepared for the Better Regulation for Growth (BRG) Pro- now turn to more specific indicators of the qual- gram, available at www.ifc.org/brg ity of regulatory reform and RIA programs. 12 WB204_RQI.indd 12 5/17/10 3:49:18 PM OECD: Indicators of government in implementing policy for better regulatory gov- capacity to produce high-quality ernance. Recently, the OECD has run a principal component analysis of the self-assessed indicators regulations to derive patterns and build clusters of countries. It is the pioneering OECD project on a database To control for consistency with other sets of indi- on RIA and regulatory reform. There are three cators, the OECD has performed a correlation versions of this database (1998, 2000, and 2005). analysis with Doing Business and other datasets Although explicit benchmarking is not unani- (OECD 2008). mously endorsed by OECD members, this is an obvious practical way to use these data. The OECD defined regulatory quality in the basis of Surveys of independent regulatory the following principles: necessary, efficiency, authorities and regulators effectiveness, transparency, and protection of public interest. The OECD approach to regula- The rationale for surveys of regulators is simple. It tory process is innovative not only for fixing prin- is difficult to establish a causal relationship between ciples of better regulation but also the way these regulatory reform tools and final economic out- principles have been used in a more comprehen- comes. But it is simpler to assume that if regula- sive transnational evaluation exercise. An over- tors make use of the tools for a certain number of view of good practice provides normative hints years, their attitude towards regulation, stakehold- for defining the principles that in turn are also ers, consultation, and overall governance will standards of checklists, used by the OECD in its change. These surveys are designed to capture how peer review of regulatory processes and gover- the attitudes of regulators compare to an "ideal" nance. attitude, assumed to be the best for supporting regulatory excellence. The causal chain between To assess each country's regulatory governance, a reform and change is shorter in this case. detailed 2005 self-assessment questionnaire was sent to the formal group of directors and experts We often hear the slogan that better regulation for better regulation and RIA programs. It was "makes institutions think." Surveys of regulators composed of five sections: inform us on whether and how beliefs and attitudes are changing. Having established their importance Content of regulatory policies at the conceptual level, when we turned to empiri- cal analysis we did not find many examples. Regulatory quality tools Two studies can be included in this category. In Institutional arrangements to promote regu- June 2005, NERA, a London based economic latory quality consultancy group, published the result of a sur- vey of regulators conducted for the World Bank Dynamic aspects of regulatory quality on the topic of transparency. In operational terms, transparency has been defined in a relatively broad Performance/outcome indicators manner, as the centerpiece of "good regulation" The last two sections are clearly a recent evolu- (NERA, 2005, 11). In turn the seven characteris- tion of the OECD benchmarking exercise. The tics of good regulation (clarity of roles and objec- database is composed of around 80 items. tives; autonomy; predictability; transparency of decisions; accountability; participation, and open Governments in this case are evaluated on the access to information) have been transformed in basis of the political attention and commitment specific concepts of regulatory transparency. 13 WB204_RQI.indd 13 5/17/10 3:49:18 PM The survey identifies "the tools that regulators ment capacity to produce high-quality regula- currently use to improve transparency, their tions (recently dubbed Indicators of regulatory views on how effective such tools are and on the management systems), the OECD has compiled barriers they face to improve transparency" different data summarized in the following (NERA, 2005: 20). Another survey of regulators report/documents: the OECD report on Ex Post was conducted by Wallsten and his associates Evaluation of Regulatory Policies (2003) and the with the aim of providing "new tools for study- RIA Inventory. At the EU level, the European ing network industry reform" through cross- Commission and the member states have been country or cross-industries empirical analysis active in collecting cross-countries data on the (Wallsten et al, 2004). This project produced better regulation policies. two downloadable datasets on telecommunica- tion and electricity regulatory sectors. The cover- Stemming from the Mandelkern report (drafted age of the database is relevant: the electricity by national experts on better regulation in 2001), database has 374 variables; 178 are the variables different presidencies of the EU Council prepared of the telecommunication database. two documents: the 2002 Report of the Greek Presidency on the implementation progress of the Finally, we found two examples of surveys of regu- Action Plan of the "Mandelkern Report on Bet- lators on their capacity to analyze future regulatory ter Regulation," and the Italian, Irish and Dutch effects, conducted for the Canadian government presidencies of the Council of the EU "Regula- and the European Commission. These studies will tory Impact Assessment in 10 Countries: State of be presented in the initial part of Section 4 on the the Art, Core Aspect, and Best Practices," a report experiences of governments with measurement prepared for the ad hoc group of experts on better systems. regulation in May 2004. The comparison of EU member states according to their progress on bet- ter regulation methods has been institutionalized in the EU charter for small and medium enter- Surveys of business prises (SME) and successively in the Lisbon This is an area where governments and interna- agenda. Both policies required national govern- tional organizations have been active with postal ments to annually report on the policy progress and internet-based surveys, especially in relation put in place toward enhancing the quality of the to the measurement of administrative obligations. business regulatory environment. We will comment on this later in Section 4, illustrating how governments produce and use Academics have produced information on EU surveys.5 member states' RIA programs (Radaelli and De Francesco 2007; Radaelli, De Francesco, Troeger 2008). The European Commission's 6th frame- work program has funded two projects on RIA: Country fact-sheets on regulatory Evaluating Integrated Impact Assessments (EVIA) reform, Better Regulation and RIA and the European Network of Better Regulation (ENBR). The first project produced 27 EU mem- This group is the most numerous. In addition to ber states, country fact-sheets and 22 case studies its review of member states' regulatory reform on the use of RIA in Poland, the UK, and the and the above-mentioned indicators of govern- European Commission. The project also included 5 Currently the European Commission is undertaking a major an internet-based survey on the use of economic program of 25% reduction of burdens arising out of EU analysis by RIA desk officers and stakeholders. legislation which includes online consultation of all stake- holders (http://ec.europa.eu/enterprise/admin-burdens- ENBR is currently setting up a database of RIA reduction/home_en.htm). (the so-called DIADEM) in 25 EU member 14 WB204_RQI.indd 14 5/17/10 3:49:18 PM states and the Commission as well as Israel, them in regulatory management processes (details Moldova, Serbia, and Ukraine. Individual RIAs below in Section 4). Academic studies are gener- have been classified and scored using a slightly ally comparative and generate composite indica- different version of the 22 indicators of regula- tors. As section 4 and Table 2 show, each aggregate tory quality design by Radaelli and De Francesco indicator has advantages and disadvantages. The (2007) for DG Enterprise of the European Com- ultimate and perfect indicator does not exist. mission. As Van de Walle (2008a, 2008b) points out, it is advisable to focus on specific projects and areas of reform, because of the limitations of comprehen- Datasets on RIAs6 and methodology sive measures of the performance of governments Indicators of the quality of RIAs are the micro and the public sector (Van de Walle 2008a; level in the assessment of regulatory quality. Such 2008b). Subjective and composite indicators are indicators gauge quality according to the simple useful because they focus on the real world out- assumption that systematic analysis of regulatory come as perceived by stakeholders, especially impacts would produce better regulatory out- businesses. But they may not be so informative comes. It is obvious that this assumption holds for governments in search of tracking regulatory only for regulations reviewed and cannot provide reform. Objective composite indicators are more any hint on the overall regulatory environment. informative, given their capacity to refer to spe- In this group we find the former AEI-Brookings cific regulatory areas and better regulation tools Joint Center with its database on U.S. RIAs, the but they cannot deliver an overarching evalua- already mentioned DIADEM segment of ENBR tion. However, the Doing Business indicators and the cost effectiveness analysis register. The cover many regulatory sectors affecting the life OECD Red Tape Assessment project is an attempt cycle of a business, and address this concern. to further develop the Standard Cost Model Only academics supported by international orga- (SCM), a methodology that emerged in the nizations are able to constantly update regulatory Netherlands to quantify administrative burdens governance and reform variables. for business in a more comparative setting. On the other hand, the majority of datasets pro- duces single measures generally generated by gov- ernment surveys and not integrated with the An overview business perception surveys (see Table 3). Few The purpose and content of the databases vary databases aim directly to improve the methodol- considerably, especially among international ogy of economic analysis, collecting RIAs and organizations. The World Bank has produced additional useful data. ranking indicators and new approaches to mea- suring complex concepts (the reference is to the Thus, an officer who is developing regulation on, NERA survey). The OECD has relied on a say, the disposal of refrigerators in country A can- descriptive/comparative approach, and, when not make use of the data and analysis produced allowed by the member states, benchmarking. A by regulators in countries B, C, D when they similar approach has been taken by the European faced the same regulatory problem. Since most Commission. In the EU, the main aim is to insti- regulatory problems are similar, and some regula- tutionalize measurement exercises and to embed tory solutions are made similar via harmoniza- tion, mutual recognition, WTO jurisprudence and other tools of regulatory rapprochement, this 6 For further discussion of RIA, and in particular its applica- tion in developing countries, see the RIA Light paper, part type of datasets would be useful. Diadem ­ the of the BRG program. main product of the European Network of Better 15 WB204_RQI.indd 15 5/17/10 3:49:18 PM WB204_RQI.indd 16 16 Table 2: Advantages and Disadvantages of Different Types of Measures Relation with Type of measure Scope Index Strengths Weakness dimensions Relation with tools Pryor (2002) Composite Across-the-board rules Laissez faire The concept of ­ Quality is related Real-world outcome No specific relation (general economic regulatory quality to only one (business's to any tool subjective regulation) is comprehensive stakeholder, i.e., perceptions) business ­ This relationship between regulation and quality is indirect as it is mediated by the index of govern- ment's effectiveness ­ The index is not informative for policymakers Kaufmann et al. Composite Across-the-board rules Index of regulatory ­ All stakeholders' ­ The index may not Real-world outcome No specific relation (2008) quality as a perceptions are be sufficiently to any tool subjective component of the taken into account informative for overall index of ­ Margins of error policymakers good governance and ideological ­ It may not achieve biases are taken the learning into account purpose at the ­ Possible bench- level of civil marking tool servants ­ Integration between regulatory quality and wider reforms of governance Koedijk and Kremers Composite Product market and Index of product ­ Regulatory quality ­ The analysis is Real-world outcome No specific relation (1996) employment market, employment, in EU context focused on the to any tool, except objective regulation and overall market ­ Regulatory quality extent of the indicator of regulation differs by cluster. regulation, not business establish- Hence the indexes on quality ment and shop hours are sensitive to ­ Quality of that can be context (worlds of statistical analysis associated to regulation) can be questioned simplification ­ Data out of date 5/17/10 3:49:18 PM WB204_RQI.indd 17 Knack and Kugler Composite Regulation of entry, Index of good ­ Methodological ­ The index does not Real-world outcome Simplification (2002) and legal framework governance, approach provide any useful objective regulation (contract regulation of entry, input on the enforcement) and contract understanding of enforcement regulatory quality Conway et al. Composite State control over Overall index of ­ Three-stage ­ Limited coverage Real-world outcome Simplification (2005) market, regulation of economic regulation, methodology leads of regulatory areas objective RIA entry, regulation of and employment to evaluation of ­ The indexes do not trade and interna- protection legislation tools and policy refer to macroeco- tional investment, ­ Flexible system of nomic performance and employment measurement that ­ Presence of legislation can be applied to subjective specific policies discretion (e.g. Innovation policy) Djankov et al. Composite Regulation of entry Index of regulation ­ Good data-collec- ­ Wide scope ­ Some limited Simplification (2001) of entry tion methodology analysis dimensions of objective ­ Methodology is ­ Narrow scope of real-world outcome easy to replicate quality (regulation ­ Activity ­ The index is of entry) precise and comprehensive ­ Clear relation to simplification Botero et al. (2003) Composite Employment Index of employment ­ Good data- ­ Wide scope Real-world outcome Simplification legislation legislation collection analysis objective RIA methodology ­ Analytic focus on ­ Statistical the extent of methodology is regulation easy to replicate ­ Ideological bias ("less regulation is always good regulation") Simple measures Objective or ­ Compliance costs ­ Easy to understand ­ Often difficult to Real-world outcome RIA subjective ­ Costs of administer- ­ Targeted gather with Simplification ing regulation ­ Lists of simple precision, ­ Cost effectiveness measures (in the accuracy, and of regulatory Yes/No format) possibility to program are very informa- replicate ­ Measures of tive on the quality ­ Silent on simple CBA quality of RIA and measures of access simplification and consultation 17 Source: Adapted from Radaelli and De Francesco (2007). 5/17/10 3:49:19 PM WB204_RQI.indd 18 18 Table 3: Purpose, Aim and Types of Measurement of Databases Integrated with larger Aimed at BR tools or Measure of gover- sets of data or regulatory reform Objective or nance or regulatory comprehensive Dataset Data utilization outcome survey-based Simple or composite environment assessments World Bank: Indicators of Ranking regulatory reform survey-based Composite Governance YES governance and institutional outcome quality (regulatory quality) World Bank: Doing Business Ranking Both Objective Composite Regulatory environment NO indicators OECD Product Market Cross-countries Both survey-based Composite Regulatory environment NO Regulation comparison and policy recommendation OECD: Indicators of Descriptive and BR tools survey-based Simple Governance NO regulatory management benchmarking systems (2007) or indicators of government capacity to produce high-quality regulations World Bank: Regulatory Data generation and regulatory reform survey-based Simple Governance NO transparency (NERA concept formation outcome Consulting) Telecommunications and Data generation regulatory reform survey-based Simple Governance NO electricity regulation database outcome (AEI Brooking) Canadian surveys Evaluation OECD Implementation Methodology regulatory reform survey-based Simple Regulatory environment NO Guidelines of the Multi-Country enhancement outcome Business Survey: benchmark- ing regulatory and administra- tive business environment (1998) 5/17/10 3:49:19 PM WB204_RQI.indd 19 OECD: Regulatory Perfor- Descriptive and BR tools survey-based Simple Governance NO mance: Ex Post Evaluation of benchmarking Regulation Policies (2003) OECD RIA Inventory Descriptive BR tools Objective Simple Governance NO Report of the Greek Presidency Descriptive BR tools survey-based Simple Governance NO on the implementation progress of the Action Plan of the `Mandelkern Report on Better Regulation' ­ Spring 2002 European Commission: Report Descriptive and policy BR tools survey-based Simple and composite Governance YES on the implementation of the orientation European Charter for Small Enterprises in the Member States of the European Union Indicators of regulatory quality Cross-country BR tools survey-based Simple Governance NO comparison EVIA Cross-country BR tools Both Simple Governance NO comparison ENBR/DIADEM Cross-country BR tools Objective Simple Governance NO comparison OECD Red Tape Assessment Methodology BR tools Objective Composite Regulatory environment NO project enhancement The cost effectiveness analysis Methodology Both Not applicable Simple Regulatory environment NO register (Tufts Medical Center, enhancement USA) 19 5/17/10 3:49:19 PM Regulation (see Section 2) is one of the few data- necessary to take into account many measures, sets developed with this specific aim. It is still cumulating different bits of information, and pos- under construction, but it definitively goes in the sibly correcting for the bias of one measure with right direction. another measures. Overall, the databases measure both the quality of Great attention should be paid to this process, since regulatory process and its institutional input, and wrong choices in the construction of the menu of the quality of regulatory environment as the real indicators may lead to adding the bias of one mea- world outcome of better regulation programs (see sure to the bias of other measures: two wrongs do Table 3). There is still room for improvement both not make one right. Further, there is a gap con- in term of quantity and quality in each category of cerning the integration of these measures with a indicators, especially in better understanding the broader measurement system. With the exclusion relationship between quality of governance and of Kaufmann's indicators and the EU Commission quality of the regulatory environment. But here measurement of narrow regulatory sectors and our point is that each indicator represents only a their integration in the Lisbon agenda for growth specific discernment and representation of quality. and jobs, the measures are isolated and not con- In order to have a broader appraisal of quality it is nected with broader and different measures. 20 WB204_RQI.indd 20 5/17/10 3:49:20 PM USE OF INDICATORS After having identified, mapped and classified making, identifying reasons for no or mini- the main databases, this section reviews how gov- mal impact of economic analysis and RIA on ernments, international organizations and aca- the decision making. demics use data and measures related to regula- tory quality and governance. c. Accuracy studies ­ These assess the quality of ex ante regulatory analysis by comparing pre- dicted versus actual impacts. How governments use existing Conduct studies require one of the following indicators measures: indicators, scorecards, process standards or checklists. Using the latter methodologies, sev- We review the measures and tools related to RIA eral independent audit offices have reported on programs in eight countries, namely, Australia, regulatory quality. Surveys are generally used to Belgium, Denmark, Canada, the Netherlands, conduct impact as well as accuracy retrospective Sweden, the UK, and the United States. Essen- studies. Ex post data are essential for setting a tially, measures and tools can be grouped in three sound (regulatory) budget system. With the types of retrospective studies (OECD 1999, 19): exception of few studies conducted in the United States (OTA, 1995 and the Pollution Abatement a. Conduct studies ­ By assessing whether and Control Expenditures calculated by the U.S. appropriate guidance or generally accepted Census Bureau7), accuracy studies are not com- practices were followed, these studies identify mon within governments. possible improvements in the methodology to carry out ex ante economic analysis. The following sections introduce and examine the tools used in conduct studies. b. Impact studies ­ These studies focus on the impact of regulatory quality tools on decision 7 www.census.gov/econ/overview/mu1100.html 21 WB204_RQI.indd 21 5/17/10 3:49:20 PM Set of indicators programs, its questions on regulatory programs are really detailed. Monitored by OMB, ques- There is only one case of use of a coherent set of tions "are written in a Yes/No format and require simple indicators of regulatory quality. In Australia, the user to provide a brief narrative explanation departments and agencies are required to compile of the answer, including any relevant evidence to every year a set of Regulatory Performance Indica- substantiate the answer." Responses should be tors (RPI). The Office of Small Business (OSB) has evidence based and not rely on impressions or developed nine RPIs, inspired by the six objectives generalities (OMB, 2002: 2). As a result, these of regulatory policy. The explicit link between what performance indicators are categorical and the country wants from the regulatory overview designed to be objective. They provide "a consis- system (the six objectives) and the indicators is an tent approach to rating federal programs," paying element of the strength of this approach. The Aus- particular attention to the individual program tralian indicators are rich in information, easy to results (GAO, 2004: 9). Indeed, the principle understand, and are monitored by the government. guiding the rating exercise is that regulatory pro- Indicators are therefore inserted organically into grams are appropriate and deserve funding when regulatory reform. However, a publication recently they show a clear evidence of effectiveness and concluded pessimistically that "the indicators were are capable of maximizing the benefits to society. of limited value, caused by the neglect of output and impact measures, the limited range of regula- Scoreboards are also common across stakehold- tion covered and a failure to integrate the indicators ers. The British Chambers of Commerce (BCC) into departmental performance management sys- (2004) has collected information on simple mea- tems" (Carroll, 2008: 2). sures of output, such as percentages of RIAs that present a summary of consultations; illustrate how proposed regulations have changed in Scorecards response to consultation; consider non-regulatory options; include sunset clauses; quantify costs, on Scorecards are tools widely used to assess the business, environment, citizens, and government; quality of RIA. Scorecards are composed of a quantify benefits, on business, environment, series of yes/no questions that generate simple citizens, and government; quantify net benefits; measures. A feature of these measures is that and report measures of compliance costs for they can be weighted and aggregated in an over- SME. Scorecards are also used outside govern- all composite indicator. This differentiates ments by academics (Cecot et al., 2007; Hahn et scorecards from the checklists. The latter are al., 2000; Hahn, Malik and Dudley, 2004; Lee usually a set of single measures that are not and Kirkpatrick, 2004) and think tanks (Vibert, aggregated. 2005; 2004; IEEP, 2004). The United States provides an interesting exam- ple of a scorecard aimed at ex post evaluation. Regulatory process standards and The Program Assessment Rating Tool (PART) is checklists "a diagnostic tool meant to provide a consistent approach to evaluating federal programs as part In Canada, the oversight body, currently the Trea- of the executive budget formulation process" sury Board Secretariat (TBS, 1996) is in charge of (GAO, 2004: 2-3). It systematizes the Office of assuring compliance with the federal regulatory Management and Budget (OMB) performance process management standards. These standards measurement. Although PART is a tool to evalu- provide checklists for each stage of the regulatory ate broad government programs that could be process. In particular, the quality assurance process spending, regulatory, or a mix of both types of terminates with regular internal self-assessments of 22 WB204_RQI.indd 22 5/17/10 3:49:22 PM performance and policy compliance. Canada has accompanied by semi-structured interviews recently introduced a formal model based on a within the departments and at the Better Regu- set of questions covering broad types of impacts lation Executive. In 2008, NAO has also assisted across the federal government. Regulators have the House of Commons Inquiries on better reg- to answer these questions at an early stage. TBS ulation in the UK and on the burdens reduction analysts also go through the questionnaire and program, sharing background data and provid- seek agreement between the department and ing state of the arts papers to parliamentary TBS. The agency and TBS sign off on triage clerks (House of Common Regulatory Reform jointly. Committee, 2008). Sweden and the Netherlands have for several In Canada, the auditor general (2000) has formu- years used checklists to exercise oversight on the lated recommendations to enhance the regulatory quality of RIAs produced by department and to quality of health and safety regulations, without monitor compliance with regulatory process proposing any indicator to measure it, however. standards. The OECD has discussed in its Work- Although Canada and other countries do not use ing Parties and proposed for adoption to the external review to propose indicators, these exer- member states a checklist drawn from the OECD cises provide standards and checklists. Typically, standards of regulatory quality. these are recommendations for changing guidance, improving training and rising the quality of resources dedicated to regulatory quality programs. Review by auditing bodies, internal watchdogs, and external In 2006, the European Commission created a body ­ the Impact Assessment Board ­ reporting evaluations directly to President Jose Manuel Barroso. The The review of RIA programs and burden reduc- Board includes the deputy secretary general of tion initiatives by courts of auditors is a common the Commission and four directors from the practice. The main aim of these studies is to under- Directorates General for Economic and Financial stand to what extent administrative requirements Affairs, Enterprise and Industry, Employment applied on the rule/law making process have an and Social Affairs, and Environment. This body impact on the way decisions are made. In the reviews the quality of the RIAs produced by the United States, using different methodologies, the Commission. It can make negative comments Government Accountability Office (GAO) pro- and ask for more analysis. The IAB is chaired by duces on several occasions reports on the effective- the Deputy Secretary General of the European ness of regulation and RIA in the context of the Commission ­ thus making a link between the general auditing of the executive's activities (see for mechanisms of control that flow through the Sec instance, the 2000 report on the analytical require- Gen and the IAB (http://ec.europa.eu/gove ments in rulemaking process GAO, 2000). rnance/impact/iab_en.htm). The IAB perfor- mance has been recently assessed in-house with In the UK, the role of the National Audit Office some descriptive indicators in the year-one report (NAO) in the quality assurance process has on the Board, produced by the Commission itself become more important since 2001. In that (Impact Assessment Board, 2008). year, the NAO (2001) produced a report on good RIA practice, drawing some lessons from a The European Commission has also externally review of a sample of 23 RIAs. NAO has also evaluated its RIA program. This study was carried evaluated the quality of guidance for RIA. One out by the Evaluation Partnership, a London-based interesting feature of the work conducted by consulting company. The approach of the evalua- NAO is that desk analysis of specific RIAs is tion was three-fold, covering the institutional 23 WB204_RQI.indd 23 5/17/10 3:49:22 PM design and the extent of implementation (process model is based on a random sample of 1000 and methodology, support structure, and central firms. These are then surveyed on how the stock overview), output, and impact on the policy pro- of regulation affects the daily administration of a cess. The methodology involved analysis at the IA company, providing a quantification of the total process and single IA levels. To obtain information administrative burdens. It considers the represen- on the IA process, the Evaluation Partnership tative sample according to business sectors and conducted surveys both with stakeholders and the size of the firm. In Belgium business surveys regulators as well as desk research on documents, capture the administrative burdens created by complemented with specific request of additional existing regulations. Recently, in Europe most information on guidelines, policy document, and EU countries and Norway have adopted the stan- external reports. The analysis on the single IAs dard cost-model to measure administrative bur- was conducted in three stages. First, 150 IAs were dens. The model is often based on interviews car- identified, selected, and screened. Second, a more ried out by consultants with a few typical firms depth analysis was conducted on 20 IA. This led per sector ­ hence it is not based on surveys of to the final stage of identification of 6 case studies representative samples. In some cases, however, analysis, supported by expert interviews in order surveys of business have been used to answer the to understand the magnitude of the impact of IA question whether firms feel that there has been a on the policy-making process. reduction in administrative burdens or not (NAO, 2007). Surveys The accuracy of survey results depends on the intrinsic quality of the questionnaire and on the Governments use two types of surveys: business presence of mechanisms that validate the infor- surveys on regulatory costs and surveys of regula- mation provided by businesses. However, ex ante tors on the impact of RIA programs. The United survey methods can have disadvantages too. First, States provides an example of business survey that they often rely on subjective data stemming from measures the cumulative impact of environmental a typical regulated party: the business. Businesses regulations, that is, the Pollution Abatement Costs tend to over-estimate regulatory costs and, ex and Expenditures (PACE) Survey.8 The survey ante, there will be limited mechanisms to vali- collects data on pollution abatement and preven- date their claims. Hence regulatory bias can tion, capital expenditures and operating costs for affect the outcomes of surveys, especially when air, water, solid waste, and multimedia. The sur- this bias has the potential to be large due to pre- vey also collects data on disposal, recycling, site dicted outcomes being assessed rather than cleanup, habitat protection, environmental moni- observed outcomes. In the absence of ex post toring, and testing and administrative costs as well assessment, predicted outcomes cannot be vali- as other payments, such as permits, fees, fines, dated against observed outcomes to allow for penalties, and tradable permits bought or sold. correction in the bias. Second, experience shows that all too often, survey data are not compared The Danish business test panels revolve around a with (and connected to) other research or sources collection of business administrative cost estima- that use different methodologies (SQW, 2005). tions in each regulatory sector, the aim of which This would be useful since it would take account is to select a regulatory sector in which simplifica- of different sources of bias. tion can yield the most efficient results. The selec- tion relies on information gathered by business Turning to surveys of regulators, in Canada, surveys, the so-called model companies. This interviews of a sample of regulators have mea- sured the change in their perception of the use- 8 http://www.census.gov/mcd/pace.htm fulness of RIA (in the context of a systematic 24 WB204_RQI.indd 24 5/17/10 3:49:23 PM external evaluation of RIA, Regulatory Consult- annual regulatory costs and benefits - in aggregate, ing Group Inc. and Delphi Group, 2000: 48-50). by agency and program, and by major rule. Sev- Canada shows how structured interviews can be eral indicators are considered. One is the sum of usefully employed to look inside government costs and benefits of regulations in a year calcu- rather than outside. The reliability of these indi- lated via RIA. The report adds up the results of the cators increases with the repetition of surveys last 10 years to provide an estimate of the total across time. impact of regulations over the decade. Another indicator is the overall impact of regulation on the public sector. Regulatory budgets, administrative The OMB indicators, however, are simply a burdens, and agendas compilation of RIAs. The measures of costs and The regulatory budget is a document containing benefits derived from RIAs cannot be treated as ceilings or targets of regulatory costs to be met by measures of the actual impact of regulation. RIAs the various regulatory programs or agency depart- are prepared before a regulation enters the stat- ments. It is a tool that monitors the growth of ute book and are enforced. Consequently, the regulatory costs. The regulatory budget generally costs and benefits included in an RIA are esti- takes the form of a report to the Parliament, thus mates. Actual costs and benefits can only be enhancing accountability. measured once the regulation is enforced. The literature on the accuracy of costs estimates sug- The international experience shows two different gests that the difference between a cost estimated approaches to regulatory budgets. The holistic in an RIA and the actual cost (as measured once approach accounts for all regulatory costs and the rule object of the RIA is implemented) can compares them to the regulatory benefits. The differ markedly (Harrington et al., 2000). As minimalist approach, instead, focuses on a single Parker (2003: 1367) argues, confusing ex ante component of the total regulatory costs, such as estimations with the actual regulatory costs and compliance costs or administrative burdens. benefits is a mistake similar to confusing a pre- game guess with the actual score of the game (see The United States follows the former; several also McGarity and Ruttenberg, 2002). European countries (i.e., the Netherlands, Belgium, Sweden, the UK, Denmark,) the latter, estimating The UK is the only country in Europe that has the total cost of administrative burdens. However, gone as far as to propose a cap on the annual cost in the U.S. a full-blown cross-government regula- (for the private sector) of new regulations within tory budget has never been adopted. However, the the context of a three or five-year budget structure Office for Information and Regulatory Affairs (BERR, 2008a, 2008b). The British initiative has (OIRA) ­ within the White House's Office for Prime Ministerial backing ­ operationally it is Management and Budget (OMB)9 ­ is required to carried out by the Department for Business, assess the benefits and costs of existing federal regu- Enterprise and Regulatory Reform (BERR10). latory programs and to recommend specific regula- The proposed budget works vertically (by depart- tions for reform or elimination. Congress annually ment) but issues such as energy efficiency will be receives a report issued by OMB-OIRA on the dealt with horizontally. The Conservatives have results of RIA alongside the estimate of the total pledged since 2005 that, if in government, they will impose an annual limit on the regulatory 9 The Office for Management and Budget's predominant mis- costs imposed by each department ­ and that sion is to assist the president of the United States in oversee- ing and coordinating the preparation of the federal budget, the administration's procurement and financial management, 10 Renamed Department for Business, Innovation and Skills and information and regulatory policies. (BIS) in 2009. 25 WB204_RQI.indd 25 5/17/10 3:49:23 PM "this limit will be reduced annually until over costs per administrative action are quantified. regulation is tackled" (Conservative Party, Action They are composed of three variables: the single on Deregulation, 2005, p.1). Thus, this trend is cost of action, the number of times a firm has to pretty much bipartisan. Several EU countries and provide specific information in a year (the so- the European Commission have multi-annual called "frequency"), and the number of businesses targets of reduction of administrative burdens. affected by regulation. The single cost of an action The Netherlands has completed a round of is given by the hourly labor cost and the time 25 percent reduction and is now embarking upon spent dealing with a specific information obliga- a further 25 percent reduction. tion per regulation. Time can be estimated through business surveys or alternatively through In Europe, as mentioned several countries follow a simulation of an administrative action taken by a sort of regulatory budget for administrative a model firm for complying with a regulatory costs approach. The approach is based on the provision (Legislative Burden Department, 2003: notion of administrative burdens. Interestingly, 21-22). several EU countries and the EU have established the same policy goal that is to reduce them by The experience of European countries with SCM 25 percent in a specific period of time. The strat- provides an interesting approach to total mea- egy to reduce administrative burdens targets the sures of administrative burdens. The Dutch pro- stock of existing regulations, as well as the flow of gram of 25 percent reduction of administrative new regulations. In both cases European govern- burdens has been positively evaluated by the ments use a specific tool, the so-called "standard World Bank and the OECD in 2007 (World cost model" (SCM). This accounting system has Bank Group 2007; OECD 2007). In terms of been created in the Netherlands and then has communicability, measures such as the total been adapted to the other European countries. administrative burden created by, say, transport regulation score well and are easy to link to tar- The Dutch SCM quantifies administrative costs gets. However, the standard cost model has some and divides them according to departments and limitations. the origin of regulation (EU directive, EU regu- lation, and national regulation). In the case of ex First, it is based on unrealistic assumptions about ante assessment, the SCM quantifies administra- a total compliance with administrative require- tive costs per regulatory option, showing the ments. Second, point estimates are often less lowest burden achievable. The measurement of informative than probabilistic ranges. Third, and the existing regulations is based on "zero-base- more importantly, administrative burdens are only measurement," an inventory of all information a component of direct costs. The other important obligations and administrative activities grouped component is compliance cost. The risk is one of according to the responsible ministry. The SCM focusing the public debate on a limited type of allows the Dutch government to encourage regulatory costs. Fourth, there is the political risk departments to respect ceilings of administrative of tilting regulatory reform towards only one burdens. For instance, in the Netherlands the stakeholder, that is, the firm. Interestingly, some ceilings "are being created for all departments as European countries and the European Commis- a fixed component of the budget and account- sion have manifested the intention to take care of ability system" (ACTAL, 2003: 7). burdens affecting both firms and citizens. The methodology to quantify administrative We conclude with the link between budgets and burdens involved in a single regulation is com- regulatory agendas. Holistic budgets, inclusive of mon to all countries that have adopted this benefits, could be used to support an open annual approach. There are two steps. In the first step, parliamentary debate on the Regulatory Agenda 26 WB204_RQI.indd 26 5/17/10 3:49:23 PM of the government (Doern 2007). Interestingly, turned into systems of indicators (a set of mea- Doern shows how Canada already has embryonic sures classified for their purpose, data source and processes and latent sources of information that methodology, as opposed to the use of a single could be brought together to produce an annual target) are not aware of this potential, or, if regulatory agenda. The advantage of a regulatory aware, they have not started activity along these agenda is that one can see synoptically all major lines. There is no evidence of the use of aggregate regulatory priorities, their costs and their benefits. measures. Even the simplest among the indexes, The presence of all major regulatory choices in a time and costs to comply with regulation of single document ­ Doern reasons ­ would facili- entry (Djankov et al. 2001), is not currently used tate the selection of major options, the allocation (with the only exception of Belgium). In other of resources and the solution to trade-offs between words, among Western governments, there is equally desirable goals. evidence of lack of systematic use of database reviewed in the previous section. Nevertheless, More recently, the Australian government has countries that focused on the reduction of admin- launched a performance benchmarking exercise on istrative burdens rely on simple measures amply business regulation across jurisdictions. The main discussed by the literature, such as the time spent idea supporting this measurement activity is that to deal with information requests, and hourly by benchmarking compliance costs one can iden- costs. Considering that most EU Member States tify unnecessary regulations and complement other are targeting red tape and that the literature has regulatory reform initiatives. Compliance costs discussed several indexes, there is considerable have been categorized in three groups associated to potential for a closer dialogue between policy- business activities: becoming and being a business; makers and social scientists. doing business; and doing business interstate. The project is not limited to indicators for compliance costs. It covers over time both quantity and quality of regulation. As a result, regulatory stock indica- How international organizations use tors are complemented and integrated in a broader indicators measurement framework to capture the progress and performance as well as regulatory design The World Bank, OECD and EU have designed administrative and enforcement. several sets and typologies of regulatory indicators. Their extent and purposes of utilization vary An overview of the number of countries with widely. The World Bank uses regulatory indicators measurement initiatives is increasing. The OECD for stimulating momentum for regulatory reform (2003: 10) lists nine member states with some through the collation of comparative statistics and experiences in evaluation of RIA output. More- reports, the appraisal of proposed regulatory inno- over, in 2004 a specific OECD questionnaire on vation projects in specific countries, and, together "ex post evaluation of regulatory tools and insti- with other indicators, regulatory indicators con- tutions" was sent to the national directors or tribute to major decisions on aid. experts of better regulation. The results of this self-assessed questionnaire are interesting: 8 out The OECD has typically made use of indicators of 22 OECD member states have an explicit and to organize discussions within its regulatory current policy/strategy on ex post evaluation of reform working parties. A range of OECD indi- regulatory tools and institutions; 3 out of 22 cators are collected by the member states via report that this policy/strategy is in preparation. questionnaire circulated within these working parties, so in a sense the national delegates sitting But one striking fact is that even European coun- on these committees are both the producers and tries with information that can be potentially the users of measures of regulatory quality. 27 WB204_RQI.indd 27 5/17/10 3:49:23 PM Observing the difficulty of assessing the real- and to start gathering simple measures. This was world impact of regulatory policies, the OECD in the context of the Commission (2005) trying in a recent study has further strengthened its to implement the Communication on Better practical approach to indicators (Jacobzone et al., Regulation for Growth and Jobs in the EU, which 2007) by investing recourses on indicators that contains the following wording on indicators: capture whether countries comply with the implementation of good practice as identified by "The Commission intends to discuss in this the OECD principles of good regulation. For the group the development of a coherent set of OECD, monitoring compliance with ministe- common indicators to monitor progress as rial-level OECD decisions on better regulation is regards the quality of the regulatory environ- an important way to make use of indicators ment both at EU level and in the Member (OECD, 1995). States themselves, as a basis for benchmark- ing. The Commission will encourage Mem- Discussions in OECD committees vary from the ber States to adopt such indicators to define analysis of general reform trends to more specific targets and priorities for their better regula- peer-review sessions on the reform results tion programs for the coming years in their achieved by individual countries. The OECD is national Lisbon programs" (European Com- undertaking a review of the old-15 EU member mission 2005, p. 10). states, with funding and intellectual support from the European Commission. Indicators have So far, however, the discussion has not progressed been discussed as a possible component of a steadily. The main feeling is that some govern- review process. But like in other cases individual ments are "reluctant although not opposed" to countries may object to the publication of indi- make the step of adopting common indicators cators. The general rule is that it is up to the (Radaelli, 2007). The same can be said of the members of the working parties to decide whether Council formations, where the Competitiveness indicators are published in detail ­ several OECD Council and the European Policy Committee reports on regulation report percentages but do have debated regulatory indicators on some occa- not list how individual countries fare. sions, but without taking the step of adopting a common set of measures to keep track of progress In the EU, the policy discussion on using regula- with regulatory reform. tory indicators started with the IRQ study funded by DGENTR (see Section 2 above). The study To date, the most important achievements in the was then presented by the Commission at a pub- EU context are the inclusion of measures of startup lic conference, at meetings of the High Level of firms (an indicator linked to the top priority of Group Advising the Commission on Impact the Lisbon agenda for growth and jobs in Europe) Assessment, and at a meeting of the Directors of and the adoption of targets for the reduction of Better Regulation of the EU hosted by the UK administrative burdens at the EU level and in the cabinet office in London, March 2005. The High individual member states (Council conclusions of Level Group is composed of delegates from the 8­9 March 2007). The Council agreed the follow- member states. Its aim is to advise the Commis- ing, with some caution on the commitment of the sion on strategic choices about impact assessment, individual countries. Note the caveats about the and to keep momentum for implementation of "starting points" and the "traditions": impact assessment at the EU level and in the member states. The Commission has sent discus- The European Council therefore agrees that sion papers on indicators to this body. The High administrative burdens arising from EU leg- Level Group was also asked to examine the feasi- islation should be reduced by 25 % by 2012. bility of collecting indicators country by country, Taking into account the different starting 28 WB204_RQI.indd 28 5/17/10 3:49:23 PM points and traditions the European Council How academics use indicators invites Member States to set their own national targets of comparable ambition This section reviews the academic discussion on within their spheres of competence by 2008. quality of regulation and governance. Based on Radaelli and De Francesco (2007), a synthetic Political hesitation notwithstanding, statements table (2) is presented to the state of the art of the about wanting to use indicators in EU regulatory methodological and theoretical insights of each policy are plentiful. The Four Presidencies Initia- indicator. Given space limitations and the fact tive of January 2004 declared that "the Commis- that most of the table entries have been discussed sion should propose indicators to measure progress in Section 3, we do not comment on the details with regulatory quality and reform at European of the studies, but only make references to some and Member State level for activation through the important points and classic approaches. Open Method of Coordination and for applica- tion in impact assessments." The presidencies also Knack and Kugler (2002) make the important stated that "[p]rogress can best be measured if such observation that "different indicators of good indicators are quantitative." The results of a 2007 governance are appropriate for different pur- survey held among the Directors of Better Regula- poses" (Knack and Kugler, 2002: 1). According tion in the context of the Intune research project to these authors, there are two main dimensions (www.intune.it) also suggests that, apart from the that motivate different indicators. One is con- official rhetoric, the adoption of indicators is not ventional and refers to the degree of aggregation. felt as a priority in this policy community. Only Indeed, most of the literature on indicators of 20 percent of the respondents chose regulatory governance or regulatory quality deals with dif- indicators as one of the three major items that ferent approaches to aggregation, with the should feature on the future better regulation implicit (but often flawed) notion that the more agenda. one aggregates, the better is the measure of qual- ity. Knack and Kugler warn that the choice of Yet it is the use of indicators rather than their indicators should account for the trade-off design, which is pivotal to their success as aid in between the precision of measurement (of a phe- policy processes. First, indicators are more likely nomenon) and the robustness of more highly to be used if they capture variables on which pol- aggregated indicators. The other dimension is the icymakers can intervene ­ a large number of vari- degree of transparency and the extent to which ables affecting the business and social environ- indicators are replicable. ment are not controllable by governments. Second, the types of indicators likely to win Koedijk and Kremers (1996) provide an often agreement among all political actors are also likely cited (although out of date) approach to compar- to be the ones that are amenable to various uses ative analysis of regulation and its effects on mac- and interpretations. Hence the usage of indica- roeconomic outcomes. The study was based on a tors in international organizations is about strik- regression analysis between the determinants of ing the balance between monitoring progress and macroeconomic performance (growth, produc- compliance from the center, and enabling indi- tivity and employment) and indexes of overall vidual countries to learn at their own pace and in market regulation, product market regulation relation to their regulatory reform objectives. and labor market regulation. It is important to Given this tradeoff, a major function of indica- note that this analysis dealt only with regulation tors is to provide transparency to regulatory and ignored other factors that can enhance the reform processes and to push governments to structural growth of an economy. Regression reveal their preferences by adopting specific indi- analysis was followed by cluster analysis, identify- cators (or by not adopting other indicators). ing different groups of countries, according to 29 WB204_RQI.indd 29 5/17/10 3:49:25 PM their economic performance and regulatory styles. to capture the real-world quality of regulation in This type of analysis shows that there are different terms of economic growth. In order to do so, worlds of regulatory reform. This is a useful anti- economists rely on theoretical assumptions or dote to the one-size-fits-all approach to policy empirical evidences that reveal the causal links recommendations. However, this approach has between regulation (usually less regulation) and limitations too. The focus is not on regulatory better economic performance. quality, but the quantity of regulation. The qual- ity of statistical analysis, based on crude regres- Overall, governments, academics, and interna- sions, is another weak point. To illustrate, the tional organizations have different approaches in authors make a simple average of the indexes to using knowledge stemming from indicators of construct the main composite measures. Although regulatory quality. Although using different this avoids the problem of arbitrary weights, there modalities of studies, governments are dedicated is no justification of why all the indexes should to evaluating their efforts in governing regula- weigh the same. Finally, their data are a bit out of tion. Accordingly, the assessment is essentially (in date. Regulatory reforms have changed the land- those few existing cases) an evaluation of the bet- scape of regulation in several EU countries. ter regulation policy and simplification or RIA programs. We are not aware of any evaluative A recent paper by Jalilian et al. (2006) deals with activities among developing countries. the impact of regulation on developing countries' economy. The study measures through an econo- Apart from few studies commissioned by the metric model how variations of regulatory quality Dutch government on the impact of reducing affect economic performance. The authors rely on administrative burdens on GDP, we are not also two dimensions of Kaufmann et al: regulatory aware of any attempt among the Western and quality and government effectiveness. Theoreti- OECD countries to measure the ultimate eco- cally, they rely on the concept of "growth account- nomic impact of better regulation policy on com- ing framework, where economic growth is used as petition, productivity, and ultimately on GDP the measure of economic performance and regula- and economic growth. This gap has been filled by tion is entered as an input in the [Cobb-Douglas] academics and economists, and international production function" (Jalilian et al, 2006: 91). Jal- organizations have been successful in filling this ilian and his associates recognize two caveats in gap in knowledge utilization promoting more their research design: the inability "to rigorously macro level measurement across governments. In demonstrate causation" and to solve the problem our proposal of regulatory quality indicators sys- of endogeneity due to the lack of data. The tem, we suggest that governments systematically authors, however, are confident in concluding utilize a series of economic outcome indicators that the quality of regulation positively affects that are now available at the World Bank to com- economic growth of developing countries. plement their absolutely necessary (micro) evalu- ation of better regulation programs. As already mentioned, the table shows that aca- demics rely on composite indicators that attempt 30 WB204_RQI.indd 30 5/17/10 3:49:25 PM CONCLUSIONS, CRITICAL REMARKS AND SUGGESTIONS This working paper has discussed different data- numbers of regulations administered by a sets of regulatory indicators and their usage. In department, are often not available, even in the conclusions, we sum up the main findings, the most developed countries (cfr. Jacobzone, make critical remarks on the state of the art, and et al. 2007). It takes time and political deter- formulate suggestions that link this paper to the mination ­ typically direct pressure from the approach chosen for the companion working Prime Minister - to convince ministers to paper. produce a list of their regulations. Projects aimed at the development of indicators in At the outset of this conclusion, in paragraph (a), these countries should usefully start from an we address the relationship between the literature analysis of what is available on the ground. on regulatory indicators and the specific socio- For example, in the case of a large continent economic conditions of the countries targeted by like Africa, the first step would be the identi- IC and the World Bank Group in general. We fication of 4-5 countries for which there is then move on to conclusions on the process of prima facie evidence that they have better aggregation, and formulate suggestions on knowl- quality information than the average coun- edge and policymaking by looking at a continuum try. Only by learning what is available in the of possible utilizations of regulatory indicators. first place can one then make suggestions as We consider three ideal-types of utilization: over- to how the data should be produced and sight, monitoring, and learning. gathered. In a sense, this is the approach fol- lowed by Doern (2007) and Radaelli and De a. When we deal with developing countries, we Francesco (2007) in a different context with have to be aware of the problems of data- the same analytical hurdles. Both studies look gathering, accuracy of measures, and limited at developed countries, but they start from a administrative capacity to collect informa- bottom-up approach, identifying measures tion. Apparently simple data, such as the and information that are already available 31 WB204_RQI.indd 31 5/17/10 3:49:28 PM (albeit not exploited yet) and latent processes are not designed at the same time. As shown through which measures could be used by by the example of the European Commission policymakers. and its campaigns to convince the EU mem- ber states to adopt indicators, the crucial step b. There are several datasets available, most based is to create political commitment to adopt a on simple measures rather than composite set of indicators, and to spell out clearly how indexes. International organizations have gen- they will be used. erated several regulatory indicators, and have tried to link measures both to their internal g. In this context, aggregation is most usefully decision-making processes and to processes of performed by those who are to make use of regulatory convergence across countries. The information and knowledge to take regulatory results are variable, with more progress on reform decisions. By adopting a set of indica- technical issues and less on convergence. tors out of the large number of measures theo- retically available, policy actors already make c. The integration of regulatory quality measures an important choice. By saying how and how with other components of structural reform often indicators will be used, they show com- and governance is generally low. Exceptions mitment to reform. By deciding on how a are the regulatory quality indicators of the particular aggregation enables them to track World Bank ­ integrated with other dimen- down a dimension of reform, they reveal their sions of good governance ­ and the integra- preferences and choose between one reform tion of administrative burdens indicators and trajectory and another. Thus, selection and targets with the EU Lisbon agenda for growth aggregation are mechanisms that reveal pref- and jobs. Statistical analysis shows that regula- erences. Imagine that a government has the tory reform is correlated with other dimen- following main goals of regulatory reform: sions of governance ­ suggesting that this form institutionalization, economic sophistication, of integration is worth-pursuing. accountability, interaction with the stakehold- d. Surveys of policymakers are few. We have not ers and communication. Aggregation should found systematic attempts to measure how then reflect these four dimensions, not in rela- regulatory reform is changing the attitudes tion to abstract technical properties. Thus, a towards regulation of regulators and policy set of N indicators could be broken down in officers who develop legislation. Yet if one four indexes providing information on the fundamental aspect of regulatory reform is to main goals, and suggesting where resources change attitudes and approaches to regula- have to be concentrated. tion, this dimension cannot be ignored. The state of play is better in the domain of busi- h. This brings us to a final critical point about ness surveys. usages of indicators. There are three catego- ries of usage: by the donor to evaluate a proj- e. It seems that most of the progress so far has ect ex-ante and during the implementation; been on the middle segment of the process ­ by a government to appraise reform and its that is, the technical discussion of indicators. outcomes; by constellations of governments The early steps (i.e., quality of data used to and stakeholders to promote convergence in calculate indicators) and knowledge utiliza- multi-lateral and supra-national organiza- tion have received less attention. tions. Further, the specific usages by each of these categories, but most typically govern- f. However, there is no point in designing indi- ments, are situated on a continuum that goes cators if processes of knowledge utilization from oversight and monitoring to learning. 32 WB204_RQI.indd 32 5/17/10 3:49:29 PM At the oversight pole, we find the need to j. There is no obvious choice between monitor- monitor progress in relation to pre-established ing and learning. Learning can foster conver- goals and targets. This is essentially a top- gence, as shown by studies of diffusion, under down usage of indicators. It secures conver- certain conditions. Strict monitoring and gence, but it may hinder processes of learning oversight are the most intuitive ways to seek in a flexible and more experimental ways. It convergence or at least identify the informa- also makes the assumption that there is only tion needed to foster convergence. Both work one shared set of regulatory reform goals ­ well when there is a shared set of beliefs on and this is not even true in highly integrated regulatory reform ­ which brings us back to political systems like the EU. There is a form the problem of getting the processes of knowl- of learning here, but it is about learning how edge utilization right. But the two mecha- to comply with targets and not lose position nisms are intrinsically different. Monitoring in the common march towards shared goals. is based on predictability, reporting, tasks well specified in advance, whereas learning is i. At the other extreme, we find experimental essentially disruptive of regularity and may learning. Here, different countries select lead to the breakdown of "monitorability" different sets of indicators or aggregate dif- (Sabel 1994). There are also several options ferently to reflect their own regulatory in-between, though. The continuum goes reform priorities and to experiment with from oversight to learning via traffic-light innovations in governance. Countries learn systems, competitive benchmarking, com- at they own pace and in relation to their parative benchmarking, naming and sham- objectives. To avoid a sort of "anything ing, shaming without naming particular goes" syndrome, where there is no common countries, and different indicators within a progress at all, even in the learning mode common set of guidelines. Organizations like there has to be a process in which countries IC should discuss with their target countries share their experience, and use indicators to these processes of knowledge utilization talk about their different opinions. This before they attend to the design of indicators. approach enhances transparency and It is fundamental to have a transparent dis- accountability. It relies on facilitated coor- cussion and shared beliefs about what the dination in institutions like the OECD and indicators are about and how they will be the EU to create more evidence-based dia- used by whom to set the better regulation ini- logue on regulatory reform. tiatives on the right tracks. 33 WB204_RQI.indd 33 5/17/10 3:49:30 PM BIBLIOGRAPHY ACTAL (Dutch Advisory Board on Administra- (The) British Chambers of Commerce (2004) tive Burdens) (2003) Work Programme 2004. The "Are regulators raising their game? UK regulatory Hague: ACTAL, www.actal.nl. impact assessment in 2002/3," April, www .chamberonline.co.uk/pdf/RIA_report.pdf. Auditor General of Canada (2000) 2000 Decem- ber Report of the Auditor General of Canada. Carroll, (2008) "Regulatory performance indica- tors in government departments." Paper presented December 2000 Report, Chapter 24 Federal at the Sixth biennial one-day symposium on Health and Safety Regulatory Programs, http:// Accountability, Governance and Performance, www.oag-bvg.gc.ca/internet/docs/0024ce.pdf Dept of Accounting, Finance and Economics, Griffith Business School, University of Griffith, Botero, J., S. Djankov, R. La Porta, F. Lopez-De- 15 February. Silanes and A. Shleifer (2003) The Regulation of Labor, National Bureau of Economic Research Cecot, C., R.W.Hahn, A. Renda, L. Schrefler Working Paper No. 9756. Cambridge, MA: (2007) "An Evaluation of the Quality of Impact National Bureau of Economic Research, http:// Assessment in the European Union with Lessons papers.nber.org/papers/w9756.pdf. for the U.S. and the EU." AEI-Brookings Jot Cen- ter Workg Paper No. 07-09, Available at SSRN: BERR ­ Department for Business Enterprise & http://ssrn.com/abstract=984473 Regulatory Reform (2008a) Regulatory Budgets: A consultation document. London, August, http:// Council of European Union (2007) Brussels Euro- www.berr.gov.uk/files/file47130.pdf pean Council 8/9 March 2007 Presidency Conclu- sions, Brussels, 7224/1/07, 2 May. ------. (2008b) Impact assessment of regula- tory budgets consultation. London, 5 August, Conservative Party (2005) Action on deregulation, www.berr.gov.uk/files/file47130.pdf. London. 34 WB204_RQI.indd 34 5/17/10 3:49:30 PM Conway, P., V. Janod and G. Nicoletti (2005) Hahn, R. W., R. P. Malik, and M. Dudley (2004) Product Market Regulation in OECD Countries: "Reviewing the Government's Number on Regu- 1998 to 2003, Economic Department Working lation." AEI-Brooking Joint Center for Regula- Paper No. 419, ECO/WKP(2005)6, Paris: tory Studies Working Paper, No. 04-03, January, OECD. Washington D.C., www.aei-brookings.org/ admin/authorpdfs/page.php?id=321. Damonte Alessia (2006) "Le sfide della governance europea." Presentation at the URGE Conference on Harrington, W., R. D. Morgenstern and P. Nel- Governance multilivello e politica europea, Collegio son (2000) "On the accuracy of regulatory cost Carlo Alberto, Moncalieri, Italy, 29 November. estimates."Journal of Policy Analysis and Manage- ment, 19:2, 297­322. De Francesco, F. and C.M. Radaelli (2007) "Indi- cators of regulatory quality." in C. Kirkpatrick Hellenic Presidency of the Council of the Euro- and D. Parker (eds) Regulatory Impact Assessment. pean Union (2003) Report to the Ministers Respon- Towards Better Regulation? pp. 36­55, Chelten- sible for Public Administration in the EU Member ham: Edward Elgar. States on the Progress of the Implementation of the Mandelkern Report's Action Plan on Better Regula- Djankov, S., R. La Porta, F. Lopez de Silanes and tion. Athens: Ad Hoc Group of Experts on Better A. Shleifer (2001) The Regulation of Entry. CEPR Regulation. Discussion Paper No. 2953, London: Centre for Economic Policy Research, http://ssrn.com/ House of Common Regulatory Reform Com- abstract=283839. mittee (2008) Getting Results: The Better Regula- tion Executive and the Impact of the Regulatory Doern, B. (2007) Red Tape, Red Flags: Regulation Reform Agenda. www.publications.parliament. in the Innovation Age (Conference Board of Can- uk/pa/cm/cmdereg.htm ada, Scholar-in-Residence Book). IEEP Institute for European Environmental Policy European Commission (2005a) Better Regulation (2004) "Sustainable development in the Euroepan for Growth and Jobs in the European Union. COM Commission's integrated impact assessment for (2005) 97 final, Brussels: EC, 16 March. 2003." Final report, April. GAO (General Accounting Office) (2000) Regu- IEG Independent Evaluation Group the World latory Reform ­ Procedural and Analytical Require- Bank Group (2008) Doing Business: An Inde- ments in Federal Rulemaking. Washington, DC: pendent Evaluation Taking the Measure of the US General Accounting Office. World Bank-IFC Doing Business Indicators, Washington DC: the World Bank. GAO (2004) "Performance Budgeting - Obser- vation on the Use of OMB's Program Assessment Impact Assessment Board European Commis- Rating Tool for the Fiscal Year 2004 Budget." sion (2008) Impact Assessment Board Report for the January, Washington D.C.: U.S. General year 2007. Commission Staff Working Docu- Accounting Office. ment, SEC (2008) 120, Brussels: European Commission. Hahn, R. W., J. K. Burnett, YH. I. Chan, E. A. Mader, and P. R. Moyle, (2000) Assessing the Jacobzone, S., C. Choi and C. Miguet (2007) Quality of Regulatory Impact Analyses. Harvard "Indicators of regulatory management systems." Journal of Law and Public Policy, 23:3, Summer, OECD Working Papers on Public Governance, 859-85. 2007/4, Paris: OECD Publishing. 35 WB204_RQI.indd 35 5/17/10 3:49:30 PM Jalilian, H., C. Kirkpatrick, and D. Parker (2007) McGarity, T. O. and R. Ruttenberg (2002) "The impact of regulation on economic growth "Counting the cost of health, safety, and environ- in developing countries: A cross-analysis." World mental regulation," Texas Law Review, 80:7, Development 35:1, 87­103. 2017. Kaufmann, D., A. Kraay and M. Mastruzzi (2003) NAO (National Audit Office) (2001) Better Governance Matters III: Governance Indicators for Regulation: Making Good Use of References 249 1996­2002. World Bank Policy Research Work- Regulatory Impact Assessments. Report by the ing Paper No. 3106, Washington, DC: World Comptroller and Auditor General, HC 329, ses- Bank, www.worldbank.org/wbi/governance/pdf/ sion 2001­02, London: The Stationery Office. govmatters3.pdf. ------. (2004) Evaluation of Regulatory Impact ------. (2005) Governance Matters IV: Gover- Assessments Compendium. Report 2003­04. nance Indicators for 1996­2004. World Bank Report by the Comptroller and Auditor General, Policy Research Working Paper No. 3630, Wash- HC 358, session 2003­2004, 4 March 2004, ington, DC: World Bank, www.worldbank.org/ London: The Stationery Office, www.nao.org. wbi/governance/pdf/GovMatters_IV_main.pdf. uk/publications/nao_reports/03-04/0304358. pdf. ------. (2008) Governance Matters IV: Gover- nance Indicators for 1996­2004. World Bank ------. (2005) Evaluation of Regulatory Impact Policy Research Working Paper No. 4654, Wash- Assessments Compendium Report 2004­05. Report ington, DC: World Bank, http://papers.ssrn. by the Comptroller and Auditor General, HC com/sol3/papers.cfm?abstract_id=1148386 341, session 2004­05, London: The Stationery Office. Knack, S. and M. Kugler (2002) Constructing an Index of Objective Indicators of Good Governance. ------. (2007) Reducing the cost of complying Washington, DC: World Bank, www1.world- with regulations: The delivery of the administrative bank.org/publicsector/anticorrupt/Flagship- burdens reduction programme 2007. Report by Course2003/SecondGenerationIndicators.pdf. the Comptroller and Auditor generl, HC 615 Session 2206-2007, 25 July 2007. Koedijk, K. and J. Kremers (1996) "Market opening, regulation and growth in Europe." Eco- NERA (2005) Regulatory Transparency: Interna- nomic Policy, 11:23, 443­68. tional Assessment and Emerging Lessons. A Final Report for the World Bank. London: NERA Eco- Lee, N. and C. Kirkpatrick, (2004) "A pilot Study nomic Consulting. on the Quality of European Commission Extended Impact Assessment." Impact Assess- Nicoletti, G., S. Scarpetta and O. Boylaud (2000) ment Research Centre, Institute for Develop- Summary Indicators of Product Market Regulation ment Policy and Management, University of with an Extension to Employment Protection Legis- Manchester, 21 June. lation. Economics Department Working Paper No. 226, Paris: OECD. Legislative Burden Department (2003) Focus on Burden! The Hague: Ministry of Finance, Decem- OECD (1995) Recommendation of the Council of ber 2003. the OECD on Improving the Quality of Government Regulation. Paris: OECD Publishing. Retrieved Mandelkern (Group on Better Regulation) from http://www.olis.oecd.org/olis/1995doc.nsf/ (2001) Final Report, Brussels. LinkTo/OCDE-GD(95)95. 36 WB204_RQI.indd 36 5/17/10 3:49:31 PM ------. (1999) "Guidance for conducting retro- Radaelli, C.M., F. De Francesco and V.E. Troe- spective studies on socio-economic analysis." ger (2008). "The implementation of Regulatory OECD Environmental Health and Safety Publi- Impact Assessment in Europe." Paper presented cations ­ Series on Risk Management No. 11, at the European Network for Better Regulation, ENV/JM/MONO (99)27, Paris: OECD, 25 Workshop on Regulatory Quality: Developing tools, January 2000. approaches and sources for research on impact assessment, University of Exeter, Exeter 27 and ------. (2003) Proceedings from the OECD 28 March. Expert Meeting on Regulatory Performance: Ex post Evaluation of Regulatory Policies. 22 September, Regulatory Consulting Group and Delphi Group Paris: OECD Publishing. www.oecd.org/ (2000) Assessing the Contribution of Regulatory dataoecd/34/30/30401951.pdf Impact Analysis on Decision Making and the Devel- opment of Regulation, Ottawa: Regulatory Con- ------. (2007) Administrative Simplification in sulting Group and Delphi Group. the Netherlands, Paris: OECD Publishing. Sabel, C. F. (1994) "Learning by Monitoring: ------. (2008) Indicators of regulatory manage- The Institutions of Economic Development." ment systems quality. Working Party on Regula- Neil Smelser and Richard Swedberg (eds), Hand- tory Management and Reform, Paris: OECD book of Economic Sociology, Princeton: Princeton Publishing. University Press and Russell Sage Foundation, p 137­165. OMB Office for Management and Budget (2002) Program Assessment Rating Tool - Instructions for SQW (2005) Evaluating the Impact of Regulation: PART Worksheets. Fiscal Year 2004, www.omb.gov. Developing a Methodology, report for the Depart- ment of Trade and Industry. London: SQW, Office of Technology Assessment (OTA) (1995) www.dti.gov.uk/files/file22009.pdf?pubpdf Gauging Control Technology and Regulatory Impacts dload=05%2F1371. in Occupational Safety and Health ­ An Appraisal of OSHA's Analytic Approach. OTA-ENV-635, Stake, R.E. and T.A. Schwandt (2006) "On dis- Washington, DC: US Government Printing cerning quality in evaluation." I.F. Shaw, J.C. Office. Greene and M.M. Mark (eds.) The SAGE Hand- book of Evaluation, pp. 404­18, London: SAGE Parker, R.W. (2003) "Grading the government." publications. University of Chicago Law Review, 70:3, 1345­ 422. Treasury Board of Canada (1996) Federal Regula- tory Process Management Standards ­ A Self- Pryor, F.L. (2002) The Future of U.S. Capitalism. Assessment Guide for Departmental Managers ­ Cambridge: Cambridge University Press. Compliance, Ottawa: Treasury Board of Canada Secretariat, www.pco-bcp.gc.ca/raoics-srdc/docs/ Radaelli, C.M. (2007). Whither better regulation publications/rpms_e.pdf. for the Lisbon agenda? Journal of European Public Policy, 14(2), 190­207. Van de Walle, S. (2008a, forthcoming) "Interna- tional comparisons of public sector performance: Radaelli, C.M. and F. De Francesco (2007) Regu- how to move ahead?" Public Management Review. latory Quality in Europe. Concepts, Measures and Policy Processes, Manchester: Manchester Univer- Van de Walle, S. (2008b) "Comparing the per- sity Press. formance of national public sectors: Conceptual 37 WB204_RQI.indd 37 5/17/10 3:49:31 PM problems." International Journal of Productivity Studies Working Paper, No. 04-05, Washington and Performance Management, 57 (4): 329­338. DC: AEI-Brookings Joint Center for Regulatory Studies. Vibert, F. (2004) The EU's New System of Regula- tory Impact Assessment ­ A Scorecard. London: Whitmore, E., I. Guijt, D.M. Mertens, P.S. Imm, European Policy Forum, March. M. Chinman and A. Wandersman (2006) "Embedding improvements, lived experience, Vibert, F. (2005) The Itch to Regulate: Confirma- and social justice in evaluation practice." I.F. tion Bias and the European Commission's New Sys- Shaw, J.C. Greene and M.M. Mark (eds.) The tem of Impact Assessment. London: European SAGE Handbook of Evaluation, pp. 404­18, Lon- Policy Forum. don: SAGE publications. Wallsten S., G. Clarke, L. Haggarty, R. Kaneshiro, R. Noll, M. Shirley, and L.C. XU (2004) The Tele- World Bank Group (2007) Review of the Dutch communication and Electricity Regulation Data- Administrative Burden Reduction Programme, base, AEI-Brookings Joint Center for Regulatory Washington, WBG. 38 WB204_RQI.indd 38 5/17/10 3:49:32 PM ANNEX 1: BETTER REGULATION FOR GROWTH PROGRAM The Better Regulation for Growth (BRG) Pro- The BRG Program has resulted in preparation of gram was launched in 2007 by the Dutch Minis- eight policy papers on regulatory governance try of Foreign Affairs, the UK Department for issues, covering a broad spectrum: from regula- International Development (DFID) and IC, the tory governance, links to competition policy, investment climate advisory service of the World regulatory institutions, and tools to indicators for Bank Group. regulatory quality. It has also involved prepara- tion of five country case studies on regulatory The objective of the BRG is to improve the regu- capacities in selected African countries. latory and investment climate in developing countries, thereby stimulating private sector The web portal www.ifc.org/brg is part of the investment, economic growth and poverty reduc- BRG Program and contains key documents, tion. The BRG program aims to achieve this by including references extracted from a compre- developing and disseminating for the first time hensive compendium of resources on regulatory widely practical and operational guidance that management and reform and a newly developed will help developing countries design and imple- Regulatory Impact Analysis (RIA) database. ment effective regulatory reform programs. 39 WB204_RQI.indd 39 5/17/10 3:49:33 PM WB204_RQI.indd 40 5/17/10 3:49:34 PM