Report No: ACS7828 Republic of Kosovo Statistical and Analytical Capacity Building Completion Memo December 3, 2013 ECSP3 EUROPE AND CENTRAL ASIA Document of the World Bank Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Copyright Statement: 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 International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce 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, http://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, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Kosovo Statistical and Analytical Capacity Building (P127642) Completion Memo December 3, 2013 1. Background Since 2006, the World Bank has employed a programmatic and sub-regional approach to poverty- focused analytical and policy work in the Western Balkans (comprising Albania, Bosnia-Herzegovina, Kosovo, the Former Yugoslav Republic of Macedonia, Montenegro, and Serbia). The programmatic multi-year approach has ensured continuity in long term efforts such as statistical capacity building, while also permitting a great deal of flexibility to respond to emerging developments such as the financial crisis and recession. In addition to the Bank’s own resources, the Western Balkans Programmatic Poverty Assessment (PPA) was supported by a trust fund financed by the United Kingdom’s Department for International Development (DFID, also known as UKAid) from 2006 until 2011. In early 2011 the DFID trust fund closed and DFID withdrew its operational and financial support from all Western Balkans countries except for Kosovo, which continued receiving assistance because of its greater development needs. DFID continued its financial support to the Bank’s work on poverty in Kosovo through the Externally-Financed Output (EFO) instrument, which complemented the Western Balkans programmatic poverty assessment work that has been funded out of the Bank budget. The Kosovo-specific financial support from DFID enabled the Western Balkans poverty team to provide much more regular and in-depth technical assistance to Kosovo institutions, especially the Kosovo Agency of Statistics (KAS). The first EFO from DFID funded the Kosovo Statistical and Analytical Capacity Building (SACB) project (P127642), which is the subject of this completion memo. DFID contributed GBP 345,000 for the World Bank to carry out a set of activities designed to improve the quality of survey data and expand the use of survey data for evidence-based policy making in Kosovo. The original timeframe of the project was from August 2011 through December 2012, with the end-date of DFID’s support being extended to August 31, 2013 by mutual agreement of DFID and the World Bank. DFID also financed a second EFO for the Kosovo Labor Force Survey Support task (P132962). The three main objectives of the task were: 1) To enhance the ability of KAS (formerly known as the Statistical Office of Kosovo) to collect household survey data and to conduct and disseminate survey data analysis. 2) To enlarge the pool of users of data for primary and secondary analysis, expanding access to data for users outside of KAS, as well as strengthening the capacity of users to undertake and present policy-relevant analysis. 3) To increase the use of empirical, data-based policy analysis in the formulation, implementation, and evaluation of polices in Kosovo. The identification of activities that would help achieve these objectives was informed by consultations with multiple stakeholders, including the management and staff of KAS, officials in key ministries, researchers in Kosovo, international development partners, and World Bank staff and consultants. The following set of six interrelated and mutually reinforcing activities were planned. 1 A. Continued direct support to the survey teams in KAS B. Establishment of a Data Users Network C. Series of training workshops on analyzing survey data D. Analytical work on inclusive growth, employment, poverty, inequality and other topics E. Commissioned papers and applied policy research seminars F. Conference on inclusive growth and public policy Throughout the project the PREM team worked closely with the ECA Human Development department, especially the Economics and Social Protection clusters in HD. These groups are also major users of KAS’s survey data, and they brought valuable insights on improving data processes and expanding the use of household survey data in Kosovo. The HD team helped deliver the training workshop on social protection analysis using ADePT, and was an important partner in delivering the regional conference on “Shared Prosperity, Inclusion and Jobs in the Western Balkans.” This report summarizes and reviews progress on these activities during the lifespan of the project (August 2011 through November 2013). 2. Summary of Outputs A. Direct Support to Survey Teams in KAS The KAS team for the implementation of the Household Budget Survey (HBS) is small, with only three analysts at KAS headquarters and one survey interviewer in each of the seven regions of Kosovo. One of the analysts is relatively new to the unit, and only one of the analysts has strong skills with statistical software. The direct support to the HBS team has taken the form of on-the-job training, working side by side with the team at the KAS office on the processing and analysis of current HBS data for the eventual publication of joint reports on poverty and inequality. Although some of the support was on specific data analysis techniques, we found that the most valuable contribution was helping to improve the systems and protocols for processing and analyzing data. The ability to replicate analyses is a basic standard for statistical offices, so institutionalizing these practices is essential, especially considering the small HBS team at KAS. Results1: In our assessment this activity has helped improve the practices and standards of the HBS team at KAS. The HBS team has a deeper understanding of data processing and analysis issues, and a better appreciation for documentation and institutionalizing systems. Evidence of this may be seen in the unit’s continued smooth functioning even when one of the members is away for several months on training. Even so, critical operations depend on a small group of people, and if one were to leave the unit it is likely that there would be a noticeable decline in the quality of the team’s work. B. Establishment of a Data Users Network The establishment of a community of survey data users in Kosovo is an instrumental objective of this project. The scope encompasses data users and potential users from KAS, government ministries, the central bank, international organizations, academic institutions, and private research organizations. The rationale is that such a community would conduct and share analysis using KAS’s survey data, and that this interaction would increase awareness of data sources, and elevate the evidence base for public 1 A summary of results, using the results framework from the project concept note, appears at the end of this memo. 2 policy debate in Kosovo. It is also envisioned that increased demand for KAS’s survey data would raise the stature of KAS and the public goods it provides in the form of official data, thus strengthening KAS institutionally. Moreover, by encouraging wider usage and scrutiny of survey data, it is anticipated that data users may help KAS improve the quality of their data collection. As noted in the 2012 annual review by DFID, the Data Users Network has been organized as a loose construct. The network is deliberately inclusive: in principle it is open to any professional analyst in Kosovo, with the only requirement being some degree of familiarity with the analysis of household survey data. The main mode of contact in the network has been the training workshops that are described in item 3, complemented by the regional conference (item 6) and other public events for the presentation of results. Results: The project has helped cultivate a network, albeit a loose one, of data analysts from a range of public and private institutions in Kosovo. There is greater awareness and appreciation of each other’s work, and this has contributed to policy debate. That said, the vitality of the network has been limited by the continued limited access to KAS’s survey data for analysis and research. The less than satisfactory results of the planned papers and seminars series (item 5) also represent a missed opportunity to expand the network and raise its public profile. The training workshops have been the main “glue” that links the network, and the frequency of these will decline sharply with the end of DFID’s support. To sustain the nascent network, it will help to make the network a more concrete entity, as advised in the DFID annual review. This could include active maintenance of an email distribution list and perhaps setting up an online discussion forum. The World Bank has also added network members’ names to its email distribution lists, to be informed and invited to public events that are likely to be of interest, such as the launch of the Labor Force Survey report at the end of September 2013. C. Series of Training Workshops on Analyzing Survey Data A series of six training workshops were conducted under the scope of this project. The main objectives of the workshops were to (a) acquaint potential users outside of KAS with the contents of the rich survey data sets that are collected by KAS, (b) strengthen participants’ skills in analyzing survey data and interpreting the results, (c) promoting discussion among participants about the meaning of results in practical terms, and the implications for public policy, and (d) building a community of data users (consistent with the objective of establishing data users network as described earlier). The topics and dates of the workshops were as follows: - Performing Poverty Analysis with ADePT Software (November 2011) - Gender Analysis using SPSS and ADePT (December 2011) - Analysis of Social Protection using SPSS and ADePT (February 2012) - Analysis on Migration and Remittances using SPSS Software (April 2012) - Analysis of HBS Consumption Data to Estimate Poverty Level using SPSS (February 2013) - Analysis of Labor Force Survey using SPSS (March 2013) The workshops were attended by analysts from KAS, government ministries, international organizations, universities, and private research firms, with between 9 and 16 participants per workshop. Many individuals attended multiple workshops in the series. The workshops were led by World Bank staff and consultants, with guest presentations by Kosovar researchers. KAS staff contributed to the presentations as well, especially on topics specific to the HBS or the Labor Force Survey (LFS). Each of the workshops was three days in length. The common format was to spend the first morning on presenting important 3 concepts behind the topic for the workshop, and devote the remaining time largely to hands-on exercises in the computer lab. The practical aspect of the exercises was reinforced by using actual recent survey data collected by KAS. As there was often a range of skill levels and experience among the participants, those who were more advanced were encouraged to share their experiences and insights. Participant evaluations of the workshops were very positive. The most common negative feedback was that those outside of KAS generally still do not have access to the individual- and household-level survey data from KAS. Results: The workshops have been very successful in exposing a broad range of potential users to the content and structure of KAS’s survey data sets, and also how the survey data may be used to inform policy. They have also been successful at promoting networking among analysts, including some interesting informal debates about survey results and their meaning. The workshops could have been more successful if it had been possible to include examples of more in-depth and policy-relevant analyses using household survey data in Kosovo. Even though most of the participants are not yet operating at that level, it could have been a useful (and motivating) demonstration of the power and potential for using survey data for evidence-based policy making. D. Analytical work on inclusive growth, employment, poverty, inequality and other topics National statistics offices are responsible not only for collecting and processing data, but also for reporting national statistics. With respect to the HBS, KAS has taken full responsibility for producing an annual report of the results of the HBS, including expenditure and income patterns across a standard array of covariates. Since 2002, the World Bank has had the lead role—with support from KAS—in using the HBS to produce reports on poverty, inequality, and other topics concerning the economic well-being of the population. An objective of this project has been to progressively shift primary responsibility for production of poverty reports from the World Bank to KAS, with the World Bank providing technical support as needed. This is the route that the World Bank and partners have been taking in other Western Balkans countries. Initially the reports led by the national statistical offices are less in-depth than the World Bank-led reports, but they become richer over time. During this project KAS published two short poverty reports, one based on the 2010 round of the HBS and the other on the 2011 round. Copies of the reports are attached. Good progress was also made on a longer poverty report that covers trends in poverty using comparable data from the 2009 HBS through the 2011 HBS. This report will be completed by December 2012. - Consumption Poverty in the Republic of Kosovo in 2010 (December 2012) - Consumption Poverty in the Republic of Kosovo in 2011 (March 2013) - Preparation for long poverty report covering HBS 2009–2011 Results: Considerable progress was made in shifting greater responsibility for production of standard poverty reports (“short poverty reports”) to KAS staff. To accomplish this, the format was streamlined so that the reports covered the main indicators for poverty and extreme poverty, namely, headcount ratio and depth, with disaggregations by urban/rural location, gender, household size, main source of household income, main economic activity, and highest educational level completed. Even though the HBS team at KAS took greater responsibility for producing the analysis, completion of the reports still required significant contributions from World Bank staff and especially consultants. One of the consultants most involved in helping to produce the reports is a Prishtina-based Kosovar academic, 4 meaning that the external support to KAS draws from local expertise more than in the recent past. Under the current staffing configuration of the HBS team, some form of external technical assistance will probably be required for at least the next two or three rounds of these annual reports. E. Commissioned papers and applied policy research seminars To help stimulate policy analysis and research using the household survey data that KAS collects, it was planned to commission four or five policy-focused papers pertinent topics, particularly related to inclusive growth and the employment/poverty nexus. It was also planned that these papers would be presented publicly in a seminar series, to which interested persons in both the public and private sectors would be invited. In addition to providing information that may be directly useful for decision makers in Kosovo, the paper and seminar series was expected to provide concrete examples of the potential power of analysis based on household survey data. Results: The outcome for this component of the project was unsatisfactory. Efforts to identify and engage researchers on relevant topics within Kosovo or internationally were unsuccessful. A major constraint on this front continues to be the limited access to KAS survey data. Even though KAS has been more open to requests for survey data, the process is still not standardized or transparent; some analysts who were approached about preparing papers were pessimistic about the prospects of getting survey microdata from KAS. In addition to data limitations, in retrospect it is likely that expectations about the rigor and sophistication of the papers were too high for this component. For example, UNDP conducted a research paper competition—for which they guaranteed participants access to survey data collected by UNDP—that attracted commendable submissions. A similar approach for this project may have been more successful. F. Regional conference on inclusive growth and public policy As part of the Kosovo Statistical and Analytical Capacity Building project, a Western Balkans regional conference on “Shared Prosperity, Inclusion and Jobs in the Western Balkans” was held from June 11– 13, 2013 Bečići, Montenegro. The main objective of the conference was to provide a forum for researchers, policy makers, statistical agencies, and donors to discuss changes in living conditions in the Western Balkans, in the context of recovering from its double-dip recession. As a follow-up to the very successful “Poverty and Social Inclusion in the Western Balkans” conference that was held in Brussels in December 2010, the conference aimed to stimulate the sharing of knowledge and experiences across an informal and multi-sectoral network of development professionals working in the Western Balkans. A total of 89 people participated in the conference, including officials from ministries and statistics offices, researchers, international organizations, NGOs, the European Commission, and a parliamentarian. A full list is included in the conference program (attached). The agenda comprised a mixture of plenary and parallel paper sessions. The plenaries included presentations on recent policy-focused analytical work led by the World Bank, as well as discussion panels on data and policy topics. The subjects of the plenary sessions were: - The Jobs Challenge in the Western Balkans - Poverty and Shared Prosperity - Panel Discussion on Poverty and Inclusion - How Can We Strengthen the Links between Social Science Research and Policy in the Western Balkans? 5 - Gender, Employment and Inclusion - Discussion Panel on Sharing of Experiences in Adopting the EU Statistics on Income and Living Conditions (EU-SILC) Surveys - Activation and Smart Safety Nets in the Western Balkans A Call for Papers was circulated widely in March 2013, and 28 papers were selected from the approximately 45 submissions received. Themes for the paper sessions included poverty and economic mobility, labor market policies, inclusion, gender, implementing EU-SILC surveys, and migration and remittances. Results: The principal result of the conference was increased communication and awareness among participants about work being done in other countries and other sectors. In particular, the conference very effectively communicated participants’ recent work on poverty, inclusion, and shared prosperity, jobs, gender, and smart safety nets. The active discussions and debates during the plenary and parallel sessions clearly carried over to coffee breaks and meals, and several participants went out of their way to comment that the conference had succeeded in facilitating networking. A summary of the participant evaluations is attached. 3. Dissemination and outreach The two short poverty update reports were disseminated through the usual methods, including hard copies of the reports (in Albanian, Serbian, and English), press releases, media interviews with KAS staff, and publication on the KAS and Bank web sites. As the scope of these reports was deliberately brief (consistent with transferring more of the production responsibility to the KAS team), the outreach to the media was less ambitious than was the case for the companion LFS Support project. The outreach to establish the Data Users Network was arguably as important as the publicizing of new poverty results from the HBS. Many network members had not worked directly with KAS survey data before, and were not fully aware of its richness and potential usefulness to address key policy questions. The training workshops exposed network members to the contents of KAS’s data and to a range of analytical techniques has helped not only to nurture a cadre of more informed consumers of empirical analysis, but also to build a constituency for greater access to survey data. 4. Sustainability and lessons learned This task has aimed to strengthen both the supply for and demand of high-quality survey data in Kosovo, and in the process promote the use of empirical evidence in policy making. On the supply side the KAS survey teams, particularly the HBS team, have improved their ability to analyze and interpret survey data, and have taken greater responsibility for the production of reports on poverty and well-being. They are now capable of producing standard statistics, tables, and graphs. They can also follow a simple template to provide accompanying narrative text, but for anything more in-depth outside assistance is required from a local consultant and/or Bank staff. The small size of the HBS team presents a risk if one of the members were to leave, especially as the skill depth is not as good as it could be (that is, there are some tasks that only one HBS team member knows how to do). Progress on the demand side has been much broader. Although the Data Users Network is still nascent, it reaches across an array of organizations, including KAS, government ministries, academia, private think-tanks, and international organizations. The professionals that make up the network are mostly young and eager to interact and learn. Although the training workshops will not continue in the absence 6 of additional external funding, the Western Balkans PPA team and other Bank teams can help the network continue to grow by collaborating with network members on analytical work or reaching out to them when disseminating work. Network members are especially eager to get access to KAS’s survey microdata. Although KAS has become a little more open about sharing data, it is still quite restrictive. Continued work is needed to demonstrate that wider data sharing—within wide but clear boundaries—is in KAS’s self-interest as a data producer. The apparent success of UNDP’s papers competition is also a signal that greater success in encouraging local analysis of primary data is possible. Table 1: Results matrix (from results framework in project concept note) Immediate Outcome Indicator Activities Client capacity Monitoring & Evaluation  Direct technical assistance (on-the-job-training) to HBS increased capacity increased team in KAS  Establishment of data users network  Series of training workshops on analysing survey data Knowledge deepened Facilitated exchange of  Analytical work on poverty, inequality and inclusive best practice w/ clients, growth Disseminated best  Regional conference on “Shared Prosperity, Inclusion practices and Jobs in the Western Balkans” Attachments (also filed in WBDocs) - Poverty reports - Regional conference program and participant evaluation summary 7