SOCIAL PROTECTION & JOBS DISCUSSION PAPER No. 1931 | MAY 2019 Can Local Participatory Programs Enhance Public Confidence: Insights from the Local Initiatives Support Program in Russia Ivan Shulga, Lev Shilov, Anna Sukhova, and Peter Pojarski © 2019 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: +1 (202) 473 1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, 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. RIGHTS AND PERMISSIONS The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: +1 (202) 522 2625; e-mail: pubrights@worldbank.org. Abstract retro geometric background: © iStock.com/marigold_88 Project 41595 Abstract This paper reviews the performance of the Russia Local Initiatives Support Program (LISP) as an instrument for directly addressing the needs of the population to access socioeconomic infrastructure, for increasing public confidence in self-governance frameworks and institutions through dialogue and community budgeting consultations, and for strengthening the capacity for local self-governance. The paper does this by looking at historical and survey data from the implementation of the LISP methodology as part of regional programs in Russia. JEL Classification: H53 – Government Expenditures and Welfare Programs; H75 – State and Local Government: Health • Education • Welfare • Public Pensions; I38 – Government Policy • Provision and Effects of Welfare Programs. Keywords: citizen engagement, community-driven development, participatory budgeting, initiative budgeting, Local Initiatives Support Program, Russia. 2 Table of Contents Executive Summary ................................................................................................................ 6 1. Introduction ................................................................................................................ 8 2. Evolution of LISP ...................................................................................................... 13 3. LISP design................................................................................................................. 16 4. Data Sources and Methodology ............................................................................... 21 5. LISP Results ............................................................................................................... 25 A. Improving Local Infrastructure and Services ........................................................ 25 B. Improving Communications and Dialogue ............................................................ 32 C. Contributing to Public Confidence ........................................................................ 36 6. Lessons on Sustainability and Institution Building .................................................. 41 Annex A: Survey Methodology ............................................................................................ 47 Annex B: Changes in the Levels of Public Confidence in Local Authorities and Satisfaction with Public Goods Provision ............................................................................... 57 References............................................................................................................................. 68 3 Acronyms LISP Local Initiatives Support Program LPM Linear Probability Model MoF Ministry of Finance [of the Russian Federation] PB Participatory Budgeting Glossary CITIZENS: The permanent residents of a town, city, municipality, or community, who have some form of legal local residency status that includes specific rights and obligations. INITIATIVE BUDGETING (IB): A term used to describe a variety of participatory budgeting practices that exist in Russia and share some features that are not quite common internationally, such as: (i) final approval of (micro)projects based on a set of formal criteria; and (ii) administration, financing and supervision at the regional (sub-national) rather than municipal level. Additionally, most IB practices (including LISP) require co-financing of (micro)projects by citizens and local business. LOCAL INITIATIVES SUPPORT PROGRAM (LISP): A methodology for participatory (initiative) budgeting developed by the World Bank for Russia regions and based on the principles of direct involvement of citizens in identification and prioritization of microprojects, final approval of microprojects based on a set of formal criteria, and cofinancing of microprojects by citizens and local business. MICROPROJECT: A short-term activity, selected and implemented under the LISP principles, that is meant to address a specific challenge faced by the community, and that can be managed at the local level. Microprojects can be stand-alone in nature, but they are effective building blocks for larger projects as well. PARTICIPATORY BUDGETING (PB): Participatory budgeting (PB) is a decision-making process through which citizens deliberate and negotiate over the distribution of local public resources 4 (Wampler, 2007: 21). Decision-making process includes direct participation in identification, discussion, and prioritization of projects to be financed from the public budget. SOCIAL NEEDS: The needs that create social value and opportunities for people to have an active and effective role in society. SOCIOECONOMIC INFRASTRUCTURE: In the context of LISP, socioeconomic infrastructure encompasses any of a number of assets typically used to accommodate social or public services, and that are owned, maintained, and run by the municipalities authorized by local legislation to implement LISP-type subprojects. 5 Executive Summary The Local Initiatives Support Program (LISP) is a participatory project identification and financing mechanism that is used by Russian regions to provide financial support to microprojects initiated directly by local communities. These microprojects are delivered through the joint efforts of local authorities, local businesses, and the local population. Although the LISP methodology is tailored to each region where it is implemented, it pursues two overarching objectives that cut across all regional programs: 1) To improve access and quality of local socioeconomic infrastructure and services, and 2) To improve the communication and dialogue between citizens and local authorities. The available data and information from the years of LISP experience across Russian regions reveal important differences between LISP participants and non-participating communities concerning specific outcomes. Overall, data suggest that LISP could have contributed to improvements in access to and quality of local socioeconomic infrastructure, and to the effectiveness of the communication between the population and local authorities. First, by financing microprojects, LISP-based regional programs produce numerous assets that are highly valued by the beneficiaries and result in better social services and improved small-scale infrastructure. The participation of regions and communities in LISP-based programs increased steadily over the years, and coverage grew. The very nature of the LISP process incentivized regions to implement the project in disadvantaged areas, and – although there was no poverty-based targeting – beneficiaries were often people from poor and vulnerable rural areas. Second, participation in LISP-type programs improves communication between local authorities and communities and likely leads to increased public confidence in local governance mechanisms. In LISP participating settlements, public satisfaction with local infrastructure and services is higher compared to nonparticipating communities. 6 The LISP methodology contributes to these results through its versatile design, which provides needed flexibility, while at the same time consistently employs several important principles: directly involving the population in identifying problems and their solutions; unlocking the potential of local communities and local self-governing bodies, promoting openness and transparency; and concentrating practical work at the grassroots level. Another winning element of the LISP design is embedding all key processes in national administrative, budgetary, and legal systems, which helps build institutional resilience. 7 I. Introduction Institutional and Social Context Although it is an upper-middle-income country, Russia still faces a number of development challenges at the local level. Russia is characterized by noticeable differences in social and economic indicators between its urban and rural areas. While cities enjoy better infrastructure and access to services, many rural municipalities face persistent problems with both the quality of, and access to, the basic socioeconomic infrastructure and communal services, such as water supply, roads, and wastewater disposal. Poverty rates in these areas are often higher, and social safety net mechanisms are weaker because of lower administrative government capacity, lower economic development, and physical distance from major cities. Second, while the responsibility of meeting the basic social needs of the population, including development and maintenance of the local socioeconomic infrastructure, falls on the municipal level, many small municipalities lack the resources to effectively address these needs. Under the highly centralized Russian budgetary system, most resources are accumulated at the higher levels (federal and regional), and then redistributed down to the municipal level through centralized state programs with earmarked funds (see Box 1 with Russia’s administrative structure). These programs are usually geared toward larger infrastructure of national and regional significance, and therefore have limited impact in addressing basic local needs, particularly in smaller settlement-level municipalities. On the other hand, local governments have little other sources of income to address specific needs of their populations that do not fall under the earmarked programs. Third, the local self-governance system in Russia is still in the initial stages of developing citizen involvement in decision making. While by the early 2000s Russia had put in place a framework for legislation that provided for various forms of citizen involvement in self- 8 governance (public meetings, gatherings, conferences, etc.)1, the newly available tools were not systematically used in practice. Decisions on spending budget funds are often made in a centralized way, and signals about needs and gaps voiced by the population are not explicitly Box 1. Jurisdictional Divisions of Governance in the Russian Federation Russia is a federative state comprised of two main types of administrative units: subjects (regions) and municipal entities. There are over 80 subjects of the following types: • Republics • cities of federal significance • krais • autonomous oblasts • oblasts • autonomous orkugs These six types of subjects have equal rights, and their administrative borders are enshrined in federal laws. Each subject of Russia is free to establish further internal division. The municipal (local government) entities follow a two-tier principle: high-level (mostly districts) and low-level (mostly settlements) municipalities. The law allows seven types of municipal entities: • rural settlement • urban districts with intra-city division • urban settlement • intra-urban districts • municipal district (rayon) • intra-city municipalities • urban district All 21,945 municipalities are organized within 80+ regions and cities of federal significance. Russia’s administrative system with number of units of each type 1 Federal law #131-FZ, dated October 06, 2003, “On general principles of local self-governance in the Russian Federation.” Chapter 5 of the law outlines several forms of citizens’ exercise of and participati on in local self- governance: gatherings, meetings and conferences of citizens, territorial public self-government, public polls, and other formats legitimizing the realization of grassroots participatory democracy principles. Source: Data as of January 1, 2018, from “Population of the Russian Federation within municipalities,” Official Bulletin of the Russian Federal Statistics service, http://www.gks.ru/wps/wcm/connect/rosstat_main/rosstat/ru/statistics/publications/catalog/afc8ea004d5 6a39ab251f2bafc3a6fce 9 used in prioritizing expenditures (Yushkov, Savulkin & Oding, 2017; Zubarevich, 2014; Hanson, 2006). As a result, a great number of issues that are significantly important for the population at the local level remain unresolved (Zhuravskaya, 2010: 77). According to the latest available official data over a quarter of the Russian population (36 percent of the poor population) live in rural areas with poor access to infrastructure and services: two-thirds of all rural settlements do not have a centralized water supply; 95 percent of rural settlements have no sewerage; one-third of rural settlements have no hard-surfaced access roads.2 This results in the lack of citizens’ confidence in local authorities, which is according to the Center of Sociological Research of the Russian Presidential Academy of National Economy and Public Administration, currently the weakest compared to confidence in all other levels of governance, and which keeps declining (60.3 percent in the beginning of 2015, and 46.4 percent of 2016).3 In 2005, the Government of the Russian Federation approached the World Bank with a request to develop a participatory instrument for identifying and addressing basic social issues at the local level. That instrument was later officially named the Local Initiatives Support Program (LISP). The main idea of the LISP is to directly engage the population in identifying local-level social priorities and to help channel funds directly to support these projects. Under LISP, citizens would identify and then vote for the projects aimed to improve local-level socioeconomic infrastructure, and once certain projects are selected, funds from the regional budget are transferred directly to the settlement budget to support 2 Latest available data from the Russia National Statistics Service (Rosstat), http://www.gks.ru/free_doc/new_site/population/urov/urov_52.doc; http://www.gks.ru/free_doc/doc_2016/jil-hoz16.pdf; http://www.gks.ru/bgd/regl/b13_13/IssWWW.exe/Stg/d1/06-45.htm 3 Between early 2015 and early 2016, levels of trust in the government declined from 79.8% to 59.4%; trust in the governors declined from 72.8% to 58.6%; trust in regional parliaments declined from 68% to 51.3%; and trust in municipal parliaments from 64.9% to 46.8%. See details in the article in RBC: https://www.rbc.ru/politics/29/03/2017/58dbb1ed9a7947e4c4de9dcb 10 implementation. In this way, the flow of funds bypasses all the intermediate units, including the district (rayon) level which is not the case for the regular interbudgetary transfers. Stavropol Krai was the first region in Russia to pilot LISP in 2007. Currently, the LISP is the most widespread participatory model in the country (Shulga and Sukhova, 2016b). With the support of the World Bank, LISP was implemented in 10 territorial units (regions4): Kirov Oblast, Tver Oblast, Nizhny Novgorod Oblast, Sakhalin Oblast, Stavropol Krai, Khabarovsk Krai, the Jewish Autonomous Oblast, Republic of Bashkortostan (Shulga, Fadeeva and Sukhova, 2017), Republic of North Ossetia-Alania, and Republic of Sakha (Yakutia). Since 2007, LISP in these regions has supported nearly 8,000 microprojects in about 2,000 municipalities. Microprojects included repairing roads, refurbishing water supply facilities and community centers, constructing children playgrounds and sport facilities, landscaping and territorial improvement, as well as other initiatives within the authority of local governments. Annually, around 300,000 people take part in over 3,000 community meetings in these 10 regions. At present, 29 regions of Russia actively use the LISP approach, some without direct participation of the World Bank. The purpose of this paper is to showcase how LISP5 has enhanced basic social protection by directly addressing the socio-economic needs of the population and encouraging local self- governance by helping strengthen public confidence in local authorities. By financing public infrastructure microprojects, LISP-based regional programs produce numerous assets that are highly valued by the beneficiaries and result not only in improved small-scale infrastructure but also in better access to social services. The LISP mechanism creates incentives for regions to expand coverage, and to implement the project in disadvantaged areas. The LISP-based programs ensure additionality of resources provided to poor communities and an effective use of these additional resources in improving social 4 The term “region” is used in this paper as the general term referring to the administrative level of any of 80+ autonomous subjects of the Russian federation, as described in Box 1. 5 Henceforth, the use of “LISP” in this paper will not refer to a specific program, but to the practice of participatory budgeting developed by the World Bank and based on a mechanism of direct citizens’ involvement in solving various development challenges of local importance. 11 services and increasing social capital in the target often hard-to-reach vulnerable groups. While LISP-type programs do not have poverty-based targeting, beneficiaries were mostly people from poor and vulnerable rural areas. Participation in LISP-type programs improves the capacities of local authorities to implement and provide better services to their most vulnerable and needy communities. LISP-type programs improved the communication between local authorities and communities, which likely lead to increased public confidence in local governance. In LISP participating settlements, public satisfaction with local infrastructure and services is higher compared to nonparticipating communities. The paper is based on historical data from the Government of the Russian Federation and its official statistics, World Bank and participating regions’ LISP reports, and the results of a survey of citizens’ and government representatives’ perceptions, conducted in 2017 in four regions: Tver, Kirov, Nizhny Novgorod, and Ulyanovsk oblasts6. The survey methodology is presented in Annex A. 6 The survey was conducted within the joint project of the Ministry of Finance of the Russian Federation and the World Bank’s “Strengthening Participatory Budgeting in the Russian Federation.” 12 II. Evolution of LISP The history of the Local Initiatives Support Program dates back to 2005, when it was developed by the World Bank based on the methodology that combines key principles of participatory budgeting (PB) and community-driven development, including transparency, participation, accountability, and enhanced local capacity. LISP was first launched in Stavropol Krai in 2007. Many more regions have used the LISP mechanism since then. The second region that launched LISP – Kirov Oblast – started implementation in 2010, three years after the first pilot in Stavropol Krai. Since LISP was a new instrument with little experience and evidence base, the process of engaging new regions was initially slow. Over the course of several years, the World Bank team and participants from the pilots promoted the LISP and its positive effects to the regional authorities. By 2015, LISP had been successfully launched in a dozen Russian regions. The World Bank was directly engaged in supporting LISP in eight of these regions, while in parallel some other regions had started to implement LISP-type programs on their own. The idea of citizens’ engagement in decision-making and budget planning became more and more popular, and other PB models based on different methodologies7 were developed and successfully tested in several Russian regions. At that time, it became obvious that the demand from regions for LISP-type programs was growing and that there was a need to build an institutional infrastructure that would allow the World Bank to assist interested regions in the design and launch of LISP programs. At the same time, a broad public discussion was initiated at the national level on budgetary practices involving citizens, and the term initiative budgeting was coined to signify the set of various Russia-specific participatory practices (Shulga et al. 2017). In the period 2015–16, the Russian Ministry of Finance (MoF) took note and became interested in the LISP and PB development. The MoF interest stemmed mainly from the 7 For example, a PB model developed by a design group of the European University in St. Petersburg supported by the Kudrin Fund for the Support of Civil Initiatives. 13 potential of these methods to lead to more efficient use of budget through participatory mechanisms: first, through a better focus on the issues that local populations perceived to be top priorities; and second, because of the reinforcing mechanisms of public oversight and monitoring through local participatory decision-making. MoF expressed interest in the promulgation of LISP and other PB processes across Russia. As a result, a joint project of the World Bank and the MoF “Strengthening Participatory Budgeting in the Russia Federation in 2016–20” was launched in April 2016. The main goal of this joint World Bank-MoF effort is the development of PB practices on a large scale in regions through replication of the most successful regional PB models. This is done by supporting regional pilots, strengthening the capacity of local stakeholders for implementing PB, and setting up an institutional infrastructure and a system for information exchange (including at the international level). It was expected that about 30 regions (slightly more than a third of all regions) would participate. However, demand for the project activities was so high that by the end of 2017 the number of participating regions reached 45. As of March 2019, that number had risen to 52 regions8, which represent over 60 percent of all Russian regions (see Figure 1). As a first step in the formation of the PB institutional infrastructure, a Center for Initiative Budgeting was established at the MoF’s Institute for Financial Research. In parallel, the project supported the creation of over 20 regional project centers to coordinate and support PB activities at the regional level by facilitating meetings of local communities, consulting local stakeholders at all project stages, and monitoring project implementation. Currently, LISP is the most widespread participatory model in the country. Being the first of its kind, LISP can be considered the engine of the broader expansion of PB practices across Russia. According to data from the official monitoring conducted by the MoF, 4,253 PB 8 29 out of 52 regions participating in the World Bank-MoF project implement LISP. All other 23 regions are either at the preparation or piloting phase of LISP or implement other PB models different from LISP (PORT, School PB, etc.). 14 microprojects were implemented in 33 Russian regions in 2017 (with 29 out of 33 having implemented LISP-type practices). The total budget of the Russian PB projects in 2017 was the equivalent of about USD 75 million, as compared to USD 40 million in 2015. Total cofinancing from sources other than regional budgets, such as municipal budgets, citizens, and businesses, exceeded the equivalent of USD 25 million in 2017. This amount included USD 13.5 million from municipal budgets (53 percent), USD 5.8 million from citizens (22 percent), and USD 3.5 million from business community (13 percent). Figure 1. Map of Regions Participating in “Strengthening Participatory Budgeting in the Russia Federation in 2016–20” Source: World Bank. While continuing to support the launch of LISP-type activities in more regions, the World Bank work expanded towards new participatory models. In 2018, with the support of the World Bank, the Sakhalin Oblast introduced two new PB models: Youth Budget, a PB program for school students; and Public Territorial Development (PORT), a PB practice aimed at the participatory identification of larger infrastructure projects. Both programs are based on the best practices of LISP, and on international experience. The World Bank is also working on strengthening the focus of PB practices to engage vulnerable and disadvantaged populations in the regional PB programs. 15 III. LISP design LISP Objectives and Beneficiaries The LISP approach and implementation mechanism were developed in response to the regional and local demand for a participatory framework for identifying and implementing local priority microprojects. Due to the tailored design offered to each participating region, the formulations of the LISP objectives differ slightly from region to region and do not have a universally applicable definition. Overall, however, LISP has two overarching objectives that, in varying degrees, are observed in all cases and tie in to the various LISP activities: 1) To improve access and quality of local socioeconomic infrastructure and services, and 2) To improve the communication and dialogue between citizens and local authorities. Figure 2. LISP Theory of Change The beneficiaries of LISP-based regional programs are predominantly rural communities that are vulnerable and poor and have the lowest levels of access to financing for their initiatives. These communities usually have little influence on local or regional-level decision-making. 16 Figure 2 illustrates the implicit results chain, explaining the links between LISP’s interventions, outputs, intermediate results, and desired outcomes. The derived objectives are visible at the level of the desired outcomes. On the right-hand side, the results chain also highlights longer- term outcomes to which LISP is expected to contribute. LISP Principles The LISP methodology employs the following key elements: 1. Direct and broad participation of citizens. At community meetings, the residents directly participate in identification and prioritization of LISP projects. This differs from some other formats of PB in which the selection is made by a budgetary committee or by a meeting of community delegates. The direct involvement of local residents in LISP preparation and, later, in implementation and oversight guarantees high awareness among the population of its goals, objectives, and mechanisms. This process catalyzes a rethinking of citizens’ role in local development and bolsters confidence in the existing mechanisms of local self-governance, which in turn in a positive perception of the program results among the population. 2. Competition among the municipalities for LISP resources. At the project approval stage that follows the voting procedure the regional level selection committee uses a set of formal criteria to rank the projects identified and approved by vote of the population in municipalities. These selection criteria aim to assess the actual demand for the projects and the degree of their support by the community (Box 2). It is important to note that while the competition between municipalities for LISP resources is quite strong, the majority (60 to 80 percent) of the applications submitted by municipalities, are subsequently approved to receive financing from the regional budgets. 3. Cofinancing of microprojects by local communities . LISP microprojects are expected to receive cofinancing from the community population and, optionally, from local businesses. The regional-level selection committee uses the level of community cofinancing as one of the selection criteria at the project approval stage. The main purpose of the cofinancing is not to raise additional resources but rather to improve prioritization and create ownership and the 17 right incentives for the participants (Belenchuk, Vagin, and Shulga, 2017). First, cofinancing helps to prioritize actual needs and supports those communities that are ready to contribute to solving their own problems. Second, cofinancing strengthens community ownership of the microprojects. In particular, at the implementation stage local communities take an active part in monitoring the quality and timeliness of the work done by contractors. The sense of ownership of the project by the benefitting community assures sustainability of the results after completion of the project works, including maintenance of the created infrastructure. The average level of LISP cofinancing by local communities is quite high – approximately 40 percent (an amount comprised of 21 percent from settlement budget, 11 percent from the population, and 8 percent from local businesses). For example, literature analyzing community-driven approaches (Susan Wong and Scott Guggenheim, 2018.) notes 25% cost- savings (comprised of community contributions plus procurement and other savings) as high, and the usual rates of financial contributions from the beneficiary population is rarely above 10%. Design of Regional Programs Based on LISP-type Methodology The LISP preparation and implementation stages have a clear process and sequence described in the regional operational manuals.9 At the preparatory stage, every participating region develops the details of its own regional program design and passes enabling regional regulations. The program idea and the approved parameters, principles, mechanisms, and schedule are announced to the municipal officials and experts who would be responsible for LISP implementation at the municipal level. Experts and advisers of regional project teams in charge of LISP then deliver trainings and seminars for municipal staff. Following the training, the heads of municipal entities launch the awareness campaign to announce the program in the settlements and to organize discussions with local communities on their priority projects. 9 For more detail about the LISP mechanisms and procedures as well as about the objectives of executive authorities, local self-governing bodies, and consultants, see Khachatryan et al. 2016; Dias ed. 2018; Shulga and Sukhova 2016a;. 18 This stage often involves taking opinion polls in the form of questionnaires or organizing preliminary meetings in neighborhoods or with the staff of large local businesses. At the general community/public meetings in participating municipalities (mainly small and medium-sized), direct voting by the meeting participants is used to identify and prioritize the microprojects, aimed at solving high-priority local problems. Such microprojects could include road repairs, rehabilitation or construction of water supply facilities, refurbishing community centers, landscaping and territorial improvement, as well as other matters within the competence of local authorities. Then the municipality, together with a group of civic activists (elected at the general community meetings), develops the microproject proposal to Box 2. Microproject Selection Criteria Project applications are ranked using assessment and selection criteria, which mark the extent to which a project complies with basic LISP principles as well as the level of support for the project by the community. The following are some of the main criteria used to rank proposed microprojects (the percentage in brackets provides the usual weight assigned to the specific criteria in the overall score): • Population participation in project identification (35–40 percent): looks at the share of population that participated in public discussions • Contribution of local stakeholders (35–40 percent): accounts for the contributions by population, municipal budget, private business, and other sponsors in project cofinancing (in both cash and in-kind contributions); • Socioeconomic effects (10–15 percent): estimates the share of beneficiaries as percentage of the local population, along with newly created or preserved jobs, environmental impact, availability of mechanisms, and means for effective upkeep and operation of the project; • Openness and transparency of decisions made within a project (5–10 percent): assesses various factors, including media use for informing the population. The assessment criteria are transformed into numeric values, enabling simple calculation algorithms to calculate the final score for each proposed microproject. 19 submit to the regional-level competition. The final approval of winning proposals is based on formal criteria that allow the municipality to assess the level of demand for each microproject (Box 2). Figure 3. LISP Cycle Based on the competition results, the approved project proposals receive financing from the regional budget and actual implementation starts in municipalities. At all stages of LISP implementation, program participants receive technical support from the regional project team, including awareness raising and consultations. The entire cycle of LISP implementation—from the conducting of community meetings to the hand-over of completed work—typically does not exceed one year (see Figure 3). 20 IV. Data Sources and Methodology The findings presented in the paper are based on three main sources of data: the data on microprojects carried out through LISP which was provided by MoF and the regions implementing the program; quantitative survey measuring the effects and outcomes of LISP; and а supplementary qualitative survey of beneficiaries and local authorities. Data on microprojects. This paper examines several types of microprojects and looks at both the number of microprojects and the number of their beneficiaries in order to outline the contributions LISP programs have made to small-scale local infrastructure and services. This data is provided by both MoF and regional monitoring of LISP implementation and provides the basis for quantitative survey sampling, based on the number of projects implemented and their typology. There is a two-level monitoring system in place. First, Russia MoF collects data on regional PB programs (type of PB practice, amount of money allocated, number of participants etc.) based on the questionnaires filled in by regional ministries responsible for implementing PB. This system was designed under the World Bank-MoF project on strengthening PB in Russia. Second, some LISP-implementing regions has MISs put in place to collect data on participatory process (including number of participants of community meetings and preliminary hearings, types of specific projects discussed and voted, number of beneficiaries, co-financing, etc.) and project implementation status. Quantitative survey on outcomes and effects of LISP. The quantitative survey is the main instrument informing this paper on beneficiaries’ perception of LISP effects and outcomes. It was a one-time survey that took place in November-December 2017. The survey was conducted by the Center of Sociological Research of the Russian Presidential Academy of National Economy and Public Administration (also responsible for running the Russian 21 branch of Eurobarometer10, which is a survey conducted by request of European Commission in each EC country since 1974 with the aim to monitor social, economic and political values, attitudes and practices of population). The survey sampling was done in three steps. As the first step, regions with longer continuous experience of LISP were selected, as these were the places where the social effects of the program were most likely to be visible to the population. The regions were Kirov Oblast (7 years of experience), Nizhny Novgorod Oblast (4 years), Tver Oblast (4 years), and Ulyanovsk Oblast (3 years). The second step included the selection of the settlements. The objective of the comparison was to trace the potential effects of LISP and its participatory design11. Two groups of settlements were chosen in the four regions: the treatment group (51 settlements overall that had participated in LISP) and the comparison group (6 settlements overall that never participated in LISP). Settlement selection in the treatment group was randomized. The sample included only settlements with no more than 5,000 inhabitants, where the effects of LISP on both settlement and population level could be expected to be most pronounced. The third step was the selection of respondents. This selection was randomized, with quotas on gender and age groups that corresponded to regional group means. The number of respondents in the treatment group was 2,010 and in the comparison group – 500. The information was collected through face-to-face interviews with the respondents. Individual sampling results are summarized in Annex A. The settlements in the comparison group were matched to the settlements in the treatment group using two criteria: population numbers, and overall budget size (settlement characteristics and matching methodology are described in Annex A), to reduce between- 10 See details on Eurobarometer in Russia here: https://www.ranepa.ru/nauka-i-konsalting/strategii-i- doklady/evrobarometr/evrobarometr 11 The purpose of the research is to test whether the LISP, by allowing the participatory decision making, helps municipalities identify their binding constraints achieving our outcomes of interest, regardless of how they use the funding. The focus of the survey is single treatment (LISP) rather than each type of investment. There is heterogeneity at the municipal level regarding what that investment looks like. 22 settlement variation. As a result, the settlements in the two groups are very similar to each other by size of the population and by size of settlement budget, which suggests they had equal opportunities in the provision of public goods at the time of the study. While the chosen settlements without LISP may have been implementing other regional or national program targeting basic infrastructure, the same was also true for the settlements with LISP, which makes the comparison scientifically robust. The quantitative survey data is used to determine levels of public confidence, satisfaction with the quality of public infrastructure, evaluations of both accountability and approachability of the local government and other variables on the individual level ( see Annex A, Table A-6 for details on sample statistics). Supplementary qualitative survey of the beneficiaries and local authorities. The supplementary qualitative survey was a one-time survey that took place between December 2016 and May 2017 and was conducted by regional teams of sociologists under the supervision of the authors. The survey was conducted in 4 regions with the longest continuous experience in LISP: Stavropol Krai, Kirov Oblast, Tver Oblast, and the Republic of Bashkortostan. The qualitative data summarizes 37 semi-structured interviews with local authorities (heads of the settlements and municipal districts) and 12 focus groups with the citizens. The sample of the interviews included four different municipal districts in each region, relatively equally experienced in LISP (with participation span of 1-2 years) and 2-3 settlements in each municipal district. For each municipal district, 3-4 semi-structured interviews were taken (one with the head of municipal district and 2-3 with the heads of the settlements in each municipal district). The sample for focus groups included citizens from four municipal district per region. The qualitative survey provides information on the beneficiaries’ opinions on LISP -related improvements, which supplements the overall picture of the LISP effects. To describe individual evaluations of LISP’s contributions to public confidence and effectiv e communication, both historical data and individual reflections are available from local 23 beneficiary interviews and focus groups. For the purposes of the analysis, this paper also looks at beneficiaries’ opinions of local authorities, and beneficiaries’ p erceptions of improvements in the local quality of life. Limitations of the Research The research applies to small rural settlements (with fewer than 5,000 inhabitants) and does not cover larger settlements or urban areas. Its results are based on a one-time survey that is not nationally representative, which limits the strength of the direct causal inferences. Thus, it is important to note that the survey is not an impact evaluation study of LISP, but rather a perceptions survey. This survey allows us to make some comparisons of outcomes between the treatment and comparison groups, and to estimate LISP’s likely contribution to these outcomes. The survey methodology also considered possible problems related to self-selection bias, since in theory settlements with higher preexisting levels of public confidence might be more likely to participate in LISP than those with lower levels of public confidence. However, the analysis demonstrated that such bias was unlikely, and that, on the contrary, low levels of public confidence seemed to trigger a higher probability of participation (see Annex A, Section “Self-selection analysis”). 24 VOICES OF THE PEOPLE V. LISP Results “That road is pure happiness!” – A A. Improving Local Infrastructure and resident Services LISP is regarded by both residents and local “Probably, there is nothing else that authorities as a contributor to improving gives us so much pleasure as this access to and the quality of the local community center. It was freezing there socioeconomic infrastructure and services in before, and we didn’t want to go there. participating settlements. In order to provide a Floor boards were falling through. The picture of the processes and achievements in roof leaked. Now the center is open day this area, the paper will look at the program’s and night. We are crazy about that outputs and the opinion of beneficiaries place. We all keep running there. It’s concerning these improvements, as well as at warm and beautiful!” –A resident participation rates and coverage. “...We now have a park ...our settlement is the only one in the district having such Volume of the Provided Support a park. People are taking walks in the As explained in Section 2, the number of evenings .... Earlier there were no places regions implementing LISP and other PB to go in the village ... now kids go to practices and the number of microprojects school and after school they go to the have increased gradually since 2007. Starting park. They all have roller skates now, from 26 projects in one region (Stavropol Krai) many kids skate in the park, and it’s so in 2007, in 2017 there were already over 4,253 pleasant to see how they spend their local participatory microprojects being time. It’s something we haven’t had implemented in 33 regions within the before – I mean roller skating, it’s kind framework of 38 participatory practices. In of in fashion now among kids …” A 2017, the overwhelming majority of all resident implemented microprojects (82 percent) were Note: The source for this and all other “Voices of the People” boxes is the qualitative survey completed under LISP-type schemes (3,504 conducted in 2017 in four regions of Russia. 25 out of 4,253 projects). A total of 33 PB programs (which represents 76 percent of all existing PB initiatives) are LISP programs that have been launched and successfully implemented in 29 Russian regions (see Figures 4 and 5). Figure 4. Number of Russian Regions Implementing PB, 2007–17 35 33 30 25 20 17 18 15 10 6 4 5 1 1 1 2 2 2 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Source: For 2007–14, World Bank data on 8 LISP regions; for 2015 –17, data from the Ministry of Finance of the Russian Federation, official annual survey from PB implementation regions (including World Bank LISP regions). Figure 5. Number of Implemented PB Projects, 2007–17 4500 4253 4000 3500 3000 3000 2657 2500 2000 1500 1000 793 296 444 500 43 65 228 26 29 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Source: For 2007 – 14, World Bank data on 8 LISP regions; for 2015 and 2017, data from the Ministry of Finance of Russian Federation, official annual survey from PB implementation regions (including World Bank LISP regions); for 2016, World Bank estimates. Beneficiaries and Microprojects From the beginning of the LISP program in Russia until 2017, about 12,000 microprojects with more than 7.5 million beneficiaries were implemented. Kirov Oblast is the leader among 26 regions in terms of the annual number of beneficiaries. In 2017, 820,000 of its citizens benefitted from LISP-related activities, which represents 65 percent of the total population living in Kirov Oblast.12 The typology of microprojects delivered within the LISP scheme shows that LISP can deliver a variety of types of social infrastructure needed by the local population, thus increasing the availability of infrastructure and services for local communities. Small-roads microprojects made up the largest number, followed by education, water supply facilities, and community recreation areas (see Figure 6). Figure 6. Typology of LISP Projects in 29 Regions of Russia, 2017 (percent of the total number of projects) Education 12% 18% Road repairs 5% Rehabilitation or construction of water supply facilities 5% Recreation facilities 7% Children playgrounds 18% Sport facilities 9% Public lighting 12% Cemeteries ground improvement 15% Other Source: Data from the Ministry of Finance of Russian Federation, official annual survey from PB implementation regions (including World Bank LISP regions). In the course of several years, the projects delivered as part of LISP have been able to transform the public environment in participating settlements, improving the availability of quality social infrastructure in municipal entities and in the entire region. For example, 875 rural roads and 520 water supply facilities were repaired from 2010 to 2018 in Kirov Oblast 12 Based on data from the Ministry of Finance of the Russian Federation, official annual survey. 27 alone. As of 2017, more than 200 community centers were renovated in Tver Oblast and Stavropol Krai. Because LISP targets the most acute issues faced by settlements, it is highly appreciated by the local population, an overwhelming majority of whom believe that LISP microprojects are very important and useful exactly because of their direct practical benefits (see Figure 7 for a breakdown of local opinion from Kirov Oblast). Figure 7. Public Opinion about LISP Results in 2013 by Residents of Kirov Oblast Question: “Do you use the results of the Question: “How important is the problem project (works) completed?” (percent of addressed by LISP?” (percent of respondents) respondents) 0% 20% 40% 60% Very important 55% Yes 82% Important 35% Not very important 1% No 18% Not important at all 0% No answer 8% 0% 20% 40% 60% 80% 100% Source: Sociological survey in the Kirov Oblast regarding LISP in 2013. Beneficiary satisfaction with local infrastructure and services. Citizens in communities that participate in LISP-type programs are more satisfied with local socioeconomic infrastructure and services than citizens in nonparticipating communities. In the conducted study, 60 percent of the respondents from the treatment group reported experiencing improvement in the water supply facilities, against 52 percent of respondents from the comparison group. Similarly, 28 percent reported improvement in the roads (versus 21 percent from the comparison group); 45 percent noted improvement in community centers (versus 20 percent from the comparison group); and 44 percent noted improvement in sports facilities (versus 21 percent from the comparison group) (Figure 8). 28 Figure 8. Treatment and comparison group responses to the question, “As of today, how do you estimate … the following aspects in your settlement”? (“Good” and “Very good” responses) Сondition of roads and streets 65% Water supply 60% 30% 28% 55% 52% 25% 45% 28% 23% 9% 21% 35% 20% 4% 25% 33% 30% 19% 16% 15% 15% comparison group treatment group comparison group treatment group Good Very good Good Very good Note: Chi-squared value: 14.678 with p-value = 0.005; Note: Chi-squared value: 20.914 with p-value = Mann-Whitney U value: 463180 with p-value = 0.018 0.000; Mann-Whitney U value: 446500 with p- value = 0.017 Condition of the cultural Condition of sport facilities 60% institutions 50% 45% 44 40% % 40% 21% 18% 30% 20% 21 20% 20% % 7% 6% 27% 10% 23% 14% 14% 0% 0% comparison group treatment group comparison group treatment group Good Very good Good Very good Note: Chi-squared value: 133.38 with p-value = 0.000; Note: Chi-squared value: 84.082 with p-value = Mann-Whitney U value: 261730 with p-value = 0.000 0.000; Mann-Whitney U value: 217130 with p- value = 0.000 Source: World Bank, based on Sociological survey in Russian Regions to measure social effects of Participatory Budgeting, 2017. These positive effects can be associated with LISP with some level of confidence, if properly combined with the types of project implemented in the settlements that participated in LISP. (The supplemental analysis is presented in Annex B, Section 3. ‘Linkages to Beneficiary Satisfaction with Local Infrastructure and Services.’). 29 In summary, we see that when the number of LISP projects implemented in a settlement increases, the satisfaction levels of local people with certain types of public infrastructure (roads, cultural institutions, and sport facilities) also increases. This indicates that the results of these projects are likely more visible to the inhabitants of the settlements than other types of local infrastructure (for example, water supply and communal services). Figure 9. “Do you think the quality of life Figure 10. “To what extent do you like to in your settlement during the recent 3 live in your village?” (percent of years has improved, has deteriorated, or respondents) remained unchanged”? (percent of respondents) 60% 90% 82% 50% 80% 75% 70% 40% 60% 47% 30% 50% 48% 51% 45% 40% 20% 34% 30% 28% 10% 21% 21% 20% 34% 28% 0% 10% comparison group treatment group 0% comparison group treatment group Has improved Remained unchanged Absolutely love Somewhat love Note (for “has improved”): Chi-squared value: Note: Chi-squared value: 13.942 with p-value = 34.005 with p-value = 0.000; Mann-Whitney U 0.003; Mann-Whitney U value: 442780 with p-value = value: 550920 with p-value = 0.000. 0.000. Source: World Bank, based on a sociological survey in the Russian regions to measure the social effects of participatory Budgeting, 2017. Moreover, although the survey methodology does not allow for full attribution, there is a visible correlation between the implementation of LISP-type subprojects and the positive opinion of the respondents from the treatment group concerning quality-of-life matters. For example, in LISP-participating locations, more citizens report that the quality of life in their settlements has improved (Figure 9) and that on the whole they like living in their villages 30 better (Figure 10). The rate of satisfaction with municipal services is also noticeably higher in participating settlements (Figure 11). Figure 11. “Are you satisfied with the way the issues of territorial improvement, and housing and communal services are being solved in your settlement?” (percent of respondents) 50% 46% 45% 36% 40% 35% 30% 35% 25% 20% 29% 15% 10% 5% 11% 7% 0% comparison group treatment group Completely satisfied Somewhat satisfied Source: World Bank, based on a sociological survey in the Russian regions to measure the social effects of participatory Budgeting, 2017. Note: Chi-squared value: 34.005 with p-value = 0.000; Mann-Whitney U value: 550920 with p-value = 0.000. The settlements and citizens, therefore, assess positively the approach used by LISP, seeing it as directly improving the quality of local infrastructure, and of the quality of living in their settlements. 31 B. Improving Communications and Dialogue The second important LISP contribution perceived by beneficiaries and local authorities is the improved communication and dialogue VOICES OF THE PEOPLE between citizens and local authorities. Local authorities often consider this to be the most “This is a mechanism of important objective of the program. communication enabling us to talk to A Platform for Community Participation people and receive feedback.” -- Local official The LISP approach provides a platform for community participation and buy-in. Every “People are not so resentful about the year local communities hold thousands of existing situation. People realize that LISP-related general meetings, which are they are consulted with.” -- Resident attended by more than a million people in “…residents begin to be concerned, total. Even more citizens are involved in thinking that if we have done preliminary events to discuss projects. These something, if we have contributed our include public opinion polls, street and own money, if we have given neighborhood meetings, and other meetings assistance to have something done, in small groups that involve up to 70 percent then we are to look after what has of the adult population of participating been produced.” -- Local official settlements. This direct participation of residents in determining priority problems is the key element of the program and may change people’s attitude towards their own role in territorial development and also likely influences their confidence in local self-governance. The population is involved in LISP projects not only at the stage of prioritization and selection, but also project delivery. For example, citizens provide voluntary unskilled labor for demolition, clean certain areas of construction waste, paint fences, or provide resources free of charge (such as electricity, building materials, and machinery). Each of the participating 32 municipalities also creates groups of local volunteers who are part of project preparation, implementation, and monitoring. Broad citizen involvement in these activities makes LISP well-known among the population. In the Kirov Oblast, LISP was familiar to 50 percent of the interviewed population, who could explain what LISP is about, while only 20 percent were informed about the second most well- known regional program (Table 1). In Tver Oblast, total awareness of LISP is somewhat lower than in Kirov Oblast due to the shorter duration of the program there – with one third (34 percent) of the population knowing about it. Still again, this is significantly higher than other currently or previously operating programs in this region, for which the public has levels of awareness of no more than 14 percent. Table 1. Level of population awareness of LISP and other regional programs (percent) Comparison Treatment Comparison Treatment Kirov Oblast Total Tver Oblast Total group group group group “Ensuring the safety "Youth of the and livelihoods of Upper Volga" for 18 23 17 14 10 15 the population of 2013-2018 the Kirov Oblast” “Local Initiatives “Local Initiatives Support Program 50 32 53 Support Program 34 20 37 (LISP)” (LISP)” “Development of “Supporting the transport program for system for 2013- employees 20 21 20 6 3 6 2020” engaged in the traditional sector of economics” “Supporting "Forestry of the program for region" for 2017 – employees engaged 2022 10 8 10 12 6 13 in the traditional sector of economics” I have heard about I have heard none of the above- about none of the 39 53 37 59 71 57 mentioned above-mentioned programs programs Source: World Bank, based on a sociological survey in the Russian regions to measure the social effects of participatory budgeting, 2017. 33 LISP/PB Practices That Improve Formal Interaction In the LISP-participating areas, the population participates more frequently in meetings and community consultations where issues and local problems are discussed with the local officials (Figure 12), and their residents have more access to meet and directly discuss issues with a representative of the settlement administration, if necessary (Figure 13). Dialogue That Leads to Practical Action In LISP-participating locations, the administration more frequently includes the population’s opinion in decisions on how to solve local problems and on prioritizing the timing for various interventions compared to the administrations in nonparticipating locations (Figure 14). Survey respondents also feel community meetings are an effective platform for solving issues (Figure 15). Figure 12. “Does the administration of your Figure 13. “Is it true that it is possible to settlement hold community meetings with meet with the representatives of the residents to discuss the local issues?” settlement administration, if needed?” (Percent of responses “Yes”) (Percent of responses “Yes”) 90% 80% 88% 70% 86% 60% 84% 50% 82% 88% 40% 75% 30% 57% 80% 81% 20% 78% 10% 76% 0% comparison group treatment group comparison group treatment group Source: World Bank, based on a sociological survey in Source: World Bank, based on a sociological survey in the Russian regions to measure the social effects of the Russian regions to measure the social effects of participatory budgeting, 2017. participatory budgeting, 2017. Note: Chi-squared value: 15.722 with p-value = 0.000; Note: Chi-squared value: 60.249 with p-value = 0.000; Mann-Whitney U value: 371670 with p-value = 0.000. Mann-Whitney U value: 413420 with p-value = 0.000. 34 Figure 14. “Does the administration of your Figure 15. “To what extent do community settlement take into account the opinion of meetings in your settlement allow you to the residents in solving local problems?” solve the issues being raised?” (“Allow” (Percent of responses “Yes”) responses, in percent of respondents) 50% 40% 40% 30% 30% 48% 20% 39% 20% 33% 30% 10% 10% 0% 0% comparison group treatment group comparison group treatment group Source: World Bank, based on a sociological survey in Source: World Bank, based on a sociological survey the Russian regions to measure the social effects of in the Russian regions to measure the social effects participatory budgeting, 2017. of participatory budgeting, 2017. Note: Chi-squared value: 34.19 with p-value = 0.000; Note: Chi-squared value: 23.801 with p-value = Mann-Whitney U value: 428830 with p-value = 0.000. 0.000; Mann-Whitney U value: 357540 with p-value = 0.000. 35 C. Contributing to Public Confidence VOICES OF THE PEOPLE The survey work suggests that the increased satisfaction of the population “The program is uniting people.” -- Resident with local infrastructure and services “The village is reviving! Life is boring here, and the broadened dialogue may change and now we have such a project!” -- Resident the mindsets of the people and generate “There is more trust in government and now proactivity and commitment. These are people are not as bitter as they were about likely contributors to a broader the existing situation when there is no objective, which is beyond LISP: to money.” -- Resident increase confidence between citizens “When a person, well, residents see that and authorities and to strengthen public someone is doing his/her best, that we confidence in the potential for positive participate in programs, win in competitions change. – the attitude certainly changes greatly.” -- Local official Although the goal of the LISP “Results are delivered, and every year people methodology is not the improvement of have better understanding and receive us local governance per se, LISP-based better. They find out themselves what events programs provide an opportunity for are held and when, that is, the general public participation in the decision- attitude of the population has changed.” -- making process, making it more Local official transparent to people. By giving people “People begin to be concerned, thinking that the right to decide what are the most if we have done something, if we have important needs of their settlement, the contributed our own money, if we have given LISP programs may also contribute to assistance to have something done, then we improved local governance. This possible are to look after what has been effect is observed in the survey results: if produced…”—Resident citizens are involved continuously, their 36 evaluations, in the form of public confidence and satisfaction, become more positive over time. Increased Public Confidence in Local Authorities The survey data shows correlation between c implementation of LISP and changes in public confidence indicators at the settlement level. As shown earlier, the most vivid change is the respondents’ perception of improved quality of the provision of public goods (see Figure 8). However, the effects of LISP spread beyond the purely institutional framework and its infrastructural dimension. The survey reported that LISP-type programs implemented for a long period (two or more years) affect the views and attitudes of the citizens toward local governance. For example, the survey results show that those heads of settlements who ensure that residents are engaged in discussions and resolution of local issues under the LISP-type programs enjoy greater citizen confidence, with an average score of 67% confidence level for the LISP participating communities (see Figure 16, graph on the left). These differences are striking if they are compared to the Russian average score of 52%13 public confidence in rural settlements’ authorities (using the same scale, which was provided by the Eurobarometer survey in spring 2017). The level of public confidence in local governance is positively associated with the duration of exposure to LISP. The results suggest that each implemented LISP project gradually leads to higher levels of public confidence in the head of a settlement. The level of confidence in public authorities in settlements that implemented four or more projects is above 80%, compared to 64-65% confidence levels in the localities that had 2-3 projects, and only 58% for those that had zero or one project (Figure 16, graph on the right). 13 Authors’ calculations based on the data of Russia Eurobarometer flash Survey conducted in 2017, overall sample consists of 6088 respondents. However, the sample of rural inhabitants was used for the comparison (1391 respondent). 37 Figure 16. Percent of citizens reporting “absolute” or “some” public confidence in the head of their settlement 80 90 80.4 67 80 70 59 70 65.2 64.5 60 52 57.9 60 50 50 40 40 30 30 20 20 10 10 0 0 no 2 3 4 projects projects projects projects Russian average Treatment group Comparison group and more Source: World Bank, based on sociological survey in Russian regions to measure social effects of participatory budgeting, 2017; for Russian averages, Eurobarometer omnibus survey, 2017. Note: Lefthand graph: For binary public confidence in survey data chi-squared value: 8.3799 with p-value = 0.004; Mann-Whitney U value: 365560 with p-value = 0.003. Righthand graph: Chi-squared value: 14.409 with p-value = 0.001; Kruskal-Wallis chi-squared value: 14.403 with p-value = 0.001. Since in most participating regions (and in all of the surveyed regions) each participating settlement is allowed to submit to the regional LISP competition one microproject proposal annually, the number of implemented microprojects for a given settlement is directly linked to the number of years that a settlement has participated in LISP. Although there may be other factors at play, the increase in the number of LISP cycles (and microprojects) in which a settlement participated is still strongly correlated with a higher level of public confidence in the local government. Accountability and Responsibility As an element of improved public confidence, LISP communities demonstrate a stronger commitment to common goals and to efficiency improvements. For example, in the LISP communities, 29 percent of respondents declared that they would invest personal money in a public project that would not directly benefit them, as opposed to only 15 percent of respondents in non-LISP communities giving a similar response (see Figure 17). When the question was changed to refer to efforts and time, rather than money, 46 percent of 38 respondents in LISP communities declared they were ready to invest, compared to 42 percent of respondents in non-LISPS communities. Figure 17. Readiness to invest personal Figure 18. How much responsibility do money, time and efforts to the public and you feel for what is happening in your charity projects, percent14 settlement 50% 42% 46% 46% 40% 60% 49% 40% 34% 42% 29% 30% 25% 40% 33% 30% 20% 15% 10% 20% 0% comparison treatment comparison treatment 0% group group group group comparison group treatment group Public project Charity project Ready to invest money Ready to invest time and efforts Feel no responsibility Feel responsible Source: World Bank, based on sociological survey in Source: World Bank, based on sociological survey Russian regions to measure social effects of participatory in Russian regions to measure social effects of budgeting, 2017 participatory budgeting, 2017 Note: Public project, money: chi-squared value: 39.681 Note: Сhi-squared value: 3.8094 with p-value = with p-value = 0.000; Mann-Whitney U value: 432130 0.051; Mann-Whitney U value: 255250 with p- with p-value = 0.000. Public project, labor: chi-squared value = 0.045. value: 2.3323 with p-value = 0.1267; Mann-Whitney U value: 482800 with p-value = 0.115. Charity project, money: chi-squared value: 13.979 with p-value = 0.000; Mann-Whitney U value: 458120 with p-value = 0.000. Charity project, labor: chi-squared value: 4.1292 with p- value = 0.042; Mann-Whitney U value: 476510 with p- value = 0.037. LISP communities also display higher willingness to contribute to charity, with 80 percent of the respondents declaring readiness to contribute, versus 65 percent in non-LISP communities. People in LISP communities also feel higher direct responsibility for solving the problems in their own town or village. In non-participating communities a whole 49 percent of the people felt no responsibility whatsoever about the settlement developments, and 30 14 Questions 24 and 25 in the questionnaire read as follows: “Are you ready to invest personal finance, time and efforts into a charity project (helping orphans, the elderly, people with diseases), which will not bring you benefit, but will be helpful to society, that is mainly to people you are not familiar with?”; and “Are you ready to invest personal finance, time and efforts into a social project (outdoors improvement, road construction, organization of social and cultural events), which will not bring you benefit, but will be helpful to society, that is mainly to people you are not familiar with?” 39 percent felt some responsibility. In LISP communities, respondents who stated that they were indifferent made up only 42 percent, and a higher 33 percent felt responsible (Figure 18). Government Perceptions of the Population Although the study focused more on the public confidence of the population in authorities, and the attitudes of authorities toward the population were less of a focus, the information collected revealed interesting positive tendencies in both directions. The increased commitment of the citizens towards local matters and priorities seems to also increase the positive attitudes of the local authorities towards their own constituency. As explained in Section 5B (see Figures 12, 13, 14 and 15), the increase in citizens’ engagement likely correlates to a change in the perceptions of local officials, who also show more respect for citizens’ demands and begin to rely more on their constituency for support in choosing and implementing local projects. This is confirmed by the supplementary qualitative research, where heads of settlements clearly stated their appreciation and trust in the participation of the citizens in selection and monitoring of microprojects. 40 VI. Lessons on Sustainability and Institution Building A. Sustainability of Process Initiative budgeting has become a policy priority at national (federal) level. As a result of active work and gained positive experience, the development of IB was defined as one of the priorities in the document Main Directions for Action of the Government of the Russian Federation till 2024 (published September 29, 2018). The document sets a target: that by the year 2024, 50 percent of the regions would develop and approve regional development strategies for initiative budgeting. Another important national document – the Concept Paper on Increasing Efficiency of Budget Expenditures for the Period 2019-2024 (approved by the Government of the Russian Federation on January 31, 2019) also states the need for developing mechanisms for citizen participation (on the basis of IB) as a way to finding solutions for social and economic development, and for the dissemination of regional and municipal IB practices. Changes to the Budget Code and the Law on Local Self-governance are also under preparation, to improve the regulations on co-financing initiatives of the population. Managing IB on various levels establishes it as a long-term agenda. Sustainable multi-level structures are being formed in Russia to manage IB. As a result of the LISP and IB initiatives, the Government established a National Center on Initiative Budgeting. At regional level, more than 20 project support centers were also formed, who provide consulting support to the regional authorities in the development and implementation of regional IB programs. A national system for monitoring initiative budgeting is a prerequisite for sustainability. In Russia, such a system has already been established. All IB processes and practices and legal changes are monitored by the Ministry of Finance of the Russian Federation, which every year prepares a report on the best regional practices of IB. Annually, MoF collects from regions data on parameters and financing for IB-type programs. In 2018, 8 of the 11 best practices detailed in this regular report are based on LISP. 41 Embedding IB programs in national systems makes them stable and predictable. LISP-based programs are fully embedded and are managed within the Russian administrative system. Every region implementing LISP established the needed regional-level regulatory framework, containing the rules for participation, and the criteria for microproject selection. Some regions adopted regional IB laws that set up the minimum annual investment financing amounts for IB projects. At the regional level, the programs are carried out by the competent executive authorities (normally by regional finance ministries), with local self-governing bodies being hands-on implementers. The programs’ financing flows in strict compliance with the national Budget Code, within the Russian budget system. The procurement of works and services within LISP programs is also carried out in accordance with the state procurement system. The procedures for program implementation, competitive selection, and distribution of subsidies among municipalities are regulated by every region through regional government resolutions. As a result, all (currently 29) LISP-type regional programs that were developed with or without World Bank support, are fully financed by the regular regional budgets. Improved access to infrastructure and improved local dialogue, together with increased confidence among stakeholders, are likely to result in positive and sustainable institutional changes. Since its inception, LISP has acted as a direct supporter or catalyst for the development of all the institutions (at any level of government) responsible for the financing of community needs through participatory budgeting approaches. Thus, the development of LISP stimulated federal policies to start supporting open budgets and citizen participation, which has led to the dissemination of regional participatory programs across Russia. LISP created a nationwide momentum for the promotion of PB approaches to solving local social problems. Starting with just one small pilot in Stavropol Krai in 2007, today LISP is a methodology used by regional programs operating across Russia. Its expansion is primarily driven by the interest of regions and communities, without any pressure from the national level. 42 B. Sustainability of Microprojects The established LISP mechanism guarantees follow-up budget and maintenance to the implemented microprojects. As LISP only deals with infrastructure owned by municipalities, municipalities are responsible by law to finance the maintenance of the created assets. There are specific budget lines in the municipal budgets to finance operational costs and maintenance. In all regional LISP programs, at the time of application, municipalities are requested to formally guarantee the maintenance of the LISP assets and to confirm in the application forms that funds have been secured for operational costs. Regional teams (officials and consultants) conduct physical monitoring of LISP assets to check their sustainability. LISP implementation builds and strengthens the local capacity for project implementation. LISP-based programs build the capacity of participating authorities at all levels. The full integration of LISP-based programs in national systems allows the development of existing national public institutions and municipal administrations, as well as strengthening of the executive branch and the local self-governing bodies to engage in hands-on work. At local level, LISP does not only build public confidence but also helps local government institutions develop the technical rapport, experience, and capacity to enter into a productive relationship with the local population. The groups of volunteers in each municipality are also valuable community assets. Passing through several LISP implementation cycles, they become well equipped for the delivery of any other project based on local initiatives. In addition, all participating stakeholders – including the project implementation group, as well as the municipal staff and the citizens’ initiative groups – receive training, regular consulting support, and technical assistance. This support is provided by experts from the higher levels of government, or by contracted specialists in the specialized units. C. Conclusion LISP-type programs have the potential to thrive and survive through political changes. LISP has already built significant experience through a critical mass of regions implementing the 43 program over many years. The program has already lived through changes of regional governments and has proven its sustainability through political changes. For example, Stavropol krai implements LISP from 2007 with the Bank, and from 2012 without the Bank, and while four regional governors changed during the period, LISP is still a priority project in the region. The year 2017 was officially announced in the region as “a year of LISP”. The LISP mechanism provides leverage for the institutions of local governance and creates productive relationships between those institutions and the local population. On one hand, local government institutions are the point of contact for the population’s needs and complaints. It is part of their prerogatives to find solutions to the population’s concrete problems at the level of the settlement, or even at the level of a specific neighborhood. On the other hand, under the regular system of financing, local authorities have little leverage over the flows of funds and the purposes the funds are used for. The national subsidies are channeled downstream with specific earmarked purposes attached to them, most often addressing infrastructure needs of national and regional importance. With a LISP program in place, local authorities finally have a mechanism to provide funding for those specific needs of the population for which the population holds them accountable. Relatively small overall volume of financing is able to influence and solve local socioeconomic problems. LISP allows for the prioritization and implementation of activities that address the highest-priority problems in disadvantaged communities. For many municipalities, a regional LISP-based program is the only way to solve social and small- infrastructure problems that had not been addressed for years. The amount of financing per microproject, albeit nominally small, is in many cases quite sufficient, as the priority issues can be resolved through repair and maintenance works that are less expensive than new construction. From the point of view of the local municipalities (as opposed to regions), the LISP financing is quite relevant, especially if compared to the regional financing that goes to municipal budgets. For example, in the Kirov Oblast in 2015, the LISP subsidies to municipalities on average amounted to 15 percent of the annual municipal budgets. While 44 the average subsidy received by municipalities is the equivalent of USD 15,400, the average annual municipal budget in LISP-participating municipalities was USD 105,000. The nationwide implementation of the LISP approach, however, also highlighted some challenges that need to be addressed in the future. As the LISP approach became more and more popular, receiving attention from federal authorities (including the Ministry of Finance), some regions started reporting activities as participatory, but in actuality some activities did not entail broad participation of the population, and transparency—a hallmark of LISP—was lacking. Enhanced public information campaigns to attract public interest in the topic, as well as the involvement of the academic community, including in the monitoring and impact evaluation of regional programs, would help address this issue. Another challenge has been that some regions have tried to save on implementation support, which, based on experience, is critical to the success of the LISP-type approach. Some regions have reduced the number and funding for activities related to training and consultation with participants, facilitation of public discussions, and monitoring and analysis of the results. This may negatively impact the quality of the process at the community level – mainly inclusion and transparency. This, in turn, could result in reduced trust in the program and lower impact. And finally, the tendency of local legislature to start overregulating the program is a risk. Certain regionally-imposed requirements (for example, as regards required community co- financing, or quantitative requirements on participation) could lead to reduced impact or could even block a regional program. Broader inclusion of the public and technical experts in the drafting of regional legal and regulatory documents could help address this issue It is clear that LISP-type mechanisms could be further developed to ensure inclusion of underrepresented social groups. Although the involvement of the population in the discussion of local development problems is quite high under a LISP-type program, more can be done to make them fully inclusive. Further work is needed to analyze the profiles of the 45 participants and beneficiaries, and to ensure the inclusion of special groups: for example, youth, students, vulnerable groups, people with disabilities and health problems, etc. The World Bank is actively working with regions to enhance the design of initiative budgeting (including LISP), to include the level of social inclusion. 46 Annex A: Survey Methodology Sampling The survey was undertaken in one round and took place in November and December of 2017. It was conducted by the Center of Sociological Research of the Russian Presidential Academy of National Economy and Public Administration. This center is also responsible for running the Russian branch of Eurobarometer. Survey sampling included three steps. The first step was to select the regions with relatively high experience in LISP in order to be able to trace the social effects, which could have taken several years to develop. These regions were Kirov Oblast (7 years of experience), Tver Oblast (4 years), Nizhny Novgorod Oblast (4 years), and Ulyanovsk Oblast (3 years).15 The second step included the selection of the actual settlements. Two groups of settlements were taken: a treatment group of settlements that had participated in LISP, and a comparison group of non-participants. The sample included only settlements with 5,000 inhabitants or fewer. With these two criteria, the treatment sample included 51 settlements and is summarized in Table A-1. Table A-1. Treatment group, settlement sample Region Number of settlements below 5,000 Number of settlements in inhabitants that participated in LISP the sample Tver Oblast 123 13 Kirov Oblast 233 12 Nizhny Novgorod 215 13 Oblast Ulyanovsk Oblast 75 13 15 One of the most experienced regions in LISP (Stavropol Krai) was not included due to some significant objective differences difference from the others. Stavropol Krai is a typical southern rural region, with types of settlements and specific local communities that differ from the other, more industrialized Northern or Central Russian regions. Moreover, LISP in Stavropol Krai initially targeted only the Eastern part of the region, which created significant intra-regional variance. 47 The settlements were selected randomly from those that fulfilled the criteria.16 For the comparison group, six settlements in the same regions were selected. The settlements from the comparison group had to be relatively close to the settlements from the treatment group based on two criteria: population, and overall settlement budget size. Thus, comparison settlements were relatively similar in size to the treatment ones, with little or no differences in their financial capacity. Also, the settlements in the comparison group had to be of the same type as their equivalents in the treatment group.17 The settlements in the comparison group were selected by a propensity score matching procedure developed by Donald Rubin (Rubin, 1973). Initially, the group means were calculated for both regional groups (Tver/Kirov regions and Nizhny Novgorod/Ulyanovsk regions, see Table A-2). Table A-2. Mean settlement populations and annual budgets, by regional group, 2015 Mean Mean annual budget, Regional group Settlement type population, 2015 2015 (in thousands of rubles) Village (derevnya) 524.42 2,432.84 Tver and Kirov regions Village (selo) 1,735.95 7,304.33 Township (poselok, 1,811.8 7,327.55 poselok gorodskogo tipa) Village (derevnya) Nizhny Novgorod and 1,525.75 5,867.85 Village (selo) Ulyanovsk regions 2,042.14 6,005.16 Township (poselok, 2,115.1 6,436.21 poselok gorodskogo tipa) Source: Population and budget data from Rosstat, 2015. 16 Several replacements were made for logistical reasons. Overall, 4 settlements were replaced in the Kirov Oblast, 2 additional settlements were used in the Tver Oblast, and 1 additional settlement was used in the Ulyanovsk Oblast. The logistic reasons are either distant geographical location (some of the settlements are hard to reach during winter season) or insufficient population (in some villages the inhabitants move to other places during winter and thus, the additional settlements had to be included in the sample to keep up with the number of respondents). 17 Settlement types include derevnya (relatively small rural settlement), selo (larger rural settlement) and poselok (even larger either rural or quasi-urban settlement). 48 Table A-2 shows that the villages in the second regional group on average were more populated than in the first regional group, and that their derevnya-type villages had budgets that were twice as large because the regions in the second group were wealthier in general. However, for the townships there were no significant differences between the groups (on average, townships in the second group were larger, but less wealthy). Then, for each regional group, of all the settlements without treatment 6 were selected by the nearest-neighbor method, which measured the least Euclidian distance between the group means shown in Table A-2 and the same parameters, population, and budget size of the settlements without treatment.18 First, 9 comparison settlements for each regional group were selected, and then the sample was narrowed down to 3 settlements per regional group based on the same logistical reasons as for the treatment group. The settlements in the comparison group with their main characteristics are presented in Table A-3 (also, Table A-55). Table A-3. Settlement populations and budgets, comparison group, 2015 Annual budget (in Population, Municipal district Settlement Settlement type thousands of 2015 rubles), 2015 Tver Oblast Udomelsky CD Kopachevo Village 496 2,769.8 (derevnya) Vyshnevolotsky Terelesovsky Township 1,648 7,368.5 (poselok) Kirov Oblast Kirovo-Chepetzky Kstinino Village 1,904 6,804 (selo) Nizhny Novgorod Oblast Shatkovsky Sharapovo Village 1509 6,011.3 (derevnya) Dalnekonstantinovsky Sarley Village 2114 5,930 (selo) Ulyanovsk Oblast 18 For the matching procedure, we use the package MatchIt developed for R (Ho et al., 2007). 49 Annual budget (in Population, Municipal district Settlement Settlement type thousands of 2015 rubles), 2015 Insensky Glotovka Township 2,138 5,935 (poselok gorodskogo tipa) Source: Population and budget data from Rosstat, 2015. As Table A-3 shows, in the Tver and Kirov regions the settlements in the comparison group were very close to the settlements in the treatment group in both population and budget size (with the exception of Kstinino, whose annual budget was lower by approximately 0.5 million rubles). For the other two regions, the parameters were very close as well. Individual Sample Characteristics The third step was the selection of respondents. This selection was randomized with quotas on gender and age groups that corresponded to regional group means. Overall, the number of respondents in the treatment group was 2,010 and in the comparison group it was 500. The interviews with respondents were conducted via face-to-face interactions with the interviewer. Individual sampling results for treatment and comparison groups are described in Tables A- 4 and A-5. Gender, age groups, marital status, education and income are presented in Figures A-1 to A-55. Table A-4. Treatment group characteristics, by region, 2015 Region Number of Number of Mean settlement Mean annual budget per settlements respondents population, 2015 settlement (in per settlement thousands of rubles), 2015 Tver Oblast 13 501 1,074 5,244.99 Kirovsk Oblast 12 507 1,666 6,323.17 Nizhny 13 502 1,359 5,631.1 Novgorod Oblast 50 Ulyanovsk 13 500 2,829 6,791.44 Oblast Total 51 2,010 1,732 5,997.68 Source: Population and budget data from Rosstat, 2015. Table A-5. Comparison group characteristics, by region, 2015 Region Number of Number of Mean population, Mean annual budget settlements respondents 2015 size, thousands of RUB, 2015 Tver Oblast 2 113 1,072 5,069.15 Kirov Oblast 1 99 1,904 6,804 Nizhny 2 186 1,812 5,970.65 Novgorod Oblast Ulyanovsk 1 102 2,138 5,935 Oblast Overall 6 500 1,732 5,944.7 Source: Population and budget data from Rosstat, 2015. Figure A-1. Gender in individual sampling Figure A-2. Age in individual sampling (percentage) (percentage) 60 56.22 53.8 35 29.9 30 26.72 27 28 50 43.78 46.2 24.2 25.02 25 20.8 40 18.36 20 30 15 20 10 5 10 0 0 18-30 31-45 46-60 61 and female male older treatment group comparison group treatment group comparison group Note: Chi-squared value: 0.854 with p-value = Note: Chi-squared value: 3.338 with p-value = 0.342; 0.355; Mann-Whitney U value: 514660 with p-value Kruskal-Wallis chi-squared value: 0.992 with p-value = = 0.330 0.319 51 Figure A-3. Marital status in individual Figure A-4. Education in individual sampling, sampling (percentage) (percentage) 70 50 43.2 57.6 57.1 41.1 60 40 35.2 50 30.7 42.4 42.9 30 40 19.3 30 20 12.4 9.2 8.8 20 10 10 0 0 junior secondary specialized university/ Married Not married secondary secondary degree treatment group comparison group treatment group comparison group Note: Chi-squared value: 0.026 with p-value = Note: Chi-squared value: 13.632 with p-value = 0.003; 0.872; Mann-Whitney U value: 496630 with p-value Kruskal-Wallis chi-squared value: 8.236 with p-value = = 0.833 0.004 Figure A-5. Income in individual sampling, percentage 50 44.3 39.2 37.0 40 32.0 30 23.8 23.7 20 10 0 low (up to 20000 rub) medium (20001-30000 high (30001 rub and rub) higher) treatment group comparison group Note: Chi-squared value: 4.831 with p-value = 0.089; Kruskal-Wallis chi-squared value: 1.941 with p-value = 0.163 Table A-6. Summary sample statistics Statistic N Mean St. Dev. Min Max Changes in the life of the settlement 2,442 1.912 0.730 1 3 Satisfaction with: life in the settlement 2,486 3.070 0.846 1 4 housing and communal works 2,376 2.285 0.958 1 4 52 Statistic N Mean St. Dev. Min Max Roads 2,493 2.679 1.249 1 5 water supply 2,432 3.567 1.247 1 5 cultural institutions 2,229 2.981 1.376 1 5 sport facilities 2,042 2.836 1.504 1 5 Public confidence in the head of the settlement 1,819 2.681 0.980 1 4 Responsibility (settlement level) 2,397 2.715 1.457 1 5 Responsibility (street level) 2,436 3.722 1.452 1 5 Accountability of local government 2,510 0.449 0.498 0 1 Approachability of local government 2,510 0.710 0.454 0 1 Local government holds meetings to discuss local issues 2,252 0.867 0.338 0 1 Local meetings allow to solve local issues 2,004 2.698 0.813 1 4 Readiness to invest money: public project 2,510 0.264 0.440 0 1 Readiness to invest money: charity project 2,510 0.319 0.466 0 1 Readiness to invest time and effort: public project 2,510 0.449 0.496 0 1 Readiness to invest time and effort: charity project 2,510 0.445 0.497 0 1 Gender 2,510 0.443 0.497 0 1 Age 2,510 44.649 16.487 18 86 Education (grouped) 2,510 2.685 0.867 1 4 Marital status 2,503 1.428 0.495 1 2 Income (grouped) 2,196 1.805 0.795 1 3 Self-selection Analysis Settlement selection in the sample had to address some methodological questions related to possible self-selection bias. Since settlement’s participation in LISP was voluntary and the decision to participate was always made or approved by the head of the settlement, there was a self-selection element in the proposed sample, i.e. the quality of governance or high levels of public confidence could be positively associated with the decision to participate. In other words, a ‘good’ head of the settlement would be interested in increased quality of public goods provision, hence, he or she would be willing to participate in LISP or any other program that provided an opportunity of such increase. Also, in the settlements that had high levels of interpersonal or institutional confidence, the success of LISP could also possibly be 53 higher than in the others. On the contrary, a ‘bad’ head of the settlement , who is not interested in increased quality of public goods provision, could act otherwise, refusing to participate, and the lower levels of public confidence might undermine the quality of citizen participation. This creates a question about the causal attribution of LISP effects: increased public confidence and the quality of public goods provision could be based on the specific characteristics of any given settlement, such as the personal efficiency of a certain head of the settlement, rather than on any LISP effects. To address this potential issue, the survey used two considerations. The first consideration was in fact a naturally occurring experiment, which took place without any intervention from the survey team. In 2018, the heads of 3 out of 6 settlements in the comparison group made a decision to participate in LISP. Thus, to test the self-selection hypothesis (public confidence and quality of governance influence the decision to participate), the survey compared the levels of public confidence in local governance (i.e., the head of the settlement) between settlements in the comparison group (in those who decided to participate with those who didn’t), and newcomers who joined in 2018 and the settlements in the treatment group. The results are summarized in Figures A-6 and A-7. 54 Figure A-6. Distribution of levels of public confidence in local authorities (by percent of respondents), comparison group settlements 50 46.7 45 39.3 40 35 28.7 30 24.3 25 20 18.5 17.9 16.4 15 10 8.2 5 0 Absolute distrust Somewhat distrust Somewhat trust Absolute trust Didn't join LISP Joined LISP in 2018 Note: Chi-squared value: 20.301 with p-value = 0.000; Mann-Whitney U value: 18452 with p-value = 0.100 Figure A-7. Distribution of levels of public confidence in local authorities (by percent of respondents), comparison group settlements that joined LISP in 2018 and treatment group settlements 45 39.8 39.3 40 35 30 24.3 25 22.0 22.7 20 18.5 17.9 15.5 15 10 5 0 Absolute distrust Somewhat distrust Somewhat trust Absolute trust Treatment group Joined LISP in 2018 Note: Chi-squared value: 9.6899 with p-value = 0.022; Mann-Whitney U value: 137470 with p-value = 0.032 55 The differences are statistically significant, and on a standard 4-point ordinal scale the levels of public doubt in the head of the settlement are significantly higher in the settlements that joined LISP in 2018 than in both the treatment group and in the settlements that didn’t join LISP. The second consideration was based on the results of supplementary qualitative research, which included 48 semi-structured interviews with the heads of settlements and municipal districts about their experience with LISP, including their motivation and assessment of risks. The results suggested that the biggest risk for the head of the municipality before making the decision to participate was that they might fail to collect needed contributions from the citizens. However, this risk emerges not from the existing public confidence or doubt in the head of the settlement but rather from his or her perception of the public confidence, which was evaluated as quite low by the interviewed officials. For example, in order to make the decision, local officials analyzed the experience of their neighboring administrations, and, usually, if their neighbors succeeded in their first year, this increased the probability that the local official would make a decision to participate in LISP, since they faced almost the same situation during the start of the program. Therefore, most of the interviewed heads of settlements took a risk of starting the program regardless of their perception of public confidence. These results suggest that the heads of the settlements, when making the decision to participate in LISP, were basing their decisions on low levels of public confidence. Their decision included an intention to increase community confidence in their institution by providing citizens with an opportunity to participate in setting priorities in the decision- making process over public goods provision. Therefore, the survey results tend to disprove the self-selection hypothesis that the settlements with higher levels of public confidence were more likely to participate in LISP than the ones with low levels of public confidence. In fact, the decision to participate seems to have been made under the opposite circumstances. 56 Annex B: Changes in the Levels of Public Confidence in Local Authorities and Satisfaction with Public Goods Provision 1.Public Confidence in Local Authorities The survey results show that those heads of settlements who ensured that residents were engaged in discussions and in the resolution of local issues under PB programs, also enjoyed greater public confidence (see Figure 16 in the main text). These results are encouraging, since public confidence in local governance in Russia remains quite low (see Russian average score in Figure 16 in the main text). In other words, the settlements that participated in LISP form a different pattern of public confidence in local governance, which diverges from the general tendency observed in Russia in recent years. Moreover, the effect of LISP on public confidence in local governance depended on the intensity of treatment. The results show that the settlements with longer experience in LISP have significantly higher levels of public confidence in local governance, although the pattern is not linear (see Figure B-1). Figure B-1. Levels of public confidence in the head of the settlement (percentage summing “absolute” and “some” public confidence), for settlements grouped by number of projects 90 80.4 80 70 65.2 64.5 57.9 60 50 40 30 20 10 0 0-1 projects 2 projects 3 projects 4 projects and more Source: Sociological survey in Russian regions to measure social effects of participatory budgeting, 2017. 57 With current research design, it was not possible to measure the dynamics of public confidence levels. Therefore, Figure B-1 represents a quasi-dynamic pattern, which was constructed by dividing the sample into the groups of settlements with different experience in LISP. The experience indicator is the number of projects implemented through LISP. In order to test the significance of the LISP influence on public confidence at the settlement level, the effects had to be isolated from individual effects. The concept of public confidence (institutional, interpersonal, or general) is highly dependent on the individual attitudes and values. However, it is possible to assess both settlement-level effects (including LISP) and individual-level predictors. This is achieved by running independent linear probability models (LPMs). The dependent variable is public confidence in the head of the settlement (initial 4- point scale is reduced to binary scale). The settlement-level effects, besides LISP, were the population and budget size of the settlement (budget size is taken in logarithmic scale). Also, the methodology tested discontinuities in LISP (if annual participation of the settlement had a break of 1 year or more) and the diversity of implemented projects. The results of settlement-level LPMs are summarized in Table B-1. Table B-1. Linear probability models on levels of public confidence in the head of the settlement (settlement-level effects) Dependent variable: Public confidence in the head of the settlement Model 1 Model 2 LISP experience 0.038*** 0.021** (0.007) (0.010) LISP discontinuity 0.025 0.022 (0.027) (0.027) Project diversity 0.035** (0.015) Population size -0.00004*** -0.00003** (0.00001) (0.00001) Budget size 0.026 0.012 (0.035) (0.035) 58 Dependent variable: Public confidence in the head of the settlement Model 1 Model 2 Constant 0.420 0.515* (0.285) (0.287) Observations 2,217 2,217 2 R 0.019 0.021 Adjusted R2 0.017 0.019 Residual Std. Error 0.472 (df = 2212) 0.472 (df = 2211) F Statistic 10.426*** (df = 4; 2212) 9.365*** (df = 5; 2211) * p<0.1; **p<0.05; ***p<0.01 Note: Standard errors in parentheses. Model 1 included years of participation in LISP, discontinuities in the process, and population and budget size of the settlement. The results show that every additional year of LISP experience increased the probability of higher public confidence in the head of the settlement by 3.8 percentage points. Population size was also statistically significant, but it did not have a substantial effect. However, if the diversity of projects is added into the equation (see Model 2), the results change. The diversity of projects was measured by the number of types of local infrastructure targeted by LISP (namely, roads, water supply, communal works, cultural institutions, sport facilities and others). As the results of Model 2 show, project diversity has even stronger positive effect on public confidence levels (3.5 percentage points). This means that not only the longevity of participation matters, but also the variety: if the projects were diverse and target different areas or types of local infrastructure, the citizens were more likely to have higher public confidence in their head of settlement. This can be explained by the visibility of the projects: people see how life in the settlement improves in various ways, for example, they get not only improved roads and pavements, but also renewed cultural and sport facilities. This in turn is linked with the ‘good’ governance of the head of the settlement who was actively involved in the implementation of LISP. 59 However, settlement-level effects were not the primary effects in explaining the levels of public confidence in local governance. They were highly influenced by individual attitudes and views on local governance and on the settlement itself. That is why the individual effects were modelled independently in order to reveal other kinds of public confidence predictors. The results of individual-level models are presented in Table BB-2. Table B-2. Linear probability models on levels of public confidence in the head of the settlement (individual-level effects) Dependent variable: Public confidence in the head of the settlement Model 1 Model 2 Model 3 Model 4 Age -0.0001 -0.001 -0.001 -0.001 (0.001) (0.001) (0.001) (0.001) Gender 0.008 -0.008 0.016 0.019 (0.023) (0.021) (0.020) (0.020) Education -0.002 -0.001 -0.010 -0.016 (0.013) (0.012) (0.012) (0.012) *** ** Income -0.0001 -0.0001 -0.0001 -0.00005 (0.00004) (0.00003) (0.00003) (0.00003) Knowledge 0.035 -0.024 -0.059 -0.054 (0.059) (0.054) (0.052) (0.052) Satisfaction levels: Life in the settlement 0.163*** 0.132*** 0.131*** (0.028) (0.027) (0.027) Housing and communal 0.399*** 0.331*** 0.326*** services (0.022) (0.022) (0.022) Responsiveness assessment: Accountability of local 0.214*** 0.205*** government (0.022) (0.022) *** Approachability of local 0.116 0.112*** government (0.025) (0.025) Participation: Participation in LISP 0.053** 60 Dependent variable: Public confidence in the head of the settlement Model 1 Model 2 Model 3 Model 4 (0.024) Constant 0.661*** 0.428*** 0.331*** 0.337*** (0.080) (0.075) (0.072) (0.072) Observations 1,706 1,605 1,605 1,605 R2 0.004 0.240 0.308 0.311 Adjusted R2 0.001 0.236 0.304 0.306 0.468 (df = Residual Std. Error 0.411 (df = 1597) 0.392 (df = 1595) 0.392 (df = 1594) 1700) 1.512 (df = 5; 71.967*** (df = 7; 79.026*** (df = 9; 71.812*** (df = 10; F Statistic 1700) 1597) 1595) 1594) * Note: p<0.1; **p<0.05; ***p<0.01; Standard errors in parentheses. These four models demonstrate the process of evaluation of municipal governance by the citizens on the individual level, with public confidence as the evaluation proxy. Citizens evaluate the head of their settlement according to a certain level of institutional performance in meeting the social needs of the settlement and his or her responsiveness toward public participation. Model 1 shows that there is no significant effect of demographic variables on public confidence in local governance. A knowledge variable was added, that tested whether the respondent was able to name the head of the settlement. The variable represents binary coding for wrong and right answers. However, it did not add any explanatory value, so it is safe to assume that the respondents did not evaluate the personal representation of municipal governance, but their public confidence was directed towards the institution itself. Model 2 added satisfaction rates explained in the previous sections, and the results showed that higher satisfaction with life in the settlement in general and with its housing and communal services did increase the level of public confidence in the head of the settlement. This proves that the citizens evaluated the delivery of public goods that matched their needs: if the quality of the delivery was satisfactory, the public confidence in those who were responsible for this delivery increased. 61 Model 3 added the assessment of the responsiveness of the local government: both accountability (whether local administration takes into account citizens’ opinion) and approachability (whether local government is accessible for the citizens to discuss local issues). Both of these responsiveness dimensions significantly increased the levels of public confidence in local government. Finally, Model 4 introduced the participation dummy variable, which divided the sample into two groups: people who never participated in LISP, and people who somehow got involved (donated money, participated in the local assemblies or initiative groups or even contributed with manual labor). The results of Model 4 suggest that the participation process itself increases public confidence in local government, which is not surprising since the head of the settlement or municipal officials were usually deeply involved in implementation of the projects developed through LISP. So, by participating in any kind of activity through LISP, citizens got to see local governance from within – this “opened the black box” of governance for ordinary people. By learning the details of the process, citizens became more aware of how the settlement was managed, which increased their public confidence. Thus, the participation in LISP had a small but statistically significant effect on the process of evaluation of municipal governance. Strictly speaking, all variables from Models 1 to 3 were not directly associated with LISP (however, accountability and approachability can be affected by it). However, the addition of participation in LISP in the regression models showed that participation did indeed play a role in assessing the quality of life in the settlement and was positively associated with the evaluation of municipal governance (by higher levels of public confidence in the head of the settlement). 2. Citizens’ Perception of their Role in Local Development There is also another dimension of public confidence that is linked with citizens’ attitudes to their own role in local development. The respondents reported their readiness to invest money in socially significant micro-projects, which is a function of both interpersonal and institutional public confidence (see Figure B-2). The interpersonal dimension comes from joint 62 efforts in the cofinancing process. Usually, in LISP the money contributed by people is collected by the citizens themselves: this means the members of initiative groups ask other citizens to contribute to a certain initiative. Thus, the readiness to invest personal funds can be fostered by the previous positive experience of participation in LISP: people donated money for a project and were able to observe the accomplishment of their efforts. The institutional dimension is connected to LISP itself. Since the cofinancing procedure is usually a specific characteristic of the program, an earlier positive experience from the investment could also be associated with the positive results of LISP in a particular settlement. In many respects, the citizens’ willingness to invest their own money was associated with citizens’ trust in LISP mechanisms and procedures, although this readiness went beyond this project. People also may be ready to invest in public projects outside of LISP, thus fulfilling their need for public participation, which until then had been blocked by the weakness of civil society. Figure B-2. “Are you ready to invest your personal money in the public project?” (Percent of respondents answering “Yes”) 30% 25% 20% 15% 29% 10% 15% 5% 0% comparison group treatment group Source: Sociological survey in Russian regions to measure social effects of participatory budgeting, 2017. 3. Linkages to Beneficiary Satisfaction with Local Infrastructure and Services As explained in Section 5C, citizens in communities that participate in LISP-type programs were more satisfied with local socioeconomic infrastructure and services. As these findings 63 are parallel to the public confidence in local government and self-perception, independent linear probability models were run, with the satisfaction with each component of social infrastructure treated as the dependent variable on the number of projects implemented through LISP and targeting the same type of social infrastructure. This supplemental analysis evaluated the contribution of LISP to the variance in satisfaction with public infrastructure in the settlements that were affected by LISP (i.e., the treatment group). This logic aimed to analyze whether there was a significant covariation between satisfaction in the particular type of infrastructure (roads, water supply, cultural institutions, sport facilities and communal services) and the intensity of LISP treatment (which is expressed as the number of projects implemented through LISP). The results are presented in Table B-3. 64 Table B-3. Linear probability models of satisfaction with social infrastructure on the number of projects implemented through LISP, on treatment group Dependent variables: Satisfaction with: Roads Water supply Cultural institutions Sport facilities Communal services (1) (2) (3) (4) (5) Project diversity 0.179*** -0.077** 0.047 -0.041 0.032 (0.043) (0.033) (0.043) (0.041) (0.040) Roads 0.045*** 0.035*** 0.109*** 0.139*** 0.005 (N of projects) (0.014) (0.011) (0.014) (0.014) (0.013) Water supply -0.229*** -0.108*** -0.008 0.149*** -0.043 (N of projects) (0.032) (0.026) (0.034) (0.033) (0.031) Cultural institutions -0.041** -0.097*** 0.205*** 0.113*** 0.008 (N of projects) (0.021) (0.017) (0.021) (0.022) (0.020) Sport facilities -0.005 0.133*** 0.116** 0.338*** 0.127*** (N of projects) (0.048) (0.038) (0.049) (0.046) (0.045) Communal services -0.039 -0.030 0.062** 0.007 0.012 (N of projects) (0.028) (0.022) (0.029) (0.027) (0.026) Other projects -0.062 0.100*** 0.066 0.252*** 0.079* (N of projects) (0.045) (0.035) (0.047) (0.043) (0.042) Constant 0.249*** 0.953*** 0.116* 0.104* 0.335*** (0.063) (0.050) (0.066) (0.062) (0.060) Observations 761 841 791 793 1,029 R2 0.186 0.246 0.145 0.187 0.039 Adjusted R2 0.178 0.240 0.137 0.180 0.033 Residual Std. Error 0.446 (df = 753) 0.373 (df = 833) 0.464 (df = 783) 0.453 (df = 785) 0.489 (df = 1021) 24.579*** 38.842*** 18.917*** 25.787*** 5.997*** F Statistic (df = 7; 753) (df = 7; 833) (df = 7; 783) (df = 7; 785) (df = 7; 1021) Note: *p<0.1; **p<0.05; ***p<0.01. Strongest effects in semi-bold. As Table B-3 shows, there were substantial effects on satisfaction with different types of social infrastructure, except communal services. This means that even if a certain type of social infrastructure was sufficiently developed, projects implemented through LISP still 65 contributed to citizens’ evaluation of the quality of this type of infrastructure. It is likely that this effect is observed because LISP projects reflected citizens’ priorities, which were conveyed during the process of program implementation. LISP projects demonstrated substantial effects on two types of social infrastructure: cultural institutions (most of the projects target ‘houses of culture’, i.e. community centers) and sport facilities. LISP projects had a small effect on satisfaction with roads (4.5 percentage points) and a strong negative effect on satisfaction with the water supply (10.8 percentage points). Because roads are very capital-intensive and expensive, these findings can be explained by the effects of other road-construction projects outside LISP that overshadow its effect. The second finding (on water supply), however, is more difficult to explain. Model 2 suggests that the higher number of projects were implemented targeting water supply, the less satisfied were the citizens. This might be explained by the specificity of certain projects targeting water supply. For example, in some cases, communities approved public works to repair or install the pipes that supplied water to a certain street. Thus, the citizens living on this street may feel more satisfied with the improvements in their water supply, while the citizens who did not gain access to the improved water supply may have felt deprived, resulting in a reverse effect on the beneficiaries’ opinions. Also, the number of projects targeting sport facilities had a stronger effect on the satisfaction with communal services. This may be explained by the specificity of the implemented projects: most of the initiatives in the communal sphere included beautification or improvement of certain public spaces (squares, parks, playgrounds, etc.) and often the installment of a variety of sports equipment was a part of these initiatives. Conversely, when a sports field was constructed, the project also included beautification improvements (planting lawns, flowers, renovation of pavement, etc.). Therefore, these two types of social infrastructure seem closely interdependent, which may have caused an amplification effect. 66 Therefore, it can be concluded that LISP effectively contributes to the increase in citizens’ satisfaction by providing services related to the social infrastructure (water, roads, community centers, and sport facilities). 67 References Belenchuk A., V. Vagin, and I. Shulga 2017. “Participatory Budgeting as a Tool for Making Budgetary Expenditures More Effective.” Finance Journal 8: 10–21 (in Russian). Dias N., ed. 2018. Hope for Democracy: 30 Years of Participatory Democracy Worldwide. Faro: Epopeia Records | Oficina. Hanson, P. 2006. Federalism with a Russian face: regional inequality, administrative capacity and regional budgets in Russia. Economic Change and Restructuring, 39:3-4, pp. 191-211. Ho, D., I. Kosuke, G. King, and E. Stuart 2007. “Matching as Nonparam etric Preprocessing for Reducing Model Dependence in Parametric Causal Inference.” Political Analysis 15 (3): 199-236. Khachatryan, G., I. Shulga, S. Gridin, and A. Sukhova 2016. “Operational Manual for the Initiative Budgeting Practice: The Example of the Local Initiatives Support Program.” Alex Publishers (in Russian). Rubin, D.B. 1973. “Matching to Remove Bias in Observational Studies.” Biometrics 29(1): 159-183. Shulga I., A. Fadeeva, and A. Sukhova. 2017. Local Initiatives Support Program in the Republic of Bashkortostan. Moscow: Alex Publishers. https://drive.google.com/file/d/1jxF8XbKeQA5T289aUdKQibYVbUH5bSzj/view?usp=sharing Shulga I., and A. Sukhova. 2016a. Local Initiatives Support Program: Improving Local Self- Government and Promoting Participatory Budgeting. Moscow: Alex Publishers (in Russian). https://drive.google.com/file/d/0B1mm6DBXVdy9YWd0TWM3RXdjVFE/view?usp=sharing Shulga, I., and A. Sukhova. 2016b. “Local Initiatives Support Program in Russia: An Example of the Best Participatory Budgeting Practice.” Municipal Power Journal 2–3: 8–19 (in Russian). Shulga I., V. Vagin, A. Sukhova, G. Khachatryan, N. Gavrilova, N. Shapovalova, and L. Shilov 2017. Initiative Budgeting: Russian Experience of Citizens’ Participation in Addressing Local Issues. Moscow: Alex Publishers. http://budget4me.ru/articles/school/initiative-budgeting-russian- experience-of-citizens-participation/ 68 Wong S. and Guggenheim S. 2018. Community Driven Development: Myths and Realities. Policy Research Working Paper 8435; and World Bank 2015. Participatory and Community Driven Development in Urban Areas Working Paper. Vakhshtayn V., Stepantsov P. 2017. “Confidence in State Institutions and Strategies of Citizens' Economic Behavior” (in Russian). Available online: https://www.rbc.ru/politics/29/03/2017/58dbb1ed9a7947e4c4de9dcb Yushkov, A., Savulkin L., Oding N. 2017. Intergovernmental Relations in Russia: Still A Pendulum? Public Administration Issues, Special Issue (electronic edition), pp. 38-59. Zhuravskaya, E. 2010. Federalism in Russia. Russia after the global economic crisis. PIIE, Washington, DC. Zubarevich, N. 2014. Regional Development and Regional Politics in Russia (in Russian). ECO, No. 4, pp. 7-27. Wampler, B. 2007. ‘A guide to participatory budgeting’, in: Shah, A. (ed.) Participatory budgeting. (Washington, DC: World Bank Institute, Public Sector Governance and Accountability Series: 21-54). 69 Social Protection & Jobs Discussion Paper Series Titles 2017-2019 No. Title 1931 Can Local Participatory Programs Enhance Public Confidence: Insights from the Local Initiatives Support Program in Russia by Ivan Shulga, Lev Shilov, Anna Sukhova, and Peter Pojarski May 2019 1930 Social Protection in an Era of Increasing Uncertainty and Disruption: Social Risk Management 2.0 by Steen Lau Jorgensen and Paul B. Siegel May 2019 1929 Developing Coherent Pension Systems: Design Issues for Private Pension Supplements to NDC Schemes by William Price April 2019 1928 Pensions in a Globalizing World: How Do (N)DC and (N)DB Schemes Fare and Compare on Portability and Taxation? by Bernd Genser and Robert Holzmann April 2019 1927 The Politics of NDC Pension Scheme Diffusion: Constraints and Drivers by Igor Guardiancich, R. Kent Weaver, Gustavo Demarco, and Mark C. Dorfman April 2019 1926 Setting Up a Communication Package for the Italian NDC by Tito Boeri, Maria Cozzolino, and Edoardo Di Porto April 2019 1925 Sweden’s Fifteen Years of Communication Efforts by María del Carmen Boado-Penas, Ole Settergren, Erland Ekheden, and Poontavika Naka April 2019 1924 Information and Financial Literacy for Socially Sustainable NDC Pension Schemes by Elsa Fornero, Noemi Oggero, and Riccardo Puglisi April 2019 1923 Communicating NEST Pensions for “New” DC Savers in the United Kingdom by Will Sandbrook and Ranila Ravi-Burslem April 2019 1922 Harnessing a Young Nation's Demographic Dividends through a Universal NDC Pension Scheme: A Case Study of Tanzania by Bo Larsson, Vincent Leyaro, and Edward Palmer April 2019 1921 The Notional and the Real in China’s Pension Reforms by Bei Lu, John Piggott, and Bingwen Zheng April 2019 1920 Administrative Requirements and Prospects for Universal NDCs in Emerging Economies by Robert Palacios April 2019 1919 Bridging Partner Lifecycle Earnings and Pension Gaps by Sharing NDC Accounts by Anna Klerby, Bo Larsson, and Edward Palmer April 2019 1918 The Impact of Lifetime Events on Pensions: NDC Schemes in Poland, Italy, and Sweden and the Point Scheme in Germany by Agnieszka Chłoń-Domińczak, Marek Góra, Irena E. Kotowska, Iga Magda, Anna Ruzik-Sierdzińska, and Paweł Strzelecki April 2019 1917 Drivers of the Gender Gap in Pensions: Evidence from EU-SILC and the OECD Pension Model by Maciej Lis and Boele Bonthuis April 2019 1916 Gender and Family: Conceptual Overview by Nicholas Barr April 2019 1915 Labor Market Participation and Postponed Retirement in Central and Eastern Europe by Robert I. Gal and Márta Radó April 2019 1914 NDC Schemes and the Labor Market: Issues and Options by Robert Holzmann, David Robalino, and Hernan Winkler April 2019 1913 NDC Schemes and Heterogeneity in Longevity: Proposals for Redesign by Robert Holzmann, Jennifer Alonso-García, Heloise Labit-Hardy, and Andrés M. Villegas April 2019 1912 Annuities in (N)DC Pension Schemes: Design, Heterogeneity, and Estimation Issues by Edward Palmer and Yuwei Zhao de Gosson de Varennes April 2019 1911 Overview on Heterogeneity in Longevity and Pension Schemes by Ron Lee and Miguel Sanchez-Romero April 2019 1910 Chile's Solidarity Pillar: A Benchmark for Adjoining Zero Pillar with DC Schemes by Eduardo Fajnzylber April 2019 1909 Sweden: Adjoining the Guarantee Pension with NDC by Kenneth Nelson, Rense Nieuwenhuis, and Susanne Alm April 2019 1908 The ABCs of NDCs by Robert Holzmann April 2019 1907 NDC: The Generic Old-Age Pension Scheme by Marek Góra and Edward Palmer April 2019 1906 The Greek Pension Reforms: Crises and NDC Attempts Awaiting Completion by Milton Nektarios and Platon Tinios April 2019 1905 The Norwegian NDC Scheme: Balancing Risk Sharing and Redistribution by Nils Martin Stølen, Dennis Fredriksen, Erik Hernæs, and Erling Holmøy April 2019 1904 The Polish NDC Scheme: Success in the Face of Adversity by Sonia Buchholtz, Agnieszka Chłoń-Domińczak, and Marek Góra April 2019 1903 The Italian NDC Scheme: Evolution and Remaining Potholes by Sandro Gronchi, Sergio Nisticò, and Mirko Bevilacqua April 2019 1902 The Latvian NDC Scheme: Success Under a Decreasing Labor Force by Edward Palmer and Sandra Stabina April 2019 1901 The Swedish NDC Scheme: Success on Track with Room for Reflection by Edward Palmer and Bo Könberg April 2019 1803 Rapid Social Registry Assessment: Malawi’s Unified Beneficiary Registry (UBR) by Kathy Lindert, Colin Andrews, Chipo Msowoya, Boban Varghese Paul, Elijah Chirwa, and Anita Mittal November 2018 1802 Human(itarian) Capital? Lessons on Better Connecting Humanitarian Assistance and Social Protection by Ugo Gentilini, Sarah Laughton and Clare O’Brien November 2018 1801 Delivering Social Protection in the Midst of Conflict and Crisis: The Case of Yemen by Afrah Alawi Al-Ahmadi and Samantha de Silva July 2018 1705 Aging and Long-Term Care Systems: A Review of Finance and Governance Arrangements in Europe, North America and Asia-Pacific by Laurie Joshua November 2017 1704 Social Registries for Social Assistance and Beyond: A Guidance Note & Assessment Tool by Phillippe Leite, Tina George, Changqing Sun, Theresa Jones and Kathy Lindert July 1027 1703 Social Citizenship for Older Persons? Measuring the Social Quality of Social Pensions in the Global South and Explaining Their Spread by Tobias Böger and Lutz Leisering July 2017 1702 The Impacts of Cash Transfers on Women’s Empowerment: Learning from Pakistan’s BISP Program by Kate Ambler and Alan de Brauw February 2017 1701 Social Protection and Humanitarian Assistance Nexus for Disaster Response: Lessons Learnt from Fiji’s Tropical Cyclone Winston by Aisha Mansur, Jesse Doyle, and Oleksiy Ivaschenko February 2017 To view Social Protection & Jobs Discussion Papers published prior to 2017, please visit www.worldbank.org/sp. ABSTRACT This paper reviews the performance of the Russia Local Initiatives Support Program (LISP) as an instrument for directly addressing the needs of the population to access socioeconomic infrastructure, for increasing public confidence in self-governance frameworks and institutions through dialogue and community budgeting consultations, and for strengthening the capacity for local self-governance. The paper does this by looking at historical and survey data from the implementation of the LISP methodology as part of regional programs in Russia. ABOUT THIS SERIES Social Protection & Jobs Discussion Papers are published to communicate the results of The World Bank’s work to the development community with the least possible delay. This paper therefore has not been prepared in accordance with the procedures appropriate for formally edited texts. For more information, please contact the Social Protection Advisory Service, the World Bank, 1818 H Street, N.W., Room G7‑803, Washington, DC 20433, USA. Telephone: +1 (202) 458 5267, Fax: +1 (202) 614 0471, E-mail: socialprotection@worldbank.org or visit us on-line at www.worldbank.org/sp.