Report No: AUS0000898 . Russian Federation Strengthening Social Protection Towards more effective social assistance in Russia: An update of the system performance considering new national target of halving poverty by 2024 Ruslan Yemtsov, Aleksandra Posarac, Maria Nagernyak and Irina Albegova . June 24, 2019 . SPL . 1 © 2017 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. 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. Attribution—Please cite the work as follows: “World Bank. 2019. Towards more effective social assistance in Russia: An update of the system performance considering new national target of halving poverty by 2024. © World Bank.â€? All 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: 202-522-2625; e-mail: pubrights@worldbank.org. 2 Table of Contents Acknowledgements................................................................................................................................. 4 Abbreviations and Acronyms .................................................................................................................. 5 Executive summary ................................................................................................................................ 6 Chapter One.......................................................................................................................................... 10 How Is Social Assistance in Russia Contributing to Halving Poverty by 2024? .................................. 10 Chapter Two ......................................................................................................................................... 14 Performance of the Social Assistance System: An Update ................................................................. 14 2.1. Spending: Aiming at Allocating Resources to the Highest Needs .......................................... 15 2.2. Coverage: Inclusiveness of the System for those Who Need Support .................................... 16 2.3. Targeting: Ability of the System to Direct Assistance to The Needy ...................................... 18 2.4. Adequacy: Providing Support Congruent to the Needs .......................................................... 24 2.5. Impact of social assistance in Russia on poverty reduction. .................................................. 26 2.6. International Comparisons: Russia in the Context of Global Trends and Benchmarks .......... 28 Chapter Three ....................................................................................................................................... 33 Renewing the Social Assistance System in Russia............................................................................... 33 3.1 Understanding the drivers of performance ............................................................................. 34 3.2. Towards an Effective Social Assistance System in Russia ...................................................... 38 3.3 What is next for research in social assistance in Russia? ........................................................ 40 Bibliography .......................................................................................................................................... 41 Annex: Data on social assistance in Russia ......................................................................................... 44 3 Acknowledgements The Report team was led by Ruslan Yemtsov, Global Lead for Social Safety Nets, and Aleksandra Posarac, Lead Economist and Project Manager (World Bank) and comprised Maria Nagernyak, Consultant and the Russia Social Protection Reimbursable Advisory Service (RAS) Project Coordinator, and Irina Albegova, Consultant. The analysis builds on joint activities implemented under the RAS Project with the National Finance Institute of the Ministry of Finance of the Russian Federation (NIFI) whose team included Vladimir Nazarov, Director General of NIFI, Olesya Feoktistova, Head of the NIFI Social Policy Department, Elena Andreeva, Senior Researcher, Dmitri Bychkov, Senior Researcher, and Elena Grishina, Senior Researcher with the Russian Academy of National Economy and Public Service and Associate Researcher with NIFI. The overall guidance was provided by Andras Horvai (Country Director, World Bank) and Cem Mete, (Practice Manager for Social Protection and Jobs Word Bank Group in the Europe and Central Asia - ECA Region) on the World Bank side and by Svetlana Gashkina (Ministry of Finance, Department Head). The team has benefited tremendously from multiple roundtables and discussions of the report findings, including with the Russian and World Bank leading experts, researchers and academics in the area social assistance. 4 Abbreviations and Acronyms ADEPT - World Bank’s Software Platform for Automated Economic Analysis ALMP - Active labor market programs ASPIRE - Atlas of Social Protection Indicators of Resilience and Equity B40 - Bottom 40 percent of the income distribution BRICS - Brazil, Russia, India, China, and South Africa CIT - Corporate income tax ECD - Early childhood development ECA – Europe and Central Asia GDP - Gross Domestic Product GRP--Gross Regional Product HBS – Household Budget Survey IMF – International Monetary Fund LFS - Labor Force Survey LITS - Life in Transition Survey MOF – Ministry of Finance NIFI – National Finance Research Institute of the Ministry of Finance of the Russian Federation OECD - Organization for Economic Co-operation and Development PPP - Purchasing power parity RAS – Reimbursable Advisory Services RF – Russian Federation RLMS - Russia Longitudinal Monitoring Survey Rosstat - State Statistics Service of Russian Federation SA – Social Assistance SCD - Systematic Country Diagnostic SPJ – Social Protection and Jobs Group at the World Bank SSN – Social Safety Net T60 - Top 60 percent of the income distribution VAT - Value-added tax WDI - World Development Indicators WBG – World Bank Group 5 Executive summary This report updates the assessment of the performance of the social protection (SP) system in Russia, using the 2014-2017 rounds of the Survey on Incomes and the Participation in Social Programs (VNDN in Russian) by the State Statistic Service of the Russian Federation (Rosstat), the largest household annual monitoring survey in Russia, and the main source of official data on poverty. The 2017 VNDN Survey round has a sample (about 150,000 households) that is 3 times larger than its standard sample size and it allows for a comprehensive assessment of the SP system performance both at the national and the regional levels. The update looks at the trends, using data from 2014-2017 and provides international comparisons setting Russia’s salient features of its social assistance system against global benchmarks. To assess performance of social transfers, the report also uses a new dataset of regional social spending collected by the National Finance Research Institute of the Ministry of Finance (NIFI)1, and harmonized dataset based on VHDN with data from other countries in ASPIRE (Atlas of Social Protection and Resilience).2 Comparing Russia with global trends and benchmarks, the report identifies strengths and weaknesses of the current system. The update also uses material from several background studies, prepared by the World Bank in collaboration with the NIFI, within the Reimbursable Advisory Services (RAS) Project for the Ministry of Finance of the Russian Federation 2016-2019 on “Strengthening Social Protection System in Russiaâ€?. This includes a comprehensive study of the social assistance system in Russia published in 20173, reports for three regions where a pilot population survey on social assistance beneficiaries was conducted4, material from workshops and seminars, analytical notes and surveys. The report depicts the evolution, current state and on-going changes in the social assistance system in Russia. The context for the update is the 2018 May Decree by the President of Russia5, which defined the new strategic goals for socio-economic development and poverty alleviation in Russia. The Decree put human development and poverty reduction at the center of the Russia’s development strategy over the medium term. To achieve the May decree goals by 2024, the Government developed 12 National Projects, particularly in health, support to families and education, and allocated significant resources for their implementation. The national goals include halving poverty, increasing the quality of education and improving health status to extend longevity and economically active life. For social protection, the objectives pertain to higher efficiency and effectiveness, greater share of resources directed to families in need of assistance, better performance results and stronger alignment with human development goals. More recently, the Ministry of Labor and Social Protection launched a series of regional pilots to inform decisionmakers on good practice examples in implementing the reform in the regions. This update is intended to feed into this on-going process, enabling the World Bank team to continue to support the Government of the Russian Federation in its efforts to improve efficacy of its social assistance system. Through the RAS project, the social assistance team of the World Bank has established itself as credible source of robust analysis and solid technical advice. 1 http://aspire.nifi.ru/ 2 http://datatopics.worldbank.org/aspire/ 3 Strengthening Social Protection System in the Russian Federation (Nazarov V. and A. Posarac, eds.), World Bank and NIFI joint report. M., 2017 (In Russian) - Развитие Ñ?ффективной Ñ?оциальной поддержки наÑ?елениÑ? в РоÑ?Ñ?ии: адреÑ?ноÑ?Ñ‚ÑŒ нуждаемоÑ?Ñ‚ÑŒ, универÑ?альноÑ?Ñ‚ÑŒ / Под ред. Ð’. Ð?азарова и Ð?. Пошарац. Ð?аучный доклад. Ðœ., 2017 4 Pilots were conducted in three regions: Volgograd, Komi Republic and Leningrad. See World Bank and NIFI 2019 (in Russian). 5 President of Russian Federation Decree from 07.05.2018, â„–204 6 Why is Social Assistance Needed to Achieve Poverty Reduction Targets? In May 2018, Russia declared a national goal of halving poverty by 2024. To achieve this goal, in addition to achieving and sustaining a robust growth (at least 3 percent per annum), Russia needs to use social transfers targeted at the poor. The World Bank Russia Social Protection team jointly with NIFI, combining different sources of administrative and survey-based data, has estimated that Russia spends about 3.2 percent of GDP on non-contributory social transfers/ social assistance programs. The estimate is based on an inventory of programs counted as social assistance in Russia and it includes merit-based benefits and top ups to salaries of certain public sector employees for special working conditions (e.g. teachers and health workers in harsh climate conditions). Using a standardized (World Bank ASPIRE database6) approach to estimating spending on social assistance to enable cross country comparisons, the spending is estimated at 2.8 percent of GDP and it excludes salary top ups.7 Both estimates show spending that is significantly above the world average of 1.6 percent and above the spending in the Russia’s regional group (2.2 percent). Arithmetically speaking and assuming perfect flexibility in allocating resources between social assistance programs, spending on social assistance is Russia is more than three times higher than the combined income deficit of all poor families in the country before transfers.8 While no social assistance system in the world is solely focused on the low-income population, the social assistance performance analysis conducted for this report shows that currently, the poverty impact of social assistance programs in Russia is weak. Although there are many reasons behind this weak performance in terms of poverty reduction, the key reason is that in contrast to many other countries, the current social assistance system in Russia does not prioritize low-income population. Under current circumstances, to halve poverty by 2024 or move an additional 4 million people out of poverty,, in addition to increased growth, social assistance spending (assuming that its poverty reduction performance would not change), will have to double. This is a fiscally unaffordable scenario. Hence, the efficacy of social assistance must improve. Allocating more resources to means-tested programs, improving accuracy of means tests and increasing the value of benefits to the poor could make the national goal of poverty reduction achievable. Reforming the Russian social assistance system is not an easy task. The system has evolved for almost a century, reflecting different needs of different times and different approaches to the roles, functions and objectives of the social assistance system. Over time, new programs have been introduced, without a single social assistance program being discontinued, creating a maze of mostly small interventions (“measuresâ€?9). The complexity of the social assistance system has reached a point of the system becoming unmanageable, and difficult to assess. The stocktaking conducted within Russia Social Protection RAS found 800 “measuresâ€? legislated and financed at the federal level plus over 13,000 regional “measuresâ€? (on average each region in Russia has over 150 different social assistance 6 World Bank (2018). The State of Social Safety Nets 2018, World Bank. Washington, D.C. 2018. 7 This internationally harmonized definition is used for performance analysis using survey data and for international comparisons, unless otherwise specified. The merit-based programs are part of social assistance in many countries, hence included in performance analysis and international comparisons. 8 Freije-Rodriguez, Samuel, Aleksandra Posarac, and Apurva Sanghi·2019. Can Russia halve poverty by 2024? Brooking Institute Blog, May 9, 2019. https://www.brookings.edu/blog/future-development/2019/05/09/can- russia-halve-poverty-by-2024/ 9 Social assistance benefits in Russia are called “measures of social supportâ€? and each is regarded as a separate unit, governed by a separate law, even though most entail meager cash payments with only minimum variation in rules. 7 benefits). Only a handful of those are means tested and targeted10 to the poor. Majority of benefits are allocated to beneficiaries using a categorical approach, i.e. benefits are targeted to specific groups or “categoriesâ€?, and everyone within the same group is entitled to the same benefit, regardless of actual need. None of the social assistance programs has a clear theory of change and none has been consistently evaluated. To complicate matters further, in early 2000s, a law was passed that prohibits any change to social assistance “measuresâ€? that would make the current beneficiaries worse off.11 Institutional inertia and tendency to preserve inherited entitlements of “categoriesâ€? of beneficiaries has led to the proliferation of programs that are not oriented towards low-income population. Only a small set of measures is aimed explicitly at the poor and vulnerable population and amounts to just 0.4 percent of GDP. But, even this part could perform better in terms of targeting performance, particularly in some regions. How is Social Assistance Changing in Russia: Performance Indicators are Improving, But Impact on Poverty Remains Weak The changes already underway go into the direction of making the system more efficient. In that, the system builds on its strengths. Russian social assistance achieves very high coverage: 81 percent of the poor are receiving some form of social assistance benefit12. There is some variation across regions, but all cover at least 70 percent of the poor13; none, however, covers all the poor: maximum coverage is 94 percent in the northern oil-producing territory of the Nenetsky Autonomous Region. Comparisons over 2014-2017 show that the coverage of the poor has increased in the country as a whole and in most regions, while there are fewer wealthy people (in the top 20 percent of the distribution) who benefit from social assistance. These shifts are clearly in the positive direction. Still, as many as two-thirds of the Russian population live in families receiving some form of social assistance, reflecting the transfers that are counted as social assistance including the above- mentioned merit-based benefits and top-ups to salaries and predominantly categorical approach to targeting, i.e. allocating transfers to an entire group of the population, irrespective of the income level of beneficiaries. Hence, it is of no surprise that performance and distributional analyses show high coverage of the total population and significant beneficiary and benefit incidence among top deciles of the income distribution. The ability of the system to direct resources to the lower end of income distribution has also improved. In 2014, only 5 percent of all social assistance transfers were received by the poor, by 2017 this share has doubled, to 10 percent. The share of the lowest two quintiles has also increased: from 28percent to 35percent. But these results are weak and falls well below those in most countries, signaling that in its current form and understanding, Russia’s social assistance system does not prioritize the poor. This lack of targeting, (targeting understood as considering means/ income of beneficiaries as a criterion for participation in the program) does not come from the lack of capacity or know how on how to target. The targeting outcomes of means tested social assistance are comparable to the best international benchmarks. They have also improved over the last few years, and now 47percent 10 The translation of the English word targeting as adresnost in Russian has created a certain amount of confusion. In Russia, adresnost is understood as accuracy in the benefit delivery, i.e. that everyone who is supposed to receive the benefit is receiving it, not as selection method of potential beneficiaries. Using a bit longer phrase “identifying beneficiaries taking into account their means/ incomeâ€? is more likely to be understood correctly when targeting poor population is concerned. 11 Article 153 of the Federal Law “On Social Assistanceâ€? of 22.08.2004 â„– 122-ФЗ. 12 Performance analysis presented in this report is based on the Rosstat household survey, according to which, reported income from social assistance amounts to about 2.4percent of GDP. This is considered a very good data “captureâ€? result. 13 With exception of St Petersburg City, a region with the lowest poverty rate (2percent); it covers only one half of its poor. 8 percent of all means-tested benefits goes to the poor; bottom 40percent receives 75percent of all means tested benefits. In some regions of Russia, these results are even more impressive: in 4 regions more than 80percent of means tested benefits goes to the poor. There are however regions (six of them) where this indicator is below 30percent. This shows a lack of consistent application of the same principles of targeting, and tremendous scope for improvement by improving administration of targeted assistance and by allocating more resources towards means-tested benefits. Weak targeting of the overall social assistance system is combined with low adequacy of benefits. On average, the aid provided to the poor is only 17 percent of their pre-transfer incomes – not enough to bring them out of poverty. Over time there was little improvement in the real value of benefits for the poor. The higher income quintiles receive comparable levels of support, but for them these transfers represent an almost irrelevant share of their budgets. Benefits levels vary a lot, with no meaningful link to the needs of their recipients. Poor regions have fewer and the least generous benefits to fit into their meager budgets, while richer regions have plenty of programs for everyone and much higher adequacy for “theirâ€? poor. In poor regions, families excluded from assistance are often those that need the support the most: those living in remote areas, incomplete families, and the unemployed. The combination of poor targeting and low adequacy leads to weak impact on poverty. The national poverty headcount is reduced compared to its pre-transfer level by 21percent - this is a decrease in performance compared to that in 2015 (when it was somewhat higher, 24percent). Despite spending of over 2 trillion Rubles on social assistance, Russia is reducing the poverty deficit by only 254 billion Rubles, and the poor are still missing as much as 421 billion to just meet their minimum basic needs, as defined by the poverty line. The shortfalls of the social assistance system to address poverty were exposed during the 2014-16 economic crisis, when poverty increased by 30percent, despite increased spending on social assistance. Consolidating Progress in Social Assistance Reform: Areas for Further Improvement The population generally accepts the social assistance system principles (merits-based benefits and categorical targeting) and is afraid of abrupt changes. At the same time, recent policy initiatives aimed at improving efficacy of social assistance and achieving better results in terms of poverty reduction demonstrate that a gradual shift in spending towards programs targeted to the poor is possible and is already under way. What is needed is a coherent program of reforms and their acceleration. Reallocation towards means-tested programs will improve performance, but not to the needed/ possible extent due to large exclusion and inclusion errors, fragmentation and small benefits. For greater impact, targeting performance must improve and the level of benefits should be increased significantly. Furthermore, changing the structure and resource allocation of social assistance in favor of the pro-poor programs would need to be done smartly, gathering population support for such a change. The emerging directions for change14 to a more impactful and efficient social assistance system include shifting the social assistance spending towards the low-income population, integrating the social assistance system internally and aligning it with higher-level objectives of poverty reduction, investment in human capital and productive inclusion, using the principles of conditionality; and improving targeting. These actions are already included in the Unified Plan for Implementing National Projects of the Government of Russia (May 2019). In addition, increasing the flow of federal transfers to the regions 14 See final synthesis report “Modernization of the social assistance system to improve its performance for reaching national objectives of halving poverty in Russia by 2024â€? prepared under RAS to MoF «Strengthening Social Protection System in the Russian Federation» by the World Bank, June 2019. 9 conditional on improved performance of their social assistance system can direct more public resources where they are needed the most and are used with greater efficacy. Chapter One How Is Social Assistance in Russia Contributing to Halving Poverty by 2024? Poverty reduction in Russia was remarkable during the years of high economic growth at the beginning of the 2000s but slowed down after the 2008-09 crisis. The positive trend was reversed after 2014, despite increased public spending on social assistance. The economic downturn in 2014-16 exacerbated the system’s weaknesses. At the same time, it has also provided an opportunity for systemic changes. Dramatic decline in living standards in Russia at the onset of transition pushed over one-third of the population below the poverty line. Throughout the 1990s, poverty remained high, the recession in 1995 and the financial crisis in 1998 saw poverty increasing again – to almost 25percent and 30percent, respectively. Economic recovery and rapid growth in the first decade of the 2000s, brought about a remarkable 15 percentage points reduction in poverty incidence, allowing Russia to cut the poverty rate by 65percent. In 2000, according to the Rosstat data, 42.3 million Russian citizens (29percent of the population) had incomes below the national poverty line (subsistence minimum). By 2007 this figure was more than halved to 18.8 million people or 13percent of the population. The true poverty reduction (using a constant poverty line) was probably even more rapid, given two changes in the methodology that have resulted in the real increase in the value of the poverty basket.15 This poverty reduction momentum was sustained during the economic crisis of 2008-09, which did not lead to the fall in living standard. However, by 2012, with a slowdown in economic growth, poverty reduction stalled. The poverty rate started to increase in 2014 after the economic shocks, inflation and fiscal crisis: from 10.8 percent to 13.3 in 2015-17. Poverty in Russia has been studied extensively.16 Most of the studies agree that the main driver of the poverty trend in Russia has been changes in real wages. The secondary, but significant channel is pensions. Changes in real value of pension (indexation and re-valorization), as well as setting the minimum pension above the poverty line have contributed to poverty reduction. The fear that the transition would cause massive unemployment did not materialize – unemployment rates remained low, and employment rates remain relatively stable. The shocks have been continuously absorbed through changes in real wages, so employment appears as a non-significant macro-economic level driver of poverty. Changes in the composition of employment and its productivity strongly affect risks of poverty. Key anti-poverty policies employed by the Government include setting the minimum wage and the minimum pension with the reference to the level of the poverty line. Such efforts have been limited by the available fiscal space (for the State employees and pensioners) and relative ineffectiveness in 15 Poverty line in Russia is regulated by a set of laws and monitored by the Russian statistical office (Rosstat). It uses an absolute approach to determine the poverty line, using the cost of the minimum food basket as an absolute anchor and the non-food part based on norms established by a commission. See detailed discussion of methodologies in Russian Federation Systematic Country Diagnostics: Pathways to Inclusive Growth (2017); see Annex 1.2 16 For the list of references see Lokshin and Yemtsov (2012). Latest review is provided in Russian Federation Systematic Country Diagnostic: Pathways to Inclusive Growth (2017), http://www.worldbank.org/en/country/russia/publication/systematic-country-diagnostic-for-the-russian- federation-pathways-to-inclusive-growth 10 enforcing any regulations on private sector wage setting. However, considerable political efforts have been deployed to avoid massive lay-offs, including through the administrative means. The 1998 crisis led to a sharp drop in the value of the key minima during 1998-2001. Interestingly, even after growth resumed, the minimum pension and the minimum wage were not adjusted and continued to fall in real terms until 2007. As a result, the minimum pension fell to just a third of the minimum subsistence level, and the minimum wages and unemployment assistance covered hardly a quarter of the poverty line. Over the same 2002-2007 period, the poverty rate in Russia dropped by almost half. In response to the 2008 crisis, the Government became much more concerned with protecting the social minima. Unemployment assistance was de-linked from contributions and increased. All pensions, including minimum pensions were raised (through the valorization of accumulated contributions and introduction of the guaranteed floor for pensions linked to the poverty line) at the additional cost of about 2 percent of GDP.17 These policies have remarkably prevented the increase in poverty and maintained the level of real incomes (despite the economic slowdown the poverty did not change). Due to the available fiscal space, Russia weathered the 2008 crisis relatively smoothly, and, as far as the social assistance system is concerned, without the pressure to improve its efficiency and effectiveness. Table 1: Poverty risk by location (settlement size) and assets for rural population Poverty% Population Poor % Urban < 50,000 15.7% 20,423,655 3,213,755 17% Urban 50,000– 99,999 11.0% 10,756,029 1,183,129 6% Urban 100,000-249,999 9.5% 13,522,783 1,283,418 7% Urban 250,000- 499,999 7.6% 14,771,703 1,126,677 6% Urban 500,000- 999,999 7.3% 14,392,489 1,050,110 6% Urban 1 mln> 3.1% 33,602,427 1,027,196 5% Rural <200 28.6% 1,420,268 405,849 2% Rural 201 – 1000 30.5% 12,892,526 3,930,086 21% Rural 1001-5000 24.4% 13,846,394 3,385,217 18% Rural >5000 26.2% 8,992,567 2,356,770 12% Total 13.1% 144,620,841 18,962,207 100% Rural + small cities 23.1% 57,575,410 13,291,677 70% + have land 25.6% 41,261,985 10,552,739 79% + have unemployed work capable 29.8% 26,257,539 7,818,590 59% Sourceâ€? Author’s calculation using Rosstat Survey of Incomes and Social Programs Participation, 2017; official poverty line, regionally differentiated. The situation changed starting in 2014, when Russia not only faced accelerated inflation, currency devaluation, and decreased real incomes of the population, but also fiscal tightening, caused by a drastic drop in oil prices. The lack of resources to expand social protection brought back the scenario 17 In 2010, the Pension Fund of the Russian Federation for people who retired before 2002 re-calculated (“valorizedâ€?) pension entitlements. For those with work histories before 1992, each year of service was counted as a 1percent increase. This led to a significant increase in a basic pension, especially for old-age retirees. See http://www.pfr.kirov.ru/node/532. 11 of falling real incomes and increased poverty – for the first time since the mid-1990s. To effectively use social assistance as a poverty mitigation/ reduction instrument, it is important to understand the poverty characteristics of the population. Rosstat’s national regionally representative surveys and especially the large 2017 survey reveal large differences across Russian regions in terms of the poverty headcount, ranging from <1percent to over 50percent of population (Table 1). Most importantly, there are peculiarities in the poverty profile that reveal a picture different from what was observed in the 1990s and early 2000s in several important ways. First, the poverty depth or income deficit of the poor is larger than previously assessed18: before receiving social assistance, in many regions an average poor household has monetary income 30- 40percent below its subsistence level; this is RUB 3,000-4,000 per month per household member. Second, there is a high poverty prevalence among the working poor in Russia. As many as 2/3 of the poor live in households where at least one member is working; and as many as 1/3 reside in households, where 2 or more adults are gainfully employed. This striking result can be explained by low salaries: i.e. one-fifth of the poor report wages below the poverty line.19 At the same, Rosstat data show that the majority of poor households (note the number of poor households, not individuals) – have able bodied members who do not work (they are unemployed or inactive). The number of work able unemployed poor in Russia is estimated at 5.2 million, many of whom are women. Table 2 shows the most recent poverty profile in Russia (by location, presence of work capable members and land use in rural areas). The gradient of poverty to be located in deeper, difficult to reach pockets is clear, as well as its connection to some productive assets. Almost 80 percent of rural poor have land and 60 percent have unemployed family members: among all Russia poor they constitute 70 percent of the headcount. Table 2. Poverty risk and children in Russia: 2000-2015 Poverty risk* 2000 2008 2015 All Children (<16) 33.7 18.3 21.4 <7 26.9 15.3 17.3 7-16 36.8 20.8 25.4 All population groups 29.0 13.4 13.3 Poverty risk of children /average 1.16 1.37 1.61 Of which children below 7 0.93 1.14 1.30 Children 7 - 16 years old. 1.27 1.55 1.91 Source: Rosstat. Note* percentage of the group with incomes per person below official subsistence minimum. Third, there is clear social disadvantage and vulnerability among the poor, especially those with children. Every fifth child in Russia lives in a poor household, and this is considered to be a very serious policy concern, and families with children make-up almost 70 percent of the poor. The deepest 18 Previous reports on poverty in Russia – e.g. World Bank 1993, Klugman 1995, World Bank 2005 – consistently portrayed poverty in Russia as shallow. This assessment has been consistent with very high poverty prevalence and related to transition shock – many structurally non-poor households fell into poverty temporarily. However, as poverty receded to about 10 percent of the population, those who were just below the poverty line managed to escape poverty; those who stayed poor despite economic growth were chronically poor in deep poverty (Ravallion 2016). A secondary factor is that official poverty estimates that were used in these previous studies l came from Household Budget Surveys with a number of imputations. In this report we are using direct measurement of disposable household incomes without imputations. It is an indicator of welfare that is closest to the eligibility conditions used by the social assistance in Russia (it uses means test in the form of income test). 19 This is an important finding that merits further investigation. It indicates that the minimum wage policy is not enforced at the regional level (in Russia the minimum wage is set at the level of the regional subsistence minimum, which is also used as a regional poverty line). 12 poverty is found among single parents and families with 3 and more children (Table 2). Especially vulnerable in this group are single parents with young children: 18-36 months of age. Lack of affordable care arrangements for children in this age (18-36 months; prevents many mothers from coming back to work, after paid maternity leave is over (first 18 months; mothers can stay out of work for another 18 months, until a child reaches 36 months, but with symbolic compensation). In addition, lack of skills sought by employers is another barrier for work. Deep poverty experienced by children at such a young age can create irreversible damage to their development later in life.20 Fourth, the survey conducted in three pilot regions within SP RAS and reports based on them demonstrated that the architecture of social protection in Russia is still heavily tilted towards support for the elderly. The pension system is the main contributor to poverty reduction, and some of its features make it a rather powerful instrument. For example, the minimum pension guarantee ensures that every pensioner is entitled to a benefit that is at least equal to the subsistence minimum. If the pension system would stop its payments, poverty in Russia would increase from its current 18 million people to 29 million. However, using this instrument to reduce poverty by increasing pensions will not be effective. As reports have demonstrated, most pensioners are already above the poverty line and expanding assistance to them will have no further impact on poverty. There is a remarkable variation of poverty rates across Russian regions. Those closely mirror differences in economic capacity (as measured by Gross Regional Product—GRP) and real incomes per capita. Figure 1 shows a tight correlation between these indicators. Figure 1. Regional GRP in RBL per capita and regional poverty rate 60 50 40 30 20 10 0 100000 1000000 10000000 Regions have very different fiscal resources depending on their GRP. Figure 1 below depicts the regional difference in the income level and poverty, showing a very close correlation between economic development and poverty directly and indirectly - lack of nation-wide equalization mechanisms that would ensure broadly shared benefits of growth. Social assistance is one of such tools, and this report discusses in detail how it can become more powerful. 20 Grosh et al. 2018 13 Chapter Two Performance of the Social Assistance System: An Update Performance measurement is the process of collecting, analyzing, and/or reporting information regarding how well a system achieves its objectives, or how it delivers on intended results. By measuring performance, informed decisions can be made, and interventions carried out to improve programs. The analysis of the performance of social assistance programs presented in this update relies on several key indicators.21 These include: coverage, beneficiary/benefit incidence, benefit size/adequacy, and poverty/inequality impact. These core performance indicators are found in the ASPIRE database. The impacts of social assistance programs on health or education outcomes, saving behavior, labor supply, fertility, and migration, among others, are not considered; these impacts can only be measured through rigorous impact evaluations, and Russia is behind other countries in terms of collecting and using such evidence. This report presents a set of comparable core indicators over time and vis-à-vis other countries, allowing us to see how the performance of social assistance system in Russia evolves over time and how it fits the global trends. Spending on social assistance (non-contributory programs providing transfers/ income support and social care services) is an important indicator reflecting its scope in each country. Public resources to finance social protection in general and social assistance specifically is an indicator of prioritization in the public policies. For ease of comparisons over time and with other countries it is expressed as percent of GDP. Coverage indicates the absolute number of program beneficiaries or percentage of the population or a given population group that benefits from a given program or combination of them. Coverage is very important because it indicates the size of the program “blanketâ€? in both absolute and relative terms. It is important to know how various population groups (e.g., poor versus non-poor) are covered by the same program. Coverage, in combination with benefit size/adequacy, is determining the program impact. Benefit incidence shows which segment of the population the program benefits go to, e.g. tell what percentage of the total amount of benefits goes to the poor. The calculation of this indicator requires the use of the household survey data that includes the indicator that allows to determine welfare of a household. Benefit level indicates the amount of the benefit, while benefit adequacy is a measure of the relative benefit level. The main purpose of estimating benefit adequacy is to get some idea to what extent the benefit size is “smallâ€? or “largeâ€? in comparison to a benchmark (e.g., average income/consumption in a country, poverty line, minimum subsistence level, minimum wage, per capita GDP). The impact evaluation literature often finds that fragmented/small benefits fall short of achieving desired developmental effects. Poverty impact reveals the distributional effects of the benefit. When it comes to the poverty impact, two indicators are often looked at: (i) reduction (percentage) in the poverty headcount (prevalence) due to the benefit; and (ii) reduction (percentage) in the poverty depth (distance to the poverty line). As empirical evidence around the world suggests, many SSN benefits help poor people become less poor (i.e., reduction in the poverty gap/depth) rather than graduate them entirely from poverty. 21 Aligned with the World Bank data base of indicators: the Atlas of Social Protection: Indicators of Resilience and Equity (ASPIRE) , continuously updated and now containing data from more than 120 countries. Russia is part of the database and in international comparisons harmonized definitions are used. 14 2.1. Spending: Aiming at Allocating Resources to the Highest Needs Consolidated spending on social protection and labor market policies in Russia in 2017 is estimated at 14.3 percent of GDP. Subtracting expenditures on contributory pensions administered by the Russian Federation Pension Fund and contributory programs related to sickness and disability administered through the Social Insurance Fund, one arrives 3.2percent of GDP. Note that there is no officially published figure on this spending, as Statistical Classification adopted in Russia does not distinguish between contributory and non-contributory sources of financing while classifying a social protection program.22 A significant portion of resources to finance social assistance comes directly from the federal budget23. In addition, the federal budget co-finances some of the spending at the regional level (federally mandated benefits), which is shown as regional spending in the consolidated spending figures. As shown in Table 3, the structure of spending at the federal and consolidated regional level has several similarities, as well as differences. Federal budget spending is tilted towards programs aimed at disability, fertility incentives (maternity capital) and compensations due to loss of health, while regional budgets seem to be heavily tilted towards programs of rewards and merit-based payments. Table 3. Social assistance spending by category, consolidated, Federal and regional budgets, in 2017 (percent of GDP) Federal Regional Total Special merits and rewards for service 0.13 0.54 0.67 Special conditions of work and incentives 0.18 0.19 0.37 Children and support to families - Maternity capital 0.39 0.22 0.61 Other Family, maternity and children 0.05 0.27 0.32 Compensation of losses of health and property 0.11 0.09 0.20 Disability and other forms of impairment 0.50 0.10 0.60 Poverty targeted (means tested) 0.04 0.40 0.44 Labor market support and assistance - 0.06 0.06 Other, not classified 0.01 0.10 0.11 Total 1.39 1.81 3.20 Source: Authors’ estimate. Notes: Social services are included in different categories of programs depending on which function (or risk) they cover, total spending on services is 0.4percent of GDP. Regional allocations are estimated based on detailed structure of spending from ASPIRE. Targeted child allowances are included in, not in the “Families and childrenâ€? category. Source: Authors based on the NIFI and the Ministry of Finance data. For detailed description of each category see Annex. The estimates presented in Table 3 show that significant resources are allocated to programs that conceptually do not fit the understanding of social assistance as public interventions aimed at assisting the poor and vulnerable population. As already noted, there are two groups of such programs in Russia: benefits to certain categories of public sector employees and benefits to award/ recognize citizens with merits. Excluding these two expenditure groups from the estimated aggregate results in a significantly smaller public spending on social assistance – 2.16 percent of GDP. This is still considerable spending: above the global average and at the average for the countries in the Europe and Central Asia Region. 22 For similar attempts by Russian researchers see Report by Independent Institute for Social Policy (L. Ovcharova, editor) – see bibliography. 23 Overall, almost 22 million beneficiaries receive social assistance directly from the Federal budget. 15 In the analysis of social assistance system performance, the ASPIRE standard classification was used and salary top ups were excluded from the total social assistance. Excluding merit-based benefits is complicated and will require significant additional time and effort, while making comparisons between Russia and other countries less robust – many countries, similar to Russia count merit-based benefits as part of social assistance system. The Rosstat survey data that are used to analyze performance reflect rather well the budget spending data, as shown in Annex Table 2. We focus on means tested poverty targeted benefits to contrast their performance to the overall spending on social assistance minus salary top ups. We will refer to this aggregate as “coreâ€? social assistance. Merit based benefits represent a significant share of resources counted in Russia as spending on social assistance - over 0.6percent of GDP. Their size and directions are determined by policy priorities and in Russia the often compensate gaps in the social insurance systems (old age and disability pensions). However, even though they are not targeted at the poor, merit-based benefits are considered as part of social assistance in most countries and, hence, affect overall performance of the system and its efficacy from the point of view of impact on poverty reduction. The rest of social assistance spending mostly goes to families and children, individuals and families with disability and to those whose incomes fall below the poverty line (the latter overall claims 0.44percent of GDP). The federal spending on maternity capital constitutes the majority spending on child related benefits (with the regional analogs, maternity capital programs account for 0.6 percent of GDP). Compensations for various disaster related damages comprise 450 measures for 130 categories of the population, but they account for only 0.2percent of GDP. 2.2. Coverage: Inclusiveness of the System for those Who Need Support Coverage measures the percentage of the intended beneficiaries who receive transfers provided through a social assistance program. Often, the coverage is calculated as a share of direct beneficiaries in the population below the poverty line or among the bottom 10 or 20 percent of the population ranked by income or consumption. In this update the enumerator to estimate the coverage comprises both direct and indirect beneficiaries (e.g. the direct recipients and their family members). Figure 2 presents data on social assistance coverage of the poor and other population subgroups over time. The system achieves a very high coverage of the poor – 85 percent, but only slightly over 40 percent of the poor are covered by means tested poverty benefits. The coverage has improved over time, especially for the poor. The coverage of the top 20 percent, on the other hand, has been reduced. 16 Figure 2: Coverage of social assistance in Russia in 2014-2017 (direct and indirect beneficiaries) 85% 80% 75% 70% 65% 60% 55% 50% 45% 40% 2014 2015 2016 2017 Coverage of the poor Coverage of bottom 40% Coverage of the richest 20% Source: Authors’ calculation using Rosstat Survey of Incomes and Social Programs Participation, 2014 -2017. Notes: **Using official poverty line. Indirect beneficiaries are all family members of direct recipients of social assistance. Figure 3 shows the evolution of coverage by regions of Russia. In most cases we see big improvement in coverage of the poor – especially in the Komi Republic, Kaluga and Tambov regions. Some rare cases of a decrease in coverage are in regions with low poverty or special conditions (e.g. Tyumen region and St. Petersburg). Most regions have also improved coverage as it is visible from comparing two dotted lines. There is also less variation in coverage across regions in 2017 compared to 2014 – showing an improved consistency of policies and performance. Hence, the social assistance coverage indicator has some further room for improvement, as no region has yet achieved 100 percent coverage. There are some “last mileâ€? delivery issues such as outreach. All social programs are characterized by heavy bureaucracy, even after the introduction of one-stop or one window integrated offices, improved government services for obtaining documents and strict adherence to service standards (which all represent good practice). Many business processes are antiquated and lead to errors. But overall, the system has a strong foundation to improve. One of important steps is the launch of the new unified registry (ЕГИССО, http://egisso.ru/site/), which brings all data from all social assistance programs on their beneficiaries into an interconnected live database. 17 Figure 3. Changes in Social Assistance Coverage of the Poor by regions of Russia, 2014-2017 100% 90% 80% 70% 60% 50% 40% Altai region Krasnoyarsk Region Astrakhan Region Tomsk Region Republic of Tyva Lipetsk region Murmansk Region Komi Republic Chuvash Republic Irkutsk Region St. Petersburg Kaliningrad region Sakhalin Region Smolensk Region Republic of Buryatia Stavropol Region Kaluga region Republic of Adygeya Republic of Khakasia Amur Region Vladimir Region Nizhny Novgorod Region Kemerovo Region Moscow Kurgan region Novosibirsk Region Tyumen Region Orenburg region Pskov region Belgorod Region Vologda region Kostroma region Chelyabinsk region Jewish Autonomous Region Ulyanovsk Region Chukotka Autonomous District Republic of Kalmykia Penza Region Ryazan Region Republic of Mordovia Republic of Kabardino-Balkaria Republic of Karachay-Cherkesia 2014 2017 Linear (2014) Linear (2017) Source: Authors calculation using Rosstat Survey of Incomes and Social Programs Participation, 2014 -2017. Notes: **Using official poverty line. Indirect beneficiaries are all family members of direct recipients of social assistance. 2.3. Targeting: Ability of the System to Direct Assistance to The Needy Targeting indicator looks at how much of resources allocated to social assistance is received by the poor or lower end of the distribution (in this update, bottom 20 percent or 40 percent of the population). In a broader sense, targeting means ability to reach intended beneficiaries, whoever those may be, for example disabled children, regardless of their living conditions. But the objective of most social assistance programs is to provide assistance to those who need it the most, and the poor are the group that experiences such need. Hence, we will focus our targeting assessment on the benefit incidence or the share of payments that goes to poor families. Targeting is not a salient feature of the social assistance system in Russia. Only 5 programs are means tested: general poverty targeted benefit, child allowances, housing subsidies, allowances for pregnant and lactating mothers, and stipends for higher education for students from less affluent families.24 24 A pension top-up to bring the pension level to the regional poverty line for pensioners, in case it falls below it, is financed and implemented by the Pension Fund of the Russian Federation but governed by law on Social Assistance (since January 2010- amending law 12.1 FZâ„– 178 of 1999). The Pension law does not guarantee this minimum; it is matter of social assistance policy, although one can discuss the merits of targeting this transfer 18 These benefits take several forms. The child allowances that are provided by regions to the poor families according to their own definition of poverty are partially subsidized from the Federal budget. All these targeted programs combined account for 0.4 percent of GDP or 12.5 percent of total estimated spending on social assistance (3.2 percent of GDP). While social assistance coverage of the population is excellent, the ability of the system to direct resources to those most in need has historically been well below the comparators. An influential report published in 200125 looked at the impact of social transfers on poverty. It estimated that only 25 percent of all funds allocated to social assistance programs were going to the poor, concluding that social assistance was not effectively targeted. Subsequent reports have corroborated these findings.26 In 2005, when the social assistance system already had the key features of the present day, the World Bank report on poverty found that still only a quarter of all social assistance benefits was directed to the poorest quintile. It used the first especially designed survey to collect data on the social programs’ participation – NOBUS (conducted in 2003; a predecessor to the Rosstat National Survey of Incomes and Social Programs Participation, which was launched in 2013). In 2016, an assessment by Russian researchers found that social assistance in Russia was not serving the poor.27 The targeting results vary by the type of program, but almost uniformly the non-poor get 70 plus percent of the program budget and in the case of the utility subsidy this percentage jumps even higher – to 85 percent.28 However, once reviewed at the regional level, the detailed program-by-program analysis has revealed that there are some excellent targeted programs that direct close to 90percent of their budget to the poor.29 Figure 4 compares several rounds of the same Rosstat survey covering 2014-2017. One observes more or less a similar situation, although with some changes: in 2014 the share of all social assistance benefits accruing to the poor was only 5 percent, and by 2017 it increased to 10 percent. Partly, this is the result of the increased number of the poor (from 10 percent in 2013 to 13 percent in 2015). There are also some signs of improved targeting performance in means tested benefits administered by regions (especially poverty benefits, child allowances), and less stringent rules for using maternity capital to finance current consumption for the poor families compared to 2013. Figure 4. Russia - Targeting performance of social assistance as a whole in 2014-17. based on household income. While such spending can be traced in the budget data, it cannot be identified in household surveys, as pension payments do not specify the share that is coming as such top-up. Therefore, the pension top-ups cannot be unequivocally classified as targeted or means tested programs. It is however included in the total spending on targeted programs, the aggregate amounts to 0.062percent of GDP. 25 Ovcharova, L.N. et al. 2001. Working Towards a Poverty Eradication Strategy in Russia: Analysis and Recommendations Regional Office of the ILO. 26 Child Welfare Outcomes during the 1990’s World Bank 2003; and Incomes and Social Services: Inequality, Vulnerability, Poverty (2005). 27 Report by Independent Institute of social policy (L.Ovcharova, editor) – see bibliography. 28 Strengthening Social Protection System in the Russian Federation (Nazarov V. and A. Posarac eds), World Bank and NIFI joint report. M., 2017 (In Russian) - Развитие Ñ?ффективной Ñ?оциальной поддержки наÑ?елениÑ? в РоÑ?Ñ?ии: адреÑ?ноÑ?Ñ‚ÑŒ нуждаемоÑ?Ñ‚ÑŒ, универÑ?альноÑ?Ñ‚ÑŒ / Под ред. Ð’. Ð?азарова и Ð?. Пошарац. Ð?аучный доклад. Ðœ., 2017 29 World Bank and NIFI 2019: "Modernization of social assistance in Russia: Survey of households in the Volgograd region. Analytical report "2018," Modernization of social assistance in Russia: Survey of households of the Komi Republic. Analytical report "2018," Modernization of social assistance in Russia: Survey of households in the Leningrad region. Analytical Report» 2019. (in Russian) 19 40% 35% 30% 25% 20% 15% 10% 5% 0% 2014 2015 2016 2017 Share of the poor in total SA transfers Share of bottom 40% Share of the richest 20% Source: Authors’ estimates based on Rosstat Survey of Incomes and Social Programs Participation, 2014-17. Note: Poverty estimate is based on official poverty line and definition of the poor (about 13 percent poverty incidence). Figure 5. Russia - Targeting performance of means tested social assistance in 2015-17. 80% 70% 60% 50% 40% 30% 20% 10% 0% 2015 2016 2017 Share of the poor in means tested transfers Share of bottom 40% in MT programs Share of the richest 20% in MT programs Source: Authors’ estimates based on Rosstat Survey of Incomes and Social Programs Participation, 2014-17. Means tested programs show much better performance relative to the overall social assistance system, but big room for improvement exists. Means tested benefits cover 13 percent of the population, almost exactly as many poor Russia has; however only 44 percent of all poor are included (Figure 5). Most importantly, almost 80 percent of the resources accrue to the bottom 40 percent of the population. However, the leakage of one fifth of resources in these poverty targeted programs to the top 60 percent shows that there is room for tightening. Over time this indicator has improved (Figure 5). Figure 6 shows concentration curves for the main types of social assistance benefits in Russia. Child allowances are the best targeted group of programs; but even there only 36 percent of resources are received by the poorest population quintile and only 53 percent by the poorest 40 percent. 20 Resources are spread over many programs, often contributing little to the income of the beneficiaries: most programs constitute 2-3 percent of the income of their recipients; the highest being 8 percent in the case of children related benefits. This is well below the well performing programs benchmarks of at least 15 percent of the income of the poorest. Figure 6. Concentration Curves for Main Types of Social Assistance 100 90 80 Cumulative percent of income 70 60 50 40 30 20 10 0 0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 Deciles of Per Capita Consumption All labor market programs All social assistance Child allowances of all types Targeted poverty benefit Housing allowances Stipends Source: Income and Program Participation Survey, Rosstat 2016. Note: The main categories of transfers based on Rosstat classification. Looking across regions, one can see significant scope for improvement. Figures 7 and 8 summarize targeting outcomes by regions for the system as a whole and for the means tested Looking at the share of total assistance payments going to the poor by regions, one observes a better picture than in Russia on average.30 If for Russia, only 10 percent of social assistance resources accrues to the poor, among regions a typical share is closer to 15 percent (Figure 7). There are many regions where this share is close to or above 20percent. There has also been marked improvement over time as shown by the trend lines. Figure 8 depicts the same parade of targeting accuracy, this time only for means tested programs. As for the country as a whole, targeting is away better and shows signs of significant improvement. Some regions have achieved close to 90 percent accuracy, and the trend also is strongly positive for most regions, showing more resources flowing to the poor. Figure 7. Changes in Targeting Accuracy of Social Assistance by Regions of Russia, 2014-2017. 30 Results are available on the ASPIRE web site of NIFI. 21 0% 10% 20% 30% 40% 50% 60% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Altai region Altai region Krasnodar region Krasnodar region Krasnoyarsk Region Krasnoyarsk Region Primorsky Krai Primorsky Krai Stavropol Region Stavropol Region Khabarovsk Edge Khabarovsk Edge Amur Region Amur Region Arkhangelsk Region Arkhangelsk Region Astrakhan Region Astrakhan Region Nenetsky Autonomous District Nenetsky Autonomous District Belgorod Region Belgorod Region Bryansk Region Bryansk Region Vladimir Region Vladimir Region Volgograd region Volgograd region Vologda region Vologda region Voronezh Region Voronezh Region Nizhny Novgorod Region Nizhny Novgorod Region Ivanovo region Ivanovo region Irkutsk Region Irkutsk Region Republic of Ingushetia Republic of Ingushetia Kaliningrad region 2014 2014 Kaliningrad region Tver Region Tver Region Kaluga region Kaluga region Kamchatka Region Kamchatka Region Kemerovo Region Kemerovo Region Kirov region Kirov region Kostroma region Kostroma region Samara Region Samara Region Kurgan region Kurgan region Kursk region Kursk region St. Petersburg St. Petersburg Leningrad region Leningrad region 2017 2017 Lipetsk region Lipetsk region Magadan Region Magadan Region Moscow Moscow Moscow Region Moscow Region Murmansk Region Murmansk Region Novgorod region Novgorod region Novosibirsk Region Novosibirsk Region Omsk Region Omsk Region Orenburg region Orenburg region Orel Region Orel Region Penza Region Penza Region Perm Region Perm Region Pskov region Pskov region Rostov region Rostov region Ryazan Region Ryazan Region Saratov region Saratov region Sakhalin Region Sakhalin Region Sverdlovsk Region Sverdlovsk Region Linear (2014) Linear (2014) by the region-specific poverty line and their disposable money incomes. Smolensk Region Smolensk Region Tambov Region Tambov Region Tomsk Region Tomsk Region Tula Region Tula Region Tyumen Region Tyumen Region Khanty-Mansiysk a O Khanty-Mansiysk a O Ulyanovsk Region Ulyanovsk Region Yamalo-Nenets Autonomous Yamalo-Nenets Autonomous Chelyabinsk region Chelyabinsk region Trans-Baikal Region Trans-Baikal Region Chukotka Autonomous District Chukotka Autonomous District Yaroslavl Area Yaroslavl Area Republic of Adygeya Republic of Adygeya programs use poverty as eligibility criterion (see Nazarov and Posarac eds for definition). Republic of Bashkortostan Republic of Bashkortostan Republic of Buryatia Republic of Buryatia Republic of Dagestan Republic of Dagestan Republic of Kabardino-Balkaria Republic of Kabardino-Balkaria Linear (2017) Linear (2017) Republic of Altai Republic of Altai Republic of Kalmykia Republic of Kalmykia Republic of Karelia Republic of Karelia Komi Republic Komi Republic Republic of Mari El Republic of Mari El Republic of Mordovia Republic of Mordovia Figure 8. Changes in Targeting Accuracy of Means Tested Programs, 2014-2017. Republic of North Ossetia Republic of North Ossetia Republic of Karachay-Cherkesia Republic of Karachay-Cherkesia Republic of Tatarstan Republic of Tatarstan Republic of Tyva Republic of Tyva Udmurt Republic Udmurt Republic Republic of Khakasia Republic of Khakasia Chechen Republic Chechen Republic Chuvash Republic Chuvash Republic Note: Targeting accuracy is defined as a share of total social assistance transfers accruing to the poor as defined Targeting accuracy is defined as a share of total social assistance transfers accruing to the poor. Means tested Source: Authors’ estimates based on Rosstat Survey of Incomes and Social Programs Participation, 2014-17. Source: Authors estimates based on Rosstat Survey of Incomes and Social Programs Participation, 2014-17. Note: 22 Republic of Sakha (Yakutia) Republic of Sakha (Yakutia) Jewish Autonomous Region Jewish Autonomous Region Figures 9 and 10 put the targeting accuracy outcomes in the context of prevailing regional poverty. Regions in Russia have hugely divergent poverty rates. Directing a certain percentage of benefits to the poor is “easierâ€? in a region with 50 percent poverty than in a region with 2 percent poverty. To simplify the assessment of how strong the targeting differential is, the figures contain “random allocationâ€? lines, showing expected share of the benefit accruing to the poor in each region had it been done universally for everyone in the same amount regardless of need of social status. Figure 9 shows the share of the poor in 2017 in each region versus the regional poverty rate for all social assistance transfers. When compared to a “random allocation lineâ€? we see a complete absence of any prioritization of the poor in actual distribution of social assistance payment. This is especially noticeable among poor regions, where poor are receiving less than even their share in the population. Only 15 regions in Russia, as shown in this graph, are above the “random allocation lineâ€? attaching some priority to the poor, and these are not the poorest regions. This calls for attention to the overall architecture of incentives and budget transfers as a potential instrument for stronger social assistance poverty orientation. Figure 9. Poverty rates and share of social assistance accruing to the poor, by regions, 2017 60% Share of social assistance transfers going to the poor percent Random allocation 50% 40% Targeting outcome 30% 20% 10% 0% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% Poverty rate Source: Authors’ estimates based on the Rosstat Survey of Incomes and Social Programs Participation, 2017. Note: Targeting accuracy is defined as a share of total social assistance transfers accruing to the poor. Figure 10 does the same exercise for means tested social assistance. As for the aggregate picture that we have discussed above, this one shows radically different outcome. No region is below the “random allocationâ€? line. Best performers achieve close to 90 percent of their benefits going to the poor. The main concern is huge variation: among the regions with the same poverty rate the share of benefits going to the poor in programs designed to be serving them varies dramatically, for instance among 23 regions with close to 20percent poverty rate from barely 20 percent (Ulyanovsk oblast) to 85 percent (Primorsky kray). Figure 10. Regional poverty rates and share of means tested benefits to the poor, 2017 100% Targeting outcome Share of social assistance transfers going to the poor percent 90% 80% 70% 60% Random allocation 50% 40% 30% 20% 10% 0% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% Poverty rate Source: Authors’ estimates based on the Rosstat Survey of Incomes and Social Programs Participation, 2017. Note: Targeting accuracy is defined as a share of total social assistance transfers accruing to the poor. 2.4. Adequacy: Providing Support Congruent to the Needs Adequacy is measured as the share of social assistance in the income of beneficiaries. Sometimes the shares (in total, by groups of programs, or by individual programs) are also compared to the poverty line, a useful comparison to assess the potential of social assistance to assist those below the poverty line in a meaningful way. Adequacy is determined by the size of the transfer. In Russia, social assistance programs come in all sizes - from very generous (e.g. maternity capital of about US$ 8,000 per mother; or grants for productive engagement through the so called “social contractsâ€? that may be as high as US$ 20,000 per beneficiary) to practically immaterial (the monthly child allowance, which has the highest coverage provides only $2.5 a month per child).Using the Rosstat Survey on Incomes and Social Programs Participation, 2016, we have estimated that all social assistance in Russia make up about 25 percent of the pre-transfer income of the poorest quintile. The average income deficit of the poor in Russia after transfers is only slightly below 20 percent of the poverty threshold. Hence, the benefits provided through assistance to the poor are inadequate to help them to move out of poverty. This figure is especially striking when compared with the sum of all social assistance transfers for the poor, which average RUB 1,000. Thus, an average poor person, even if correctly identified by the means-targeting system as in need of assistance will remain poor after receiving all available benefits. 24 At the same time, some among the poor, especially the pensioners, have much smaller poverty deficit (RUB 100-500), but they receive bigger transfers (RUB 2,000 per month per person). This lack of connection between the need and the level of assistance is the main source of low efficacy of social assistance in Russia in terms of its impact on poverty reduction. The only way to address this gap is to: (i) identify correctly income of low-income households (in other words, identify correctly means with no disregards and exceptions); and (ii) calibrate assistance as difference between means and the poverty threshold. In other words, introduce a guaranteed minimum income for low income households. This would imply that in many cases the level of assistance would need to go up. Figure 11 illustrates how adequacy of social transfers has changed in Russia over 2014-17. The Figure show that there was practically no progress towards improving adequacy over time. Figure 11. Adequacy (share of social assistance transfers in incomes of beneficiaries), 2014-2017 35% 30% 25% 20% 15% 10% 5% 0% 2015 2016 2017 Adequacy of total SA transfers for the poor Reduction of poverty headcount Reduction of poverty deficit Source: Authors’ estimates based on the Rosstat Survey of Incomes and Social Programs Participation, 2014- 2017. Note: Targeting accuracy is defined as a share of total social assistance transfers accruing to the poor. Most beneficiaries below the poverty line receive very little assistance that is not enough to mitigate their poverty in a meaningful way. For instance, poverty benefit does not exceed 15 percent of the poverty line, monthly child allowance is only 3 percent of the beneficiaries’ income in the poorest 20percent (families with children constitute the majority of the poor in Russia). In contrast, benefits that are not means tested are often equal or higher than the poverty line (e.g. merit-based benefits). Figure 12 describes regional performance in terms of adequacy, presenting regions (sorted by their income level). The presented statistic is comparing the sum of all social assistance transfers and the poverty deficit. The regions are sorted from less poor (left) to the poor (right). While the social assistance overall budget in the richest regions in Russia far exceeds the needs of the poor, resources in the poor regions are insufficient to fill in the poverty gap. 25 Figure 12. Social Assistance Overall budget in Relation to Poverty Deficit, by regions, 2017 400% Social Assistance transfers to overall poverty deficit 350% 300% 250% 200% 150% 100% 50% 0% St. Petersburg Magadan Region Kursk region Vladimir Region Chelyabinsk region Kaliningrad region Kemerovo Region Vologda region Kostroma region Amur Region Tyumen Region Smolensk Region Republic of Khakasia Republic of Sakha (Yakutia) Republic of North Ossetia Chechen Republic Murmansk Region Tomsk Region Sverdlovsk Region Ivanovo region Penza Region Kamchatka Region Orel Region Altai region Republic of Kalmykia Moscow Leningrad region Kaluga region Yaroslavl Area Tambov Region Udmurt Republic Tver Region Republic of Tatarstan Nizhny Novgorod Region Chukotka Autonomous District Novgorod region Sakhalin Region Lipetsk region Krasnodar region Rostov region Volgograd region Omsk Region Orenburg region Pskov region Belgorod Region Voronezh Region Republic of Bashkortostan Khabarovsk Edge Republic of Mordovia Ryazan Region Bryansk Region Primorsky Krai Ulyanovsk Region Astrakhan Region Republic of Adygeya Kurgan region Republic of Karachay-Cherkesia Arkhangelsk Region Nenetsky Autonomous District Trans-Baikal Region Republic of Altai Tula Region Yamalo-Nenets Autonomous Samara Region Perm Region Chuvash Republic Republic of Karelia Republic of Tyva Moscow Region Khanty-Mansiysk a O Saratov region Komi Republic Novosibirsk Region Stavropol Region Kirov region Irkutsk Region Krasnoyarsk Region Republic of Mari El Republic of Dagestan Republic of Kabardino-Balkaria Republic of Buryatia Jewish Autonomous Region Republic of Ingushetia Regions sorted from wealthiest to the poorest Source: Authors’ estimates based on the Rosstat Survey of Incomes and Social Programs Participation, 2017. 2.5. Impact of social assistance in Russia on poverty reduction. To present the impact of social assistance on poverty reduction one may look at variations across regions. Figure 13 presents the impact of social assistance transfers on poverty by regions. While the variation in terms of how much poverty reduction is taking place due to transfers is really large, most regions do not achieve much. And over time the trend is negative: less impact on poverty in 2017 on average, compared to 2014. Aggregating this to all Russia, Figure 14 compares the pre-transfer poverty gap of all poor households in Russia (about 600 billion Rubles, or 0.8 percent of GDP) to social assistance transfers (whose total budget in 2014 was about 2.0 trillion Rubles), and the poverty gap reduction (by 260 billion rubles: to 337 billion, or 0.45 percent of GDP). The rest of the public spending on social assistance (1,623 billion Rubles accrues to the non-poor population.31 Figure 13. Reduction in the headcount (percent) due to poverty assistance in Russia, 2014-2017 31 According to the official data, poverty gap was at its highest level in 1992, at 6.2 percent of total income of the population, and has been falling since then, especially after 2000. By 2010, it stood at 1.2 percent. By 2015, the estimate provided by Rosstat was less than 1 percent of GDP. It means that under the assumption of perfect targeting it would be sufficient to mobilize 1.2 percent of total household income (less than 1 percent of GDP) to bring all the poor above the poverty line. 26 32 0% 10% 20% 30% 40% 50% 60% 70% 80% Altai region Krasnodar region Krasnoyarsk Region Primorsky Krai Stavropol Region Khabarovsk Edge Amur Region Arkhangelsk Region -1000 0 500 1000 1500 2000 2500 -500 Astrakhan Region Nenetsky Autonomous District Belgorod Region reduced by 25percent Bryansk Region Vladimir Region Volgograd region Vologda region Voronezh Region Nizhny Novgorod Region Ivanovo region Irkutsk Region Republic of Ingushetia Kaliningrad region 2014 -596 Tver Region Kaluga region Kamchatka Region TRANSFERS Kemerovo Region Kirov region Kostroma region SOCIAL ASSISTANCE INCOME GAP BEFORE Samara Region Kurgan region Kursk region St. Petersburg Leningrad region 2017 Lipetsk region Magadan Region Moscow Moscow Region Murmansk Region Novgorod region Novosibirsk Region Omsk Region Orenburg region Orel Region -337 Penza Region Perm Region Pskov region Rostov region Ryazan Region Saratov region ASSISTANCE TRANSFERS Sakhalin Region Sverdlovsk Region INCOME GAP AFTER SOCIAL Linear (2014) Smolensk Region Tambov Region Tomsk Region Tula Region Tyumen Region Khanty-Mansiysk a O Ulyanovsk Region Yamalo-Nenets Autonomous Chelyabinsk region Trans-Baikal Region Chukotka Autonomous District Yaroslavl Area 259 Republic of Adygeya To Non BUDGET Republic of Bashkortostan Poor 1623 Republic of Buryatia Republic of Dagestan Republic of Kabardino-Balkaria SOCIAL ASSISTANCE Linear (2017) Republic of Altai Republic of Kalmykia Republic of Karelia Komi Republic Figure 14. Social Assistance and Poverty Income Gap in 2014 (Billion Rubles) Republic of Mari El Republic of Mordovia Republic of North Ossetia Republic of Karachay-Cherkesia Republic of Tatarstan Republic of Tyva Udmurt Republic Source: Authors’ estimates based on the Rosstat Survey of Incomes and Social Programs Participation, 2014 and 2017. Republic of Khakasia Using official poverty line definition. Using international standard in ASPIRE (20percent) the poverty gap is Chechen Republic Assistance (this report). Note: Official poverty line. Social assistance overall budget is as captured in the data. Chuvash Republic Source: Rosstat Survey of Income and Social Programs Participation, 2015 and NIFI compilation of data on Social weak outcome relative to the overall spending (2.8percent of GDP excluding salaries top ups and most 27 Republic of Sakha (Yakutia) Hence, the social assistance in Russia reduces the poverty deficit by about 28 percent,32 which is rather Jewish Autonomous Region of merit-based benefits; or 3.2 percent including the full count of transfers counted as social assistance transfers in Russia). Countries with similar level of spending and even countries with a much lower level of spending achieve significantly more in terms of poverty alleviation. The fact that significant resources allocated to social assistance in Russia do not do much to alleviate poverty raises concerns about the system design and performance. Only a fraction of the total consolidated social assistance budget (0.4 percent of GDP) is spent on programs with some form of a means test. Delivering all means tested programs in Russia to the poor (perfect targeting) will not be sufficient to cover the deficit of the poor and eradicate poverty, as the budget for all means tested programs (256 billion Rubles is way below the income deficit after social assistance transfers (337 billion Rubles). The system needs to deliver more resources to the poor, which would require some reallocation across programs. Russia can aspire to achieve better benefit incidence and progressivity of the system as a whole. Unfortunately, in many regions the social assistance system performs even below the Russia’s national average, suggesting that there is a huge untapped potential at the regional level to improve the results of social assistance policies. 2.6. International Comparisons: Russia in the Context of Global Trends and Benchmarks Russia’s level of spending on social assistance is well above the global average of 1.6 percent of GDP.33 Even compared to richer than average countries with higher fiscal capacity, this level is high. Russia’s spending on social assistance, either as full account of all non-contributory social transfers (3.2 percent of GDP) or WB ASPIRE methodology adjusted spending (2.8 percent of GDP) is above that in Brazil, China, Portugal and Mexico, higher than in most former socialist economies, on a par with Chile, similar to South Africa, and relatively comparable to Greece, Croatia and Italy. It is significantly lower than in Germany and Belgium and Nordic European countries (Figure 15). Figure 15. Spending on non-Contributory Social Assistance and Social Services Programs, in Percent of GDP 10 9 8 7 6 5 4 3 2 1 0 Brazil Turkey Serbia Vietnam Romania Chile Danemark Thailand Latvia Italy Finland China Poland Moldova Mexico Bulgaria Greece Croatia Germany Tunisia Azerbaijan Estonia Belgium Peru Portugal Pakistan Russian Federation Tajikistan Uruguay Armenia Sweden South Africa Source: World Bank’s Atlas of Social Protection: Indicators of Resilience and Equity (ASPIRE). In all cases countries’ own definition of social assistance is used. 33 World Bank 2015 The State of Social Safety Nets 2015. 28 Russia runs an inordinate number of social assistance programs. An inventory of social assistance programs undertaken for this study counted 756 federal program (rounded at 800 above) and on average about 150 regionally mandated programs in region34. Some of the programs are tiny with just a few beneficiaries, some cover over half of the population. Typically, each is governed by its own law/ legal act stipulating a complex set of rules and instructions. The number of social assistance programs in Russia compares to an average of 40 social assistance programs in a given developing country (based on ASPIRE, World Bank 2018). Russia tends to be on the higher end of the spectrum among the countries with a high degree of fragmentation in social assistance programming. For example, Greece has about 200, India has 950 federally financed schemes, Mexico has close to 3,000 Federal, state level and municipality level social assistance programs. On the other hand, many countries avoid creating multiplicity of programs while committing serious resources to the social assistance: e.g. Turkey has 40 programs that cover most of the poor, providing various forms of assistance covering key socio-economic risks35. The graphs below (Figure 16) depicts coverage comparing Russia to a group of benchmark countries. To make data comparable, here we use a different definition of poverty: in all countries, it is the poorest 20percent of the population before social assistance transfers. It shows that Russia’s social assistance coverage is outstanding – comparable to the best performing countries in the world and exceeding the rates of many OECD countries. However, given the legal guarantee that the system should cover everyone with low incomes, there is still a coverage gap. Figure 16. Coverage of the poorest 20percent of population by social assistance programs, latest available year 100 90 80 70 60 50 40 30 20 10 0 Peru Latvia Romania Italy Poland Argentina Bulgaria Greece Moldova Sweden Portugal Hungary South Africa Turkey Albania Vietnam Finland Tajikistan Armenia Brazil Estonia Belgium Danemark Mexico Georgia Chile Croatia Phillippines Uruguay Russian Federation Pakistan Source: Authors’ calculation based on ASPIRE and Rosstat data for Russia (2016); income is post-transfer. 34 Ranging from 37 programs in the Baikal Krai to 187 programs in the Moscow city. 35 The World Bank 2017. Turkey: Social Assistance Review. Report. Ankara. World Bank Country Office. 29 Figures 17-19 show the coverage across quintiles of income distribution in Russia and a group of comparator countries (Figure 17), distribution of social assistance across quintiles of the population (from poorest to richest) – Figure 18, and the level of social assistance transfers as a percent of incomes of beneficiaries (Figure 19). While Russia, as pointed above, stands out as a country with high coverage, is also exhibits one of weakest targeting outcomes (note extremely high inclusion rates for the top richest quintile, while the poorest 20 percent of the population represents just 25 percent of all beneficiaries). Figure 17. Benchmarking Social Assistance in Russia: Coverage and Benefit Incidence Panel A. Coverage percent of the population by quintile 100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 Russia South Hungary Mexico Peru Latvia Poland Turkey Argentina Brazil Serbia 2016 Africa 2007 2012 2014 2009 2012 2014 2013 2012 2013 2010 Total Q1 poorest Q2 Q3 Q4 Q5 richest Source: Authors calculation based on ASPIRE and Rosstat data for Russia (2016); income is post-transfer. Figure 18. Tageting: Percent of Beneficiaires by Quintiles of Pre-Transfer Income 100 90 80 70 60 50 40 30 20 10 0 Q1 (poorest) Q2 Q3 Q4 Q5 (richest) Source: ASPIRE. Note Pre-transfer incomes are based on V1 definition in ADEPT. 30 Russia does not seem to be an outlier in terms of the average adequacy of social assistance (Figure 19). The figure shows that in Russia social assistance constitutes on average 24 percent of the pre- social assistance transfer incomes of the poorest 20 percent of the population, which is similar to that in Chile and Belgium, lower than in Armenia, Brazil in Mexico (all of which rely on a single large program to assist poor – Targeted Family Poverty Benefit in Armenia, Bolsa Familia in Brazil and Prospera in Mexico) and significantly lower than in Denmark and Poland, for example. This comparatively favorable level of adequacy can only be explained by bundling, e.g. the fact that one household benefits from a number of programs, given that most of the programs in Russia deliver very small benefit amounts. Figure 19. Adequacy of social assistance benefits for the lowest quintile (percent to beneficiaries’ pre-transfer income): Russia compared to other countries, latest available year 50 45 40 35 30 25 20 15 10 5 0 Source: Authors based on ASPIRE and Rosstat data (2017). Note: data pertains to the post-transfer income. How does Russia’s targeting performance compare internationally? Figure 20 puts Russia in an international perspective. As done above, to enable consistent comparisons across countries, here, the poverty rate is assumed to be equal to the bottom 20 percent of the population. Figure 20. Comparative targeting performance of social assistance systems. Benefit incidence for the lowest quintile (percent): Russia compared to other countries, latest available year 31 100 90 80 70 60 50 40 30 20 10 0 Latvia Romania Bulgaria Hungary Peru South Africa Greece Argentina Turkey Sweden Albania Moldova Italy Mexico Portugal Poland Vietnam Armenia Brazil Danemark Estonia Tajikistan Georgia Belgium Finland Russian Federation Chile Uruguay Phillippines Croatia Pakistan Source: Authors’ estimates based on ASPIRE and Rosstat data for Russia (2016). Note: Post-transfer income. Figure 20 shows that Russia performs worse than most countries. This is in contrast to a very high coverage rate. This contradiction: high coverage, low targeting makes Russia an outlier, because typically, countries with high coverage also deliver more benefits to the poor: Denmark, Peru, Poland, and Vietnam. In Russia, due to the design of the system, high coverage of the poor is not translated into higher share of benefits accruing to the poor. While no country in the world focuses its social assistance system exclusively on the poor, Russia’s weak orientation towards lower income groups stand out, suggesting a huge room for improvements. What do these comparisons tell about modern social assistance, and what trends are important to follow? Countries are very diverse in the way they use social transfers in their social policy, with no unified patterns. However, there are important common tendencies that help better understand trajectories of each national system. The three modern global trends in social assistance are: (i) ensuring that no one is left out, (ii) helping beneficiaries graduate from poverty, and (iii) increasing efficacy and flexibility through the use of modern technology. First trend is the extension of coverage, to ensure that anyone who needs social protection can access it at any time and receive adequate support. This is sometimes called “universal social protectionâ€?36 and refers to the integrated set of policies designed to ensure income security and support across the life cycle. As we have shown, Russian social assistance system achieves very high coverage, although there is a gap to the full coverage that must be bridged. At the same time, the system does not provide adequate support to those who need it most. Second trend is to achieve greater results in terms of poverty reduction, tailor the assistance to specific circumstances and use transformative force of social protection to help beneficiaries graduate from poverty or achieve other higher-level goals, such as investment in human capital, productive inclusion and social cohesion. Participation in social protection programs is a tool for changing behavior to break the poverty and social exclusion cycle. One of the specific forms of such trend is expansion of conditional cash transfers programs or social safety nets with accompanying measures tailored to the specific circumstances of beneficiaries. Russia does not have conditional cash transfers and its nascent graduation programs remain small. 36 Joint Statement by World Bank Group President Jim Yong Kim and ILO Director General Guy Ryder 32 Third trend consists of wider use of technological solutions to make social protection more efficient, effective, flexible, dynamic and responsive at a minimum cost. Technology can help alleviate administrative burden for case workers, free up time for critical tasks, improve decision-making, and deliver better and faster services while reducing costs to beneficiaries. Russia is in the process of building nation-wide integrated social registry of social protection beneficiaries. The single window operations of social assistance offices form part of a broader e-government strategy. However, more can be done to streamline administration processes and use of technology along the delivery chain to minimize costs.37 Chapter Three Renewing the Social Assistance System in Russia To understand the key determinants of today’s performance of the social assistance system in Russia, it is important to keep in mind that this system has developed for almost a century and that what one observes today amalgamates different political and socioeconomic priorities from very diverse times. Russia has inherited a social assistance system from the Soviet Union: social benefits and guaranties existed alongside the policies of full employment, government regulation of prices and wages and extensive subsidies for consumer goods and services. During the transition period to the market economy, faced with severe crisis during the 1990s, the policy response was to expand the number of social protection benefits, as well as the categories of the population entitled to them, without exploring options to reorient the system towards lower income strata. Today, the system is faced with a critical challenge in its evolution. In the mid-2018, the Russian Government announced its strategic development goals of halving poverty by 2024- together with increased life expectancy, improving housing conditions and innovations in education. These goals were formulated as a renewal of the social contract between the government and the society and have challenged the status quo in social policy. Estimates show that return of the economy to growth is not sufficient to halve the poverty rate in six years: the incomes of the bottom quintile will have to grow at a pace of at least 5 percent per year – a pace not observed since the early 2000s. Social assistance is uniquely placed to provide much needed impulse to living standards of the poor and help achieve ambitious targets. But to play its role, the system would need to reorient itself towards the low-income end of the distribution. The recent policy decisions, e.g. to means-test new benefits to families with children, A Unified Plan for the Implementation of National Projects and instructions to regions to increase the number of means tested programs and allocate more resources to them, signal that the system may have begun to respond to the challenge. Specifically, the Unified Plan for the Implementation of National Projects of the Government of Russia38 contains specific measures to ensure greater targeting as a mean to achieve national objectives for poverty reduction: (i) direct more resources towards means tested programs, (ii) increase the adequacy of support to the poor families with children, (iii) employment support to mothers and expansion of kindergartens and child care services, (iv) promotion of self-employment, (v) establishing interconnectivity of Government databases to improve accuracy of targeting. Furthermore, a review of regional social assistance systems reveals that some regions have embarked on a reform, providing a good practice example about what and how it could be changed. 37 Leite, Phillippe; George, Tina; Sun, Changqing; Jones, Theresa; Lindert, Kathy. 2017. Social Registries for Social Assistance and Beyond: A Guidance Note and Assessment Tool. Social Protection & Labor Discussion Paper; No. 1704. World Bank, Washington, DC. 38 http://static.government.ru/media/files/j8IV1FkssLpUqI89JCXZ2mLiIiLEn7H8.pdf 33 3.1 Understanding the drivers of performance The system has been changing, but slowly: a short history since 1990s Russia has inherited a social assistance system from the Soviet Union: social benefits and guaranties existed alongside the policies of full employment, government regulation of prices and wages and extensive subsidies for consumer goods and services. During the Soviet times, the number of social protection benefits had been continuously expanded, as well as the categories of the population covered by them. As a result, in the early 1990s when Russia started its transition to a market economy, there were 150 types of social assistance benefits and 200 categories of beneficiaries. The onset and early years of transition brought tremendous economic insecurity and the poverty shot up to one-third of the population. Thus, the social assistance system faced enormous demands from the population but was limited by a severely constrained fiscal space. This period saw some partial reforms, but also incessant expansion of programs and categories of the population entitled to benefits, often without a clear strategic framework. Nonetheless, there were some positive changes. A methodology for means testing was developed in the mid-1990s and 20 regions introduced targeted social assistance programs. In mid 1990s several regions piloted a new model of targeted social assistance with the support from the World Bank (Komi Republic, Voronezh and Volgograd regions). They tested specific means testing approaches and assessed feasibility of providing assistance aimed at poverty reduction. In addition, for the first time, new technology was used to digitalized social assistance information and assist in decision making. The recommendations from these pilots were: to establish the unified principles for targeting at the Federal level, treat household as a unit of assistance, and take its overall living standards into consideration while deciding on eligibility. Partly informed by the pilot results, an important framework law was voted in 1997: The Federal Law on Subsistence Minimum in the Russian Federation. It defined a minimum subsistence level and established eligibility criteria for enrollment in income support programs for the poor. Targeted social assistance was defined as benefits in cash or in kind (goods and services) provided to those with insufficient level of income. This principle, which is in line with international practice, was also enshrined in 1999 Law â„– 178 «The Law on State Social Assistance». This law defined eligibility condition for social assistance as having per capita income in the household below subsistence minimum, defined by the governing body of the region. Main types of targeted social assistance programs defined by this law: child allowances, housing benefit (subsidy) and poverty income support, remain the key means tested programs to this day. The only addition to this list is a pension top up to bring the pensions to the poverty line introduced in 2010.39 This law, however, did not establish the new foundation of the system, as pre-existing body of legislation remained in place. The impetus for efficiency reforms weakened during the 2000-2008 period with its rapid increase in wages, reduction in poverty and comfortable fiscal space. However, two important changes took place: a shift, albeit partial, from in-kind towards cash benefits and a shift from centralized to a decentralized system. First, monetization of social assistance benefits that previously were provided in kind, although only partially done, helped to modernize the system. The share of cash benefits in total social assistance increased from 55 percent in 2008 to 72 percent in 2013. In 2005, the institutional responsibility for social assistance was devolved to the regions. In practical terms this means that a large part of the non-contributory programs became constitutional mandate that regions must provide using their own budgetary resources. This includes implementation of the federally mandated non-contributory programs (e.g. child and family allowances, housing subsidies, compensations for in-kind privileges), but also own regional programs legislated by the regional 34 parliaments: mostly pension related supplements (to the “veteransâ€? of labor), benefits for special categories, and disasters-related compensations. In such a system, poor regions have less capacity to finance adequate programs for more people in need of assistance. To mitigate this a complex system of inter-budgetary transfers was instituted in Russia to somehow equalize regions and provide more resources to those with greater needs. The decentralization and devolution of social assistance functions to the regions, while challenging regional fiscal space, particularly in less developed regions with higher poverty rates, resulted in multiplication of federally mandated benefits at the regional level. But it also has had some positive consequences. It provided an impetus for the regions to innovate: many have used information technology to improve social assistance operations and delivery. Some have replaced normative/coefficient-based subsidies to utilities providers by a compensation system based on actual use, resulting in significant savings. However, means testing as a criterion for program eligibility remains more of an exception rather than a mainstream practice. As a result, the fragmentation, overlap, as well as the lack of pro-poor orientation of social assistance measures persists causing inefficiencies and low effectiveness. In the early 2000s, Russia started implementing a form of social assistance aiming to graduate families form poverty on sustainable income generation paths. The form is called “social contractâ€? and is an agreement signed between a citizen and social protection unit that obliges the government to provide social assistance to the citizen, and recipient of this assistance has to fulfill certain steps presupposed by an agreed upon “social adaptationâ€? program (work at individual household small land plot, professional education, job search, etc.). It is among few programs in Russia that has a clear set of objectives. But only ½ of regions actively use the program form, and its coverage remains low: in 2017 only 5 percent of all beneficiaries of poverty targeted programs and 1.3percent of all poor in the country. Since 2014, some regions have used their legal power to introduce means testing in social assistance programs they are responsible for40. As many as 56 regions have tightened means-testing and/or introduced means testing in some categorical benefits. However, the power of regions to improve targeting is limited by the legal requirement not to worsen condition of any existing beneficiary. As a result, the regions typically tighten criteria for programs which are already means tested, and not for merit-based programs – in part reflecting greater political voice of special interest groups. The fiscal crisis and reduction of resources available to regions through transfers from the federal budget over 2014-2016 have led to a significant fall in the regional social assistance spending: most regions have tried to balance their budgets by reducing the size of benefits or tightening eligibility conditions. Some have used the crisis as an opportunity to restructure the system, including the following: (i) introducing means testing into previously categorically targeted programs or introducing additional conditions for eligibility discriminating against more affluent households; (ii) simplifying and unifying legal provisions through introduction of social codes, discontinuing inefficient programs or consolidating/merging similar benefits; (iii) introducing verification of information provided through unified social registries, which enable information exchanges and cross-verification from various sources; (iv) fully automatizing administration of programs through integrated management information systems. Key drivers of performance Several inter-connected features determine low efficacy of the current social assistance system: • Lack of conceptual clarity 40 Art 26.3. and 26.3-1 of the law legislated in 1999 â„– 184 «On principles of organization of l egislative and executive powers in the regions of the Russia Federation». 35 • Lack of clear objectives, targets, and weak accountability for results • Multiplicity of laws and regulations • Programs composition and resource allocation that favor untargeted programs • Differences in interpretation of targeting, inconsistency across methods for means testing • Outdated verification procedures and underutilization of modern technology • Poor outreach and communication Each of these issues is discussed below. Lack of conceptual clarity Several large groups of programs counted as social assistance in Russia have objectives different from what is typically understood as social assistance. They include top ups to salaries (mostly a hardship allowance) of some categories of public sector employees such as teachers, health workers, public order and security, etc.; rewards to citizens with merits; and demographic grant aimed at increasing fertility. Lack of conceptual clarity is manifesting itself in inconsistent interpretation of targeting both across the regions and across the programs, each differing in who is considered poor or vulnerable, and how this is established and verified. Hence, equally poor persons are entitled to very different support depending where she or he lives. Lack of clear objectives and specific targets and weak accountability Official development strategies do not set targets for social assistance in terms of poverty reduction and alleviation, or impacts on education or access to health care, or impact on child welfare - a typical array of impacts countries monitor. Particularly conspicuous is the lack of detailed action plan to use social assistance to reduce poverty – a feasible goal, as argued in this report. The lack of clarity with regards to the programs’ objectives, makes grouping the programs and defining the targets and indicators to measure the achievement of results difficult. Thus, results indicators and performance criteria that can be used to assess how well the system works do not exists. The main administrative indicators of performance are two: the amount of money spent on various programs (input) and the number of beneficiaries (output). The actual results in terms of impact, for instance the change in income, or labor market participation, or functioning and participation in case of disability, or child protection are not measured, reported and monitored. In other words, the results chain in social assistance programs in Russia is yet to be introduced and used systematically. The activities under the SP RAS were first ever attempt to assess the performance of social assistance using a set of standardize indicators and this report is a continuation of these efforts.41 Similarly, a periodic in-depth review and evaluation of programs is not practiced in Russia and neither is experimental design embedded in the new programs’ introduction. Multiplicity of laws and regulations Social assistance in Russia is regulated by hundreds of laws and thousands of by-laws, decrees, ministerial orders and instructions and other official acts. This by itself makes management of the system and even “lightâ€? monitoring difficult. At the same time, budgetary classification uses a shorter 41 Some elements of monitoring or evaluating results exist in some programs, but they have never been integrated into a coherent framework that drives budget allocation decisions and prioritization of policies. An example is the “individual social contractâ€? social assistance program. The program development took a long time with piloting, multiple critical reviews and debates on how to measure performance. As a result, the federal law that followed included detailed guidelines on assessing the social contracts’ performance. The methodology on how to assess the individual social contract performance was approved jointly by the Ministry of Labor and Social Protection and Rosstat in 2013 and included a full set of performance indicators (coverage, adequacy, impact on poverty). However, in practice, this methodology has not been applied and the program results are reported as the number of contracts, number beneficiaries and the amount of resources. 36 list of codes, which combine different programs, making even a simple expenditure analysis hardly feasible. Contradictions and inconsistencies abound (e.g. in housing subsidy poverty or need for assistance is defined differently than in the law on poverty benefit), creates a lot of uncertainty for policy makers. What appears to be a legitimate way of action under one body of legislation may be challenged from the positions of another one. This leads to a lot of litigation and court rulings across the country, impeding or reversing reform efforts. A particular complexity is added by the actions of the regional prosecutor office that has the power to challenge and stop/ reverse the changes in the social assistance policies. Anecdotal evidence suggests that this is frequently the case and that the practice on the same issue varies significantly from region to region. Institutional lock-in of inefficiencies: resource allocation in favor of untargeted programs Once introduced, entitlements are hard to modify or remove – the only change the legal system clearly allows is to expand and increase. The composition of spending at the regional level shows a very distinct pattern of spending: the largest spending goes to merit-based programs or special incentives for public sector employees (over 40 percent of consolidated regional spending). Such programs are instituted by regional parliaments and typically reflect pressures from lobbying groups. When faced with cuts in their subsidy from the federal budget, many regions tend to sequester budgets for family benefits or poverty targeted schemes, protecting spending on merits and top ups to salaries. Yet, restructuring and reforming the system proves to be a very difficult process. The experience of the monetization of categorical benefits in kind in 2004-05 illustrates well the power of interest groups which have campaigned to protect their entitlements. As a result, the fear of mass protests against the reform of categorical benefits has led the Federal authorities to introduce the new safeguard provision that de jure prohibits regions to reduce financing for existing social assistance programs (in nominal terms) or worsen conditions for eligibility for any beneficiary.42 This provision dramatically limited regional authorities’ scope for action in reforming the existing programs or reallocating funds toward more efficient use. Different understanding of the meaning of targeting and inconsistencies across methods for means testing Poverty in Russia is understood as unmet needs, and there are four different framework laws in the area of social assistance that define it. Each has its own understanding of the “needâ€? for social assistance: 1) insufficient means/ income relative to some threshold (this is consistent with international definition of poverty); 2) limitations of vital activity caused by poor health; 3) unsatisfied vital needs caused by difficult life situation as a result of a one-time external shock, not dependent on a person; and 4) additional needs due to the difficult life situation caused by the living conditions of a person. In addition to different definition of the “needâ€?, there are also differences in methods to test income/means where means test is applied. Thus, the federal programs for housing, for rental and utilities subsidies, for child allowances, all use different means testing methods. This is inefficient, because the method should be the same; what could be different is a cut-off point if the programs aim to reach different numbers of beneficiaries. Moreover, these programs are administered by different authorities, entailing significant administrative cost – having a single method and a unified registry of applicants would make administration more cost-efficient. To add to this diverse landscape, there are different definitions of "needâ€? in regional programs too. In some cases, targeting is interpreted as a listing of addresses of beneficiaries, regardless of their incomes or needs. In others, it is understood as strict adherence to means testing, i.e. allocating assistance only to those whose incomes are below the poverty line. Most of the regions determine 42 Article 153 of the above-mentioned Federal law “On Social Assistanceâ€? of 22.08.2004 â„– 122-ФЗ. 37 the need for assistance in their own way and often different understanding is used for different groups of programs. The variations include: (i) low value of assets, low earning capacity and low observable (formal) incomes; (ii) poor health status and capacity to earn a living; (iii) difficult situations due to circumstances beyond the control of an individual: loss of property in a disaster, loss of a breadwinner, loss of employment; (iv) life events and family dynamics: birth of a child, having many children, having a child with a disability, etc. Often, additional criteria are added, such as living in remote areas, poor housing conditions, absence of pensions and social insurance benefits. Only a minority of regions use a poverty line to determine an income cut off for eligibility to social assistance. The practice varies here too: some define the threshold as equal to one poverty line, while others define it as equal to 1.5, 2 or even 3 poverty lines. Most regions (43) use the average per capita income in the region as the eligibility cut off. Lack of consistency has generated court disputes with very different rulings, signaling lack of consistency and clarity in the interpretation of the laws governing social assistance. All four means tested programs in Russia – housing allowances, poverty benefits, family benefits, and social services - use different criteria for determining the target population (the population in need of assistance), and who the unit of assistance is (a household or a family). To complicate the matter further, the procedures to test eligibility and verify information all differ, and the systems mostly do not talk to each other. The ultimate result is horizontal and vertical inequity in the access to assistance and in the amounts of assistance. Outdated verification procedures and underutilization of modern technology The rules that are used to verify information needed to test eligibility to award and extend benefits in the means tested programs rely a lot on submission of paper documents that are (almost) never verified or compared to other sources of evidence. Furthermore, the benefits are often awarded for long periods, without any requirement to verify income periodically. An analysis of regional rules for the monthly child allowance showed that in many regions means-testing is carried out only once in 3 years, whereas in some only at the time of the initial application. The electronic cross-check and verification of information is not a common practice, although the number of regions introducing e- government platforms has been growing recently. These shortcomings are well known and were already reported by many previous studies (Ovcharova, World Bank and Urban Institute 2007). Their persistence suggests that there are other, deeper systemic root causes of the failure to improve the system Poor outreach and communication The complex system of social assistance regulated by thousands of legal acts and administered by multiple agencies is not well understood by the population. Vulnerable and socially excluded often do not know what types of programs they are entitled to, or do not consider costs of application worth the small benefits provided. The payments of specific benefits are often aggregated into one transfer which is automatically deposited to beneficiaries’ accounts. There is no breakdown by programs and beneficiaries often do not know how much they receive from different parts of the system, which makes it difficult for them to monitor the accuracy of payment from individual programs. 3.2. Towards an Effective Social Assistance System in Russia To be more effective in reaching the national objective of halving poverty by 2024, the social assistance system in Russia needs to deliver greater support to the poor by improving targeting accuracy (particularly, reducing the error of exclusion, as a large percentage of the poor is excluded from means-tested programs), increasing the amount of assistance the poor receive, and by a wider use of labor market participation conditions for the recipients of support. These changes can be delivered through a gradual shift of resources towards means tested programs (including through inter-budgetary transfers), building links to employment programs, and greater use of technology and interconnectivity of data. Specifically: 38 1. Align social assistance system with higher level objectives: Social assistance system should be aligned with Russia’s higher-level objectives and strategies. This would allow for conceptual clarity, focus, clear definition of targets, monitoring of performance and accountability for results. 2. Gradual approach has better chances of succeeding: Abrupt changes in the institutional design of social assistance are not likely to succeed. A shift towards greater pro-poor orientation can be carried out as fine tuning of existing programs, introducing new features that are supported by the population: simplification and unification of means testing procedures and reliance on co-responsibility of beneficiaries for improving their livelihoods. The efficiency of the system can also be improved by replacing in-kind benefits by monetary forms of assistance. 3. Improving adequacy, particularly for programs associated with poverty profile: Tighter connection between the needs of the poor and social assistance transfer levels requires better information base for determining the assistance, systematic use of evaluations, and expanded use of labor market participation requirements. The nation-wide focus on supporting families with children as part of demographic policies demonstrates that there is a commitment to increase adequacy of existing programs. New means-tested support to families with children introduced in 2018 and amendments to the rules regulating the maternal capital grant,i allowing poor families to use part of their grant for their consumption needs are illustrate this commitment. Still, many children live in poverty, event with assistance, signaling that the amount of assistance has to increase. The system is full of very small fragmented benefits provided to different social categories of recipients across different age groups. Combining them and increasing the amount, would not only improve the poverty impact, but would foster public support. The recipients often receive benefits as a bulk sum that is transferred to their account, without being able to list individual benefits, as shown during the implementation of the pilot surveys. and 4. Use inter-budgetary transfers mechanism to support the social assistance system reform: Better results in terms of poverty alleviation can be achieved by linking the budgeting of social assistance at the regional level and transfers received from the federal budget to the demonstratable improvement of efficacy, consolidation of benefits, replacing in kind inefficient subsidies by cash or near cash instruments, and greater use of means testing. The ASPIRE data base established under the RAS and hosted by NIFI, provides empirical evidence to monitor regional social assistance performance indicators and can be used to monitor the needed improvements as condition for inter-budgetary transfers. 5. Wider and tighter use of labor market participation conditions. The existing activation and graduation programs (in particular the individual “social contractâ€?) should be carefully evaluated. Social assistance in general can use greater reliance on work requirement for able bodied applicants and demand human capital investment focused actions from poor families with children. The former requires that firm links to the employment programs need to be established and properly administered. 6. Regions can significantly improve the administration of social assistance programs, including implementing a requirement that recipients renew application annually or bi-annually, updating income, demographic and other information annually, using information systems to cross-check, verify and update information, modernizing management of information flows and validation using the Unified National Registry architecture. Streamlining the legal framework is being already carried out in Russia with the Regional “Social Codesâ€?, new unified regulations replacing disparate body of law in selected regions (as of 2019 – in six regions). 7. Continue the improvement of targeting by gradually introducing means tests in some categorical measures and further improving the methodology for means testing, including by clarifying the method for calculating household income and by greater use of electronic data bases already containing data on income, taxes, employment status and property. 39 8. Move beyond resources and beneficiaries as key performance indicators: Management and budget decisions should be based on the system for monitoring and evaluating that measures performance (effectiveness and efficiency) of the social assistance system in protecting the poor and vulnerable, based on existing administrative data, and using population surveys. 3.3 What is next for research in social assistance in Russia? For almost three decades, research on the development of the social assistance system in Russia, as discussed in this paper, focused on increasing its efficacy in terms of the poverty impact, as well as on the efficiency of the delivery system. Reorienting programs towards lower income groups, introducing means testing, consolidating programs, improving adequacy of transfers and improving the delivery system including the use of IT technology have been among key policy recommendations. While this agenda will remain relevant in the years to come, new areas are emerging, including: (i) the fit between the poverty profile (working poor) and social assistance system, historically designed on the premise that those working cannot be poor; (ii) improving efficacy of social spending in the poorest regions: what and how?; (iii) importance (and in particular increasing adequacy) of child protection social assistance, including development of childcare services for children 0-36 months, for immediate and future poverty reduction; (iv)“social contractsâ€? as valuable experiments with significant potential to serve as a driver to further reorient social assistance toward working poor and strengthen mutual responsibility aspects of social assistance; and (v) labor market policies and program including better understanding of the minimum wage policies implementation/ enforcement in the regions. 40 Bibliography Alderman, Harold 1993. Multi-Tier Targeting of Social Assistance: The Role of Intergovernmental Transfers. The World Bank economic review, vol. 15, no. 1 33-53. Alderman, Harold and Ruslan Yemtsov (2013) How Can Safety Nets Contribute to Economic Growth? World Bank Economic Review June 2013 and as World Bank Policy Research Paper 6437, 2013. Alexandrova A.L., Kovalenko E.A., Kuznetsova P.O. Reforming of In-Kind Benefits at Regional Level: High Price of Modest Achievements. Analytical Report. Ðœ.: Urban Institute Fund, 2005. Asad Alam, Mamta Murthi, Ruslan Yemtsov, Edmundo Murrugarra, Nora Dudwick, Ellen Hamilton, and Erwin Tiongson. Growth, poverty, and inequality. Eastern Europe and the Former Soviet Union // The World Bank, 2005. ASPIRE: The Atlas of Social Protection (2012-18) / The World Bank: http://data.worldbank.org. Coady, David, Margaret Grosh, John Hoddinott 2004. Targeting of Transfers in Developing Countries: Review of Lessons and Experience // The World Bank, 2004. ESSPROS. Manual, The European System of integrated Social Protection Statistics (ESSPROS) // () // Eurostat. Methodologies & Working papers. European Union, 2011. Fedorov, Lev 2002. Regional Inequality and Regional Polarization in Russia, 1990–99. World Development 30(3): 443–456. Freije-Rodriguez, Samuel, Aleksandra Posarac, and Apurva Sanghi·2019. Can Russia halve poverty by 2024? Brooking Institute blog , May 9, 2019. https://www.brookings.edu/blog/future- development/2019/05/09/can-russia-halve-poverty-by-2024/ Gimpelson, Vladimir and Rostislav Kapeliushnikov 2013. Labor Market Adjustment: is Russia Different? The Oxford Handbook of the Russian Economy, 2013. Grosh, Margaret, Julieta Trias, Ruslan Yemtsov, and Ian Walker 2018. The Contribution of Social Protection and Jobs to Human Capital Formation. Social Protection and Jobs Global Practice of the World Bank. Mimeo. Guriev, S. and E. Vakulenko. (2015). Breaking out of poverty traps: Internal migration and interregional convergence in Russia. Journal of Comparative Economics, 2015. Klugman, Jenny (ed) 1997. Poverty in Russia: Public Policy and Private Responses. EDI Development Studies, World Bank Publications. Kuzminov, Yaroslav I. and Vladimir A. Mau, eds. 2013. Strategiia-2020: Novaia Model' Rosta - Novaia Sotsial'naia Politika [Strategy 2020: New Model of Growth - New Social Policy]. Moscow: Izdatel'skiÄ­ dom "Delo" (In Russian). Kuzminov J, L. Ovcharova and L. Jakobson (editors) 2018. “How to increase human capital and its contribution to economic and social developmentâ€? Report to the Annual Conference of the HSE. Higher School of Economic and Center for Strategic Studies. Lehmann, Hartmut and Anzelika Zaiceva 2013. Informal Employment in Russia: Definitions, Incidence, Determinants and Labour Market Segmentation, OECD Economics Department Working Papers No. 1098. Levin, Victoria 2015. Promoting Active Ageing in Russia: Working Longer and More Productivelyâ€? Washington, DC: The World Bank. Lokshin, Michael and Martin Ravallion 2000. Welfare impacts of the 1998 financial crisis in Russia and the response of the public safety net. Economics of Transition 8(2): 269-295 41 Lokshin, Michael and Ruslan Yemtsov 2013. Poverty and Inequality in Russia, Chapter 32 of The Oxford Handbook of the Russian Economy. Edited by Michael Alexeev and Shlomo Weber. Oxford Handbooks. Oxford and New York: Oxford University Press, 2013. Lopez-Calva, L.F., N. Lustig, M. Matytsin and D. Popova 2015. Who Benefits from Fiscal Redistribution in Russia? Background paper. Lukiyanova. Anna 2015. Earnings inequality and informal employment in Russia. Economics of Transition, 23(2), 2015. Lukiyanova, Anna, and Vladimir Gimpelson 2015. Inequality of labor market incomes, 2015. Maleva, Tatiana, Elena Grishina and Elena Tsatsura 2019. Social Policies in the Long Run: multidimensional poverty and efficient targeting. Russian Academy of National Economy and Public Administration. Ðœoscow. Matytsin, M. L. and Moorty, K. Richter 2015. From demographic dividend to demographic burden? Regional trends of population ageing in Russia. WB Policy Research working paper. Nazarov, Vladimir and Aleksandra Posarac (eds) 2017. Strengthening Social Protection System in the Russian Federation, World Bank and NIFI joint report. M., 2017 OECD 2007 Babies and bosses. Reconciling work and family life. A synthesis of findings for OECD countries. OECD: 2007. Ovcharova, L.N. (editor), Gorina E.A, Voron E.A., Smirnov S.N., Shepeleva E.V. Raising the Efficiency and Effectiveness of social protection based on development of targeted social assistance programs for the poor, 2016. Prokofieva L., Korchagina I., Mironova A., Tarnovskaya E. 2015. Social contract as a mechanism enabling households to overcome poverty in Russia. The Journal of Social Policy Studies, vol. 13, no 1, pp. 81-96 Ravallion, Martin 2016. The Economics of Poverty: History, Measurement and Policy. New York: Oxford University Press. Sobolevskaya, Olga 2015. Social Contracts: No Single Solution to Poverty. September 8, 2015 https://iq.hse.ru/en/news/177664563.html Tesliuc Emil, Pop Lucian, Grosh Margaret, and Yemtsov Ruslan. Income Support for the Poorest: A Review of Experience in Eastern Europe and Central Asia / International Bank for Reconstruction and Development and the World Bank. The World Bank, Washington DC, 2014. ILO, 2014. World Social Protection Report 2014/15: Building economic recovery, inclusive development and social justice // International Labour Office. Geneva: ILO, 2014. World Bank 1993. Russia: Social Protection during Transition and Beyond. Washington DC. World Bank 2005. Russian Federation. Reducing Poverty through Growth and Social Policy Reform // Document of the World Bank. Report No. 28923-RU. February 24, 2005. World Bank 2007. Raising the efficiency of Social assistance Programs and Employment Promotion in Poverty Reduction. Recommendations for Social Protection strategies. Summary Report. Urban Institute (USA); Independent Institute of Social Policy; World Bank. 2007 World Bank 2011 Russia. Public Expenditure Review // Document of the World Bank. Report No. 58836-RU. June 8, 2011. World Bank 2011 Russian Federation. Social expenditure and fiscal federalism in Russia // Document of the World Bank. Report No. 54392-RU. January 25, 2011. 42 World Bank, 2012. 2012-2022 Social Protection and Labor Strategy: Resilience, Equity and Opportunity. The World Bank, Washington DC. World Bank 2017 Russia: Systematic Country Diagnostics. The World Bank, Washington DC. World Bank, 2018. The State of Social Safety Nets. The World Bank, Washington DC. World Bank and NIFI 2019: "Modernization of social assistance in Russia: Survey of households in the Volgograd region. Analytical report "2018," Modernization of social assistance in Russia: Survey of households of the Komi Republic. Analytical report "2018," Modernization of social assistance in Russia: Survey of households in the Leningrad region. Analytical Report» 2019. (in Russian) 43 Annex: Data on social assistance in Russia The significance of resources allocated to social protection, including social assistance, reflects the institutional set up of the system, where the citizens are legally guaranteed public protection against a variety of social and economic risks that may cause their incomes to fall below “socially accepted minimumâ€?. It is not a surprise then that the ILO report “World Social Protection Report 2014/15â€? classified Russia as a country with comprehensive scope of legal coverage43. Similar comprehensive systems ILO finds in all 49 countries of Europe (excluding Turkey), and all OECD countries (excluding South Korea and Mexico). In addition to publicly mandated social protection, Russia features a complex system of social benefits provided by employers44. Despite this institutional and fiscal significance, the social assistance system in Russia is rarely analyzed in detail. Available studies are mostly limited to descriptions, reporting more sizable programs, number of beneficiaries and spending in nominal amounts. In part, this is caused by the lack of data and inconsistency of statistical indicators, where they are reported. Rosstat, the Treasury and Ministry of Finance differ in their reporting on social protection spending, and their classifications do not match international standards. For instance, social assistance and social insurance are blended together in Ministry of Finance and Treasury data in a concept of “social securityâ€?. In other countries, social security typically means contributory forms of social protection, where eligibility and amount of assistance are largely determined by the past contributions of a claimant. In difference to other countries, where most of social assistance benefits are extended in cash, Russia’s social assistance system still features a large number of benefits in kind. As noted above, these in- kind benefits (“privilegesâ€?, or in Russian l’goty) comprise a wide array of free or subsidized goods and services (rent payment, housing utilities, house appliances, drugs, transportation, cultural and sports events tickets, early childhood care and education services, rehabilitation and assistive devices, spa treatments, transportation, etc.). While there is almost no category of the recipients of social assistance programs that is not entitled to several of these l’goty, citizens with merits constitute their most significant recipient group. The system to reimburse the providers is extremely complicated, making reporting and accounting difficult and it likely one of the areas where significant efficiency gains are possible. In an effort to simplify and streamline the system, some of the l’goty were monetized in the mid-2000s and replaced by cash compensations called Unified Monetary Payments, amounting to about 0.9 percent of GDP. Remaining privileges in kind (there are about 600 various categories of recipients and types of such measures) account for about 1.0 percent of GDP. Looking back, the attempt to monetize the system of privileges proved to be politically controversial and was only partially implemented. More than a decade later, the Russian press still portrays the reform as a failure and laments the benefits that were monetized. Table 1 also shows a significant number of social assistance programs that either top up contributory pensions that fall short of reaching region specific poverty line for pensioners, or to provide income in the form of disability, old-age or survivorship social pension to individuals who do not qualify for such benefits under the contributory pension scheme. The social pensions essentially compensate for the gap that exists in the pension insurance coverage. While this gap has been an exception prior to the collapse of the Soviet Union, the transition to the market economy model has also brought 43 World Social Protection Report 2014/15: Building economic recovery, inclusive development and social justice / International Labour Office. Geneva: ILO, 2014. P. 191. 44 Employers provide “additionalâ€? social protection benefits to their employees. These are typically large corporations, and the benefits vary tremendously (anecdotal evidence suggests a range from $10 to $2,500 a year; with richer corporations providing a more generous package of corporate benefits). The evidence on corporate spending is scant. For the purpose of this report, we have estimated that the employers in Russia spend on social protection benefits for their employees about 0.2percent of GDP. 44 uncertainties in employment, resulting in shortened and less dense working and contribution histories. and Russia does not have contributory unemployment insurance scheme any longer. Unemployment assistance in cash is entirely financed by the state budget45. Active labor market programs such as job search services and (re)training were devolved to regions in 2005 and are not reported as a separate spending category. Hence their absence from Table 1. Programs to prevent/ mitigate the risk of Income loss form only a part of the social assistance realm in Russia. Its other significant part pertains to benefits that are conceptually hard to classify as social assistance, including a salary top ups to specific categories of public sector employees and benefits for citizens with merits before the state (e.g. war heroes and veterans, veterans of labor). Furthermore, to address the fertility crises, a number of programs whose objective is to stimulate fertility has been added to the social assistance menu. (i) Salary top ups to some categories of the public-sector employees through social assistance have a long history in Russia. They stem from the Soviet Union times and their objective is to award/ incentivize “socially significantâ€? occupations such as teachers and health care workers in rural areas, agricultural extension workers, military personnel, customs officials, law enforcement personnel). In essence, they are a hardship compensation for public sector workers’ deployment in harsh climate or hazardous working conditions and comprise both an allowance in cash and benefits in kind. Over time, the coverage has expanded not only to include more occupations, but also family members of the covered employees. While hardship allowance is a standard remuneration practice, it is not clear why it is counted and reported as social assistance. Two consequences are clear immediately: the salary portion paid as social assistance is not subject to social security contributions and income taxes and it is not included in the wage base for pension calculations. (ii) Individuals with merits constitute another group of “atypicalâ€? recipients of social assistance. They include war heroes and veterans, veterans of labor46, sportsmen and artists. While some of the benefits are paid in cash, most are in kind and include a range of free or subsidized goods and services (from transportation, to rehabilitation services and spas, to electricity and communal services, to drugs, to home appliances, etc.). None is means tested and they are mandated both by the federal and regional provisions. Because of the way how statistics on social assistance is reported in Russia, it is difficult to precisely estimate public spending on merit-based benefits. In this report, we have estimated it at about 0.9 percent of GDP. Each country in the world has its own system of merits and merit-based rewards, reflecting its history, political and cultural context and traditions. While merit-based benefits are a public transfer to their recipients, it is difficult to argue that they are part of social assistance to the extent that social assistance is understood as assistance to economically and socially less fortunate members of the society. (iii) Since 2007, stimulating fertility became a new priority area for social assistance in Russia. Between 1987 and 1999, the total fertility in Russia dropped by almost half – from 2.22 to 1.16. Combined with increased mortality rate: from 10.5 per 1.000 population in 1987 to 16.4 in 2003, it resulted in the negative natural increase of the population, causing alarm across Russia. In 2007, the new ‘Concept of demographic policies in Russiaâ€? was adopted, introducing a range of programs whose aim is to incentivize fertility. Its flagship program is “maternity capitalâ€? paid to 45 Russia used to have unemployment insurance fund (abolished in 2005), and the unemployment benefits are still provided according to the Law 1032-1 from 1991, but they are no longer associated with contributions paid. 46 A Veteran of Labor designation originates from the Soviet times and is awarded to individuals upon reaching a certain length of work tenure (e.g. 15 or 20 years). In difference to other countries where the length of tenure may be recognized by an employer, in Russia, this status is recognized by the state. 45 mothers for the birth of their third and subsequent children (subsequently – starting with the second). The grant can only be withdrawn in annual installments and spent only on certain household expenditure items such as education tuition, a house purchase or renovation and similar. The grant amount is significant. Initially, it was set up at 250,000 Rubles (about US$ 9,000) and it has been adjusted for inflation over time. It currently stands at 453,000 Rubles (about US$ 8,000). Maternity capital has become the largest social assistance program in Russia, absorbing 0.4 percent of GDP in current spending, and with several percentages of GDP as fiscal liability for future payments. Most regions have followed the introduction of the Federal maternal capital grant by introducing their own version of the program (ranging from $750 to $5,500). Annex Table 1: Russia, social assistance programs by key income loss risks Income risk Social assistance measures Long term illness - income support (federally mandated benefits for long term illness); - additional income support in case of poor health; - in-kind benefits. Long term - income support in case of work accidents (for individuals not eligible for respective social disability insurance benefits under the Social Insurance Fund of the Russian Federation); - medical care and rehabilitation services (for individuals not eligible for respective social insurance benefits); - in-kind benefits. Short term - income support in case of temporary disability (for individuals not eligible for respective social disability insurance benefits); -other programs related to short-term disability. Disability/victims - disability (social) pension (for individuals not eligible for contributory disability pension from the of natural and Pension Fund of the Russian Federation); man-made - additional paid leave compensation for care takers of a disabled child (for those who are not disasters eligible for this benefit under the social insurance provisions); - special one time or periodic benefits for victims of natural disasters/ catastrophes; - tax breaks; - in-kind benefits. Maternity and - maternity benefit for pre-natal services registration (for women not eligible for this benefit under childhood the social insurance scheme); - birth grant (for those not eligible for this benefit under the social insurance scheme); - allowance for early childhood (up to 1.5 years) (for mothers not eligible for this benefit under the social insurance scheme); - foster family allowance; - means-tested allowance for families with children; - maternity capital; - allowance for orphans; - tax breaks, etc. 46 Income risk Social assistance measures Loss of work - unemployment assistance47 Insufficient - poverty benefit; income from work - means tested housing allowance; - allowances for special conditions of work (high risk occupations, deployment in harsh climate conditions; rural public sector employees, e.g. teachers and health workers); - other benefits for selected professional groups. Insufficient - pensions top up (up to the minimum region-specific poverty line for pensioners); income in old age - old-age (social) pension (for the elderly who do not qualify for an old-age pension from the Pension Fund of the Russian federation); - additional benefits for those with long work careers (veterans of labor); - means-tested housing allowance; - special benefits for victims of political repression; - additional benefits for war veterans and victims of political repression; - other forms of in-kind benefits, etc. Loss of - survivors’ (social) pension (for survivors who do not qualify for a survivors’ pension from the breadwinner Pension Fund of the Russian Federation); - allowances for families of military personnel; - other allowances in case of a loss of a breadwinner. Source: Authors, OECD classification; excluding contributory pension insurance benefits administered by the Pension Fund of the Russian Federation and contributory social insurance benefits administered by the Social Insurance Fund of the Russian federation. Many social assistance benefits are geared towards health-related income loss risks or provide some form of compensation for damages to health or income earning capacity inflicted by natural or man- made disasters (the federal inventory of such measures counts 450 different programs48). These are applicable to a specific incident, and each time are determined differently depending on the fiscal capacity and nature of incident. Some regions have specific schemes that can be classified as migration or IDPs assistance, but there is no national program to support these vulnerable groups. To assess performance of the social assistance system in Russia, we use the most recent population survey data collected by the Russian statistical agency, Rosstat. Since social assistance is provided to families with different needs and conditions, and multiple benefits are often received by members of the same family, it is important to see who receives what at the aggregate level. Russia does not have a unified registry of beneficiaries of social protection programs, and the only way to see the social assistance programs’ aggregate impacts on the population is to use data from representative household surveys. Since 2014, Rosstat carries out annually The Survey of Income and Social Program Participation. This is a large representative survey whose data is in the public domain and this section is based on this source, using both available rounds (e.g. 2015 and 2016). 47 Contributory unemployment insurance was abolished in Russia in 2005. Non-contributory allowance in the case of unemployment is the only form of income support related to unemployment currently in place in Russia. 48 Several nuclear accidents have spurred a series of legislative acts where state took on itself considerable liability of life long support to the affected populations. 47 Data for international comparisons come from the Atlas of Social Protection Indicators of Resilience and Equity - ASPIRE, a World Bank database of social protection performance indicators49. ASPIRE includes the most recent data from 122 countries and presents a number of effectiveness and efficiency indicators. The data are analyzed at the level of the entire system, by social protection area, and by individual programs. The main indicators are (a) measures of coverage (who does and who does not receive benefits); (b) adequacy (the relative size of social protection transfers compared to household income or consumption for a group); (c) distribution of the transfers amounts (benefits), and recipients (targeting or incidence); (d) the poverty and inequality effect; and (e) measures of efficiency such as targeting performance and the share of benefits that are reducing poverty (benefit- cost ratio). Russia is part of the global database and its local replica with enhanced statistics for Russia is maintained at the NIFI web site.50 ROSSTAT, the Statistical Agency of the Federation, monitors poverty based on the annual surveys of populations. It has established itself as a credible source for analyzing poverty and inequality nationally and is the only such source for monitoring national and regional poverty profiles. It runs a continuous survey, introduced in 1952 and substantially improved and revised in 1997 and, again, in 2005. The HBS sample consists of 49,000 households drawn from the whole country, with data collected quarterly. Recently, ROSSTAT added several new regionally representative surveys to complement and deepen the characterization of poverty, most importantly survey of incomes and participation in social programs, key source for this report. The survey has been developed over 2000s through a series of pilot tests. The first step in this process was the joint work with the World Bank on NOBUS- a survey conducted in 2002 and 2003 with the new sampling methodology and new questionnaire design to capture participation of population in social programs. The survey was carried out in April and May 2003 and covered a sample of approximately 45,000 households with 118,000 individuals. NOBUS used a three-stage stratified sample design and is representative on the national, and on the regional level for 46 selected regions (GKS 2004). NOBUS gathered information about demographic characteristics of the household members, health, education, their main and secondary job activities, as well as individual sources of income and the amounts of wages, pensions and multiple other social transfers. It was the first to become an open micro data by official statistical agency and was used for many studies and reports. Its key principles remained in the new survey that started in 2013. Survey of incomes and participation in social programs, key source for this report has become the main source for official poverty statistics as of 2018 (Russian abbreviation Ð’Ð?ДÐ?). The survey series has been launched on 2013 and has a sample size of 46,000 households. It is regionally and nationally representative. The survey interviews take place annually in March of each year, with the reference period of the previous year. In 2017 Rosstat expanded the sample dramatically to 150,000 households, making possible accurate analysis of poor and vulnerable groups by regions. Data are published in open access at the Rosstat web site (). This data is the primary source of performance indicators for this report. The Table 2 below shows the comparison of administrative data and survey- based estimates of main types of programs in Russia. 49 www.worldbank.org/aspire 50 www.nifi.ru 48 Annex Table 2. Comparing budget data with survey-based estimates for Russia. Administrative data Household Survey Programs USD PPP 2011 % GDP USD PPP 2011 % GDP Unconditional allowances and other social payments for people entitled to 42,202,603,824 1.22 33,588,756,864 1.0 receive social support Child allowances 7,812,588,431 0.23 3,129,270,615 0.1 Poverty targeted cash transfers 1,612,969,255 0.05 735,618,074 0.0 Benefit for children who lost one 2,365,854,706 0.07 2,320,011,461 0.1 parent Maternal capital 13,848,847,359 0.40 23,639,050,125 0.7 Social pensions including disability, 7,945,456,430 0.23 5,731,926,259 0.2 Survey of survivorship income s and Free use of the milk kitchen 224,219,625 0.0 participation in social Housing subsidy 5,714,431,336 0.16 6,266,875,399 0.2 programs Scholarships 3,244,741,369 0.09 2,038,391,633 0.1 2016 Food and transportation allowances 93,633,032 0.0 Food and transportation 6,954,628,852 0.20 2,958,096,867 0.1 privileges/discounts Transfers for care takers of people in 2,463,910,583 0.07 1,051,567,543 0.0 need of assistance Other cash transfers for government 611,412,240 0.0 organizations Unemployment benefit 1,276,056,942 0.04 863,246,748 0.0 Total 95,442,089,088 2.75 83,252,076,484 2.4 i Starting in 2007 a large demographic grant to stimulate fertility has been provided to families starting with 3rd child. It is provided as a deposit (“maternity capitalâ€?), which can be used only for specific purposes and gradually on spending related to human capital investment. 49