Report No: ACS8094 Republic of Peru PE Poverty & Social Inclusion NLTA II ACHIEVEMENTS SUMMARY FOR FISCAL YEAR 2014 March 2014 LCSPP LATIN AMERICA AND CARIBBEAN Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750- 4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. 2 PERU POVERTY AND SOCIAL INCLUSION II (P146773) NON-LENDING TECHNICAL ASSISTANCE ACHIEVEMENTS SUMMARY FOR FISCAL YEAR 2014 1. Task summary This Non-Lending Technical Assistance (NLTA) is responding to requests by the National Statistical Institute of Peru (INEI) and the Ministry of Social Inclusion (MIDIS) to provide technical assistance to increase the government’s capacity to design, monitor and evaluate social policies aiming at reducing poverty and improving social inclusion. The activities during this fiscal year (FY14) were organized around two broad goals: (i) validating and improving poverty measurement conducted by INEI; (ii) strengthening evidence-based decision making in MIDIS. 1st component: Validating and improving poverty measurement conducted by INEI This work builds on several years of close collaboration with INEI and the Peruvian Poverty Committee. The objective of this technical support is to strengthen INEI’s capacity to produce high qua lity information for policy making in an open and transparent way. During FY14, responding to a request by MIDIS and INEI, the NLTA has been providing technical assistance and advice to INEI about poverty mapping. The poverty maps are estimated using household survey and census data that can produce poverty rates at high levels of disaggregation. The information from the poverty maps is used by some government agencies, including the Ministry of Finance and MIDIS, to guide resource allocations to poorer districts. Thus, poverty maps are extremely relevant from an operational as well as a political economy aspect. It is of no surprise that there is high interest to understand and validate the assumptions underlying the estimation of the maps, the quality of the current estimations, as well as providing advice on how often to update the series. The activities done this year ranged from completing the technical validation of the existing poverty maps, present a proposal to the Poverty Committee about the best way to estimate the maps, and opening a broader discussion about targeting in Peru. This activity has increased INEI’s capacity to independently estimate poverty maps in the future. In addition, this work has been contributing to MIDIS’s effort to develop a new targeting algorithm for its social programs. Overall, this NLTA has facilitated a more informed and deeper exchange between the Ministry of Finance, MIDIS and INEI, in the context of the Poverty Committee, about the costs and benefits of alternative targeting formulas for different resource allocations and programs in Peru. 2nd component: Strengthening evidence-based decision making in MIDIS This component of the NLTA has the overall goal of improving the performance and impact of MIDIS- led programs and policies though better M&E systems. The NLTA activities are a key part of a larger 3 package of assistance provided by the Bank to MIDIS, which includes the TAL and DPL. The work has been done in close coordination with the HD team that is leading the operations. The NLTA’s main counterpart is the General Directorate for Monitoring and Evaluation (DGSE) which is in charge of the M&E activities in MIDIS. During FY14, the NLTA has supported a series of activities with the more concrete objective of promoting the effective use of M&E information in the policy cycle. As part of this work the team has contributed to: (i) the production of a mobile app to facilitate the use and increased the transparency of MIDIS information; (ii) develop an logic framework for the M&E activities in MIDIS focusing on bridging the gap between the producers (M&E team) and the users (the social programs) of the evidence; (iii) producing analytical products on poverty and social inclusion to serve as an instrument to motivate and influence policies in MIDIS and in Peru. The work has helped organizing the way the DGSE operates and relates to the vice-ministry of social programs. In addition, the activities have contributed to consolidate the DGSE within MIDIS, thus contributing to the sustainability and functionality of MIDIS M&E system. 2. FY14 achievements 2.1 Validating and improving poverty measurement conducted by INEI 2.1.1 The request and context Since 2007, the World Bank team has been actively leading and supporting the work of the Peruvian National External Advisory Committee (AC) whose role has been to: (i) discuss and recommend methodologies to improve poverty measurement; (ii) support the work of an inter-institutional technical team that implements the recommendations (composed by INEI, World Bank and others); and (iii) build sustainability and country ownership of the poverty measurement methods and results.1 The AC model has been very successful in providing the space for dialogue and consensus building, just in-time technical advice to INEI, and a new concept for quality control. Under the work of this committee, many technical issues related to poverty measurement have been solved and Peru has re- started publishing annual poverty estimates using a comparable methodology across time. In FY12, MIDIS requested an assessment of the quality of the poverty maps done by INEI. The poverty maps are estimated using household survey and census data that can produce poverty rates at the 1 The AC is composed by experts on poverty issues from the public sector (Ministry of Economics and Finance and Central Bank), local academia (Universidad Catolica, Universidad del Pacifico, GRADE, Universidad San Martin), the French IRD, the IADB and the World Bank. In addition, to the technical support, a key objective of this inter-institutional approach has been to help INEI create an environment of openness and transparency in order to help to improve its credibility and relationship with clients. 4 district level. As mentioned above, there is high interest to understand and validate the assumptions underlying the estimation of the maps, the quality of the current estimations, as well as providing advice on how often to update the series. Based on that, at the end of FY12 the World Bank team proposed and integrated the poverty map work under the AC activities to ensure the highest level of quality and transparency. During FY13, a technical team from INEI, local academia, the IADB, and the World Bank reviewed the estimation process of the poverty maps and performed a series of validations that could inform about: (i) the best way to apply the methodology in the case of Peru; and (ii) the methodological soundness of updating the poverty map using the same population census from 2007. This work was documented in a methodological note completed in FY13. 2.1.2 Achievements during FY14 Methodological evaluation of the existing poverty maps was presented and discussed in the Peruvian Advisory Committee. The technical team held several meetings to discuss the results of the validations and presented the findings to the AC members. In December 2013, the team presented a summary of the results to the AC and the main lessons learned for the future estimation of poverty maps (See Box 1 for summary of lessons and Annex 1 with the main power-point presentations in Spanish). The AC agreed that the next step would be to start the work on a new poverty map implementing the methodological suggestions. The team did a technical assessment of the viability of updating the poverty map using SISFOH. One of the suggestions of the AC was to update the poverty map using data collected by the Household Targeting System (SISFOH). Since the objectives and field protocols of the SISFOH are different from a National Census, it was unclear whether this idea would work. Thus, the Bank team responded rapidly to this concern by estimating a preliminary version of a poverty map for 3 departments using data from the ENAHO 2012 and the SISFOH. The analysis indicated that it was feasible to estimate a new map using SISFOH and that the suggestion to move in this direction was worthwhile. (See Annex 2 for a power-point presentation). The work on poverty maps has been extremely relevant for a broader discussion on targeting of social programs. The information from poverty maps produced by INEI is currently being used by the Ministry of Finance for the allocation of the Oil Canon in some departments of Peru. Additionally, the Ministry of Finance uses the poverty maps as one of the criteria to make some specific budget modifications to favor the poorest local governments (D.S. Nº009-2012-EF). Moreover, MIDIS uses the poverty maps to target some specific social programs and as one criteria to identify their main target population (the population in process of inclusion). In the past few months, MIDIS has been evaluating a new algorithm to target its social programs and, after the presentation of the work on poverty maps, it is considering adding an updated poverty map for geographical targeting. Overall, the exchanges between the AC and MIDIS have improved understanding the data requirements, assumptions, and limitations of different targeting formulas, including poverty maps. The team has been helping to connect the discussions about poverty maps within the AC and MIDIS’ new algorithm. 5 Box 1. Main technical lessons from the validation of the poverty maps estimated by INEI The poverty maps estimated in Peru for 2007 and 2009 follow the methodology developed by Elbers, Lanjouw and Lanjouw (Econometrica, 2003).2 The poverty map methodology estimates a consumption model using household survey data and then uses that model to impute consumption in the census data. This allows estimating measures of poverty and inequality at low geographic levels using all the observations from the census. In Peru, the poverty maps were estimated using the 2007 National Census and the 2007 and 2009 National Household Surveys (ENAHO). During FY13, a team from INEI, local academia, the IADB and the World Bank conducted a series of analytical pieces to evaluate the quality of the poverty maps estimated in 2007 and 2009. Next, we summarize the main technical lessons learnt from the work. Continuing to update the map using the 2007 census may not the best strategy. Given the large development changes observed in Peru in the last years, the 2007 Census information is becoming less comparable with more recent information collected by the National Survey (ENAHO). Updating the poverty map using data from SISFOH which were collected more recently, may prove a better strategy. The next poverty map should group small districts in population to estimate more reliable and stable district level poverty rates. The experience estimating poverty maps recommends presenting estimations for geographic areas with at least 1,000 households since the error of the estimation can be large for smaller areas. In the case of Peru the map is being estimated at the district level, however, approximately half of the districts have fewer than 1,000 households. This fact may also contribute to reducing the stability of the poverty rates estimated across time. The poverty maps should include a cluster level error in the consumption models. The estimation of the 2009 poverty map did not model a cluster level effect for 16 out of 25 consumption models. The validation showed that excluding the cluster level effect was underestimating the standard errors of the poverty rates for some areas. Department level modelling seems appropriate. The current poverty maps estimate models at the department level. One of the validations compared the results of estimating consumption models for larger geographic regions but separating by urban and rural areas. The results indicate that the department level models perform better. 2Elbers, C., J. Lanjouw, and P. Lanjouw (2003), “Micro-Level Estimation of Poverty and Inequality,” Econometrica 71:1, 355–364 6 2.1.3 Next steps and additional requests The Bank team will continue supporting the poverty map work with INEI in the context of the AC. During the next months, INEI is planning to conduct a survey to government agencies on the uses of the poverty maps or similar instruments for targeting resources and/or guiding policies. The Bank team will help analyze those data and disseminate the findings. This information will contribute to the debate about the value of targeting and potential tools to do this in Peru. The team is planning to write a note with the lessons learnt from Peru to share this experience with other countries and within the Bank. In addition, INEI has requested support to standardize existing microdata documentation systems to improve the usability of the databases and information collected. Second, the team was asked to strengthen INEI’s current Application Programming Interface (API) platform to improve accessibility and user experience. This technology, similar to the one the World Bank uses under its Open Data Access initiative, allows full flexibility by the user on how data and information are accessed and visualized. The feasibility to do so using an easy, unique, interactive system in real time could introduce large efficiency gains on how information is processed and flows from INEI to the rest of the GoP and external users. Peru is one of few developing countries that are pursuing this agenda, and as such, this work will also provide key lessons on how to adopt this technology in similar settings elsewhere. Finally, the team will also provide technical guidance to INEI in its process of revising and improving the coverage, quality and timeliness of its vital statistics. Moreover, the team will help conduct pilots and analyses related to testing new household survey modules to measure food consumed outside of the home. These activities will also aim to strengthen INEI’s statistical capacity and quality of poverty measurement. A Trust Fund for statistical capacity has been secured to support part of this work since FY12. 7 2.2 Operationalizing evidence-based decision making in MIDIS This pillar of the NLTA is part of a comprehensive package of assistance provided to MIDIS by the World Bank to support their information and monitoring and evaluation systems. During FY14, responding to specific requests from MIDIS, the NLTA provided support in several areas ranging from strengthening capacity on monitoring and evaluation (M&E) to analytical products to inform policy. 2.2.1 Producing a logic framework to evaluate MIDIS M&E activities One common mistake of the M&E systems is to assume that the systems have an intrinsic value. In reality, the information generated by the system is only valuable if it is broadly used. The typical institutional arrangements for evaluation observed in LAC consist on generating evidence about the impact of policies and programs, and share that evidence with the policy makers responsible for the implementation of the policies and programs. Under this arrangement, the efforts of the evaluation area are concentrated in generating evidence and putting that evidence on the hands of the decision- makers. However, in many cases, the operative units do not have the technical capacity, the human resources, the time, or the incentives to use the evidence in a concrete and practical way. Thus, the results from the evaluations are not used as they should to improve the policies and programs. The General Directorate for Monitoring and Evaluation (DGSE) of MIDIS has been trying to incorporate a series of additional elements to improve the traditional approach. During FY14, the NTLA has supported the DGSE to develop its logic framework. Under this new framework, the operations of the DGSE have three goals: (i) producing evidence and monitoring performance of programs; (ii) management of the evidence and early warning systems about performance; and (iii) use of the evidence and information produced by the M&E system. The first goal corresponds to the traditional approach observed in many countries. The innovation is adding the management and use of the evidence and part of the activities of the DGSE. Importantly, the DGSE will be evaluated based of the achievement of the three goals. This creates incentives for the unit to work with the programs and policy-makers to develop an improvement plan that is clear, concrete, and valuable for increasing the final impact of the programs. In addition, the DGSE will develop mechanisms to measure the impact of the improvement plans that feedback the policy cycle. During FY14, the logic framework to evaluate the DGSE M&E activities was completed. The Bank team contributed to this activity by providing guidance in the development of the logical framework with clear monitoring indicators. In addition, the Bank collaborated to develop the strategy for the M&E of the indicators, defining roles and responsibilities for M&E and the sources of information (Annex 3 includes the logic framework prepared by MIDIS with the support of the Bank). 2.2.2 Developing strategies for the use of the evidence produced by M&E systems The DGSE new logic framework highlights the efforts of MIDIS to address the typical missing link in the policy cycle of feeding M&E back into policy and program planning and decision-making. MIDIS seeks to bridge this divide through the “management of evidence” (gestión de evidencia), where the 8 information produced through their M&E system (particularly with regards to evaluations) are prepared in more digestible forms for program and policy decision-makers and the DGSE accompanies the process to support their implementation. More specifically, a special sub-unit on management of evidence in the DGSE intends to take the recommendations identified through the evaluations and translate them into more feasible action items given the political, operational and financial contexts of the programs. This would form the basis for an improvement plan defined jointly between the programs and the DGSE, which would also identify necessary financial and technical resources to support these improvements. During FY14, the team started to work on the first pilot of the management of the evidence. The team has been helping MIDIS to develop the standards and protocols for the management of evidence. In particular, MIDIS intends to use planned evaluations to serve as a pilot for the application of the management of the evidence. The objective of the pilots is to explore ways to operationalize this activity and develop some guidelines on how to work with the programs. During FY14, the team worked with MIDIS in preparing a pilot for Pension 65 using information from a recently produced qualitative evaluation of the program. The team has provided guidance through the process to review the results of the qualitative evaluation and identify action items for the improvement plans. The team is currently working on preparing strategies to accompany the implementation and oversight of the improvement plans, and measure the impact of the recommendations (Annex 4 includes material prepared as part of this activity). During FY14, the Bank team organized a workshop with the Behavioral Insights Team from U.K. to inform the management of evidence. Incorporating behavioral insights into policy can be extremely valuable in some cases where psychological biases cause people to make choices that seem contrary to their best interests. For example, psychological factors, such as high levels of stress or lack of self- efficacy and aspirations create barriers to taking full advantage of a wide range of social programs, designed under the assumption that beneficiaries act rationally in the context of complete information. Recent experimental evidence from Nicaragua suggests that boosting beneficiaries’ aspirations may double the impacts of a productive grant on household incomes (Macours and Vakis, 2009).3 This idea has gained significant traction in developed countries such as the U.K. where the Prime Minister Office has a specially created a unit to work on this, the Behavioral Insights Team (BIT). That team has found positive impacts of incorporating behavioral nudges into ongoing policies in the areas of taxes, health, energy, etc. The DSGE from MIDIS expressed interest in exploring these ideas as part of the management of evidence efforts. The DGSE believes that within some of MIDIS programs, such as the CCT Juntos or Pension 65, behavioral biases could be important. During FY14, the Bank team invited the senior policy advisor of the BIT from U.K. to a one-day workshop with MIDIS in Peru. This allowed MIDIS to understand BIT’s work in the U.K. and discuss the relevance and challenges of pursuing something similar in Peru. 3 Macours, Karen and Vakis, Renos, 2009. "Changing households' investments and aspirations through social interactions: evidence from a randomized transfer program," Policy Research Working Paper Series 5137, The World Bank. 9 During FY14, the Bank team facilitated the exchange ideas on the management of evidence. The team from MIDIS was invited to co-host the IX Conference of the Monitoring & Evaluation Network of Latin America and the Caribbean, which took place in Lima, Peru from December 2-4, 2013.4 MIDIS, with the support of the Bank team, organized the second day of the conference centered on the Use of Evidence. During that day invited speakers (from the U.S. Government Accountability Office, JPAL-LAC, the U.K. BIT, and renowned M&E expert Michael Bamberger) presented the progress and challenges in this area. In addition, MIDIS shared their experience so far in institutionalizing the management of evidence with other government officials, practitioners, and academics from Peru and many countries in LAC. As a result of MIDIS’ presentation in the conference, the World Bank Series Nuts & Bolts of M&E Systems5 invited MIDIS to write a piece on this experience. The note will be completed in a couple of months. 2.2.3 Offering tools to increase the use and transparency of information Reliable and quality information is fundamental to the success of any M&E system. MIDIS faces unique information challenges as it intends to provide M&E of policies and programs. This will involve the strengthening and integration of existing databases and the technological solutions for their communication and interfaces for different users, including policy makers, M&E technocrats, and program operators. The issues related to information systems of policies and programs are largely addressed through the TAL. As such, the NLTA has focused on areas requiring more immediate attention, in particular, the development of apps for the dissemination of basic information and specific technical requests. During FY14, the beta version of the MIDIS mobile app was completed. The Bank team has provided support for the development of a user-friendly, widely accessible app to readily provide basic information on MIDIS programs, such as budgets, beneficiaries, and socio-economic data produced by INEI. The main users of the app so far are MIDIS officials who want to access to the information while they are in the field or in their daily activities. The expectation is that this app could be expanded with 4 This regional conference has been supported by the World Bank (WB) and the Latin America and Caribbean Community of Practice on Managing for Development Results (CoPLAC-GpRD) of the Inter-American Development Bank (IDB) since 2005, together with a growing group of countries in the Latin American and Caribbean region. The forum promotes the exchange of experiences, and tools and innovative approaches for M&E at the national, sectoral and sub-national levels. The REDLACME is an instrument that facilitates the regional dialogue around M&E issues among interested parties, such as academics, public servants, and practitioners throughout Latin America and the Caribbean and beyond. 5 The objective of the series is to increase knowledge about M&E systems through a regular series of papers focusing on the design, implementation and use of M&E information by governments and civil society based on best practice experiences. 10 the development of MIDIS information systems, particularly with the monitoring dashboards currently under development in the DGSE. The development of the app has helped MIDIS to organize the information to be presented and develop a system to keep the information updated. Importantly, as part of the work, the team brought together staff from INEI and MIDIS to develop the application in a coordinated fashion and exploit synergies. Another successful result of this process is that MIDIS agreed with INEI to start uploading data directly from INEI’s web service instead of loading the data manually. This will dramatically improve data sharing between the two institutions and will make updating of the data far more efficient (Annex 5 includes a presentation describing the app in Spanish). Overall, the app serves as a tool to increase transparency and accessibility of already existing information. The next phase of this work is to develop a more interactive app where beneficiaries could also receive information or provide feedback. In addition, within this work, MIDIS has expressed interest in exploring the possibility in moving towards Open Data. Box 2. The MIDIS mobile app The beta version in Android of the MIDIS mobile app presents information about MIDIS’ programs and INEI’s socio-economic indicators at different geographic levels and across time. The users are able to do specific data queries, and display indicators in tables, charts or maps. The application has been developed so that it can be used without Internet access and receive regular updates when connected to the Internet. The experience and lessons learnt from this process will be shared with other government agencies in an effort to promote the usability of the information collected by INEI and MIDIS and highlight innovative ways to share that information and receive feedback about government programs and policies. The app is available for download in Google Play. 11 2.2.4 Producing analytical and media products on poverty and social inclusion One of MIDIS’ challenges is to ensure that its social inclusion policies are relevant and effective. For that purpose, MIDIS requested support in both producing and sharing evidence on social inclusion issues in Peru. As part of this NLTA, during FY14, the team worked on a series of analytical products on poverty and social inclusion to inform MIDIS’ policy agenda and a media campaign on agency as part of MIDIS’ gender strategy. During FY14, the Bank team produced an updated poverty and shared prosperity note. In the context of the new World Bank’s twin goals and in order for MIDIS to assess the progress and the pending issues on social inclusion, it is important to have a clear understanding of the recent trends in poverty and inequality in Peru. The recently published poverty series by INEI provides an opportunity to do so. The objective of the poverty and shared prosperity note for Peru will be to describe the key trends in poverty and shared prosperity and highlight some of the factors that could be driving the observed patterns. In addition, the note highlights the key challenges ahead for Peru in terms of social inclusion. The ultimate goal of the note is to serve as an instrument to motivate and influence the poverty debate in Peru and help MIDIS prioritize its social inclusion policies. The Bank team has already completed a first draft of the note and plans to finalize the note by May, 2014. During FY14, the team produced a descriptive analysis of the spatial correlates of chronic poverty in Peru. In the last few months, the team has been producing a note describing the factors associated with the persistence of poverty in Peru using poverty maps. This analysis complements the poverty note and will also support MIDIS’ agenda on social inclusion by showing the geographic areas that are still lagging behind, highlighting the factors that may be driving the slow progress, and suggesting potential policies that could help those areas to converge. This work is also part of the Regional Study on Chronic Poverty leaded by LCSPP and will be completed by June, 2014 (Annex 6 includes a preliminary power point presentation). The production of a media campaign on agency as part of MIDIS gender strategy is underway. “Pequenas historias para aprender a ser gigantes” is an information campaign designed jointly with MIDIS to use behavioral economics insights to improve the effectiveness of its social programs, through carefully crafted stories about actual beneficiaries of social programs in Peru. The campaign promotes messages and policy lessons aimed at lifting psychological constraints on taking full advantage of the programs. The campaign is aimed at several audiences: beneficiaries, broader community, and policy makers. The campaign is generally focused on the message that small nudges, within reach of each type of audience, may trigger big changes. For beneficiaries, greater aspirations may help achieve their goals. For the community, greater acceptance of those infected with TB may help lower TB transmission rates through facilitating treatment. For policy makers, small intelligent tweaks in the policy programs may significantly increase their effectiveness. 12 There is substantive evidence that small nudges in development programs may trigger big impacts (see Datta and Mullainathan, 2012).6 A recent study in Peru (Perova and Vakis, 2013)7, based on quantitative and qualitative work with three social programs (CCT program Juntos, financial literacy program Corredor Puno Cusco and socio-economic intervention to reduce TB) is consistent with this evidence: lack of self-efficacy and belief in own ability to learn new skills interferes with take-up of financial literacy programs. Depression is a strong contributing factor to a success of a TB treatment program. Community ostracism imposes a heavy psychological tax on the beneficiaries of the flagship Juntos program. In a way consistent with behavioral economics theory, these psychological constraints may be reducing effectiveness of the social programs. Building on this work, the information campaign illustrates these policy lessons with the stories of concrete beneficiaries. The campaign aims to trigger behavioral changes through inspirational stories of real people, relying on insights from behavioral economics. Importantly, information campaign is developed to complement and work in tandem with current MIDIS programs (e.g. the CCT Juntos), fostering behavioral changes that are likely to increase their effectiveness. During FY14, the strategy and main messages for the media campaign on agency have been finalized. The Bank team has been working on producing a message that communicates in an easy but clear way the stories and the concepts. The Bank and MIDIS teams have been working very closely with the artistic directors of the campaign to make sure they understand the concepts we want to communicate. In FY14, the team completed the campaign strategy and did several field visits with the artistic directors to prepare the work. The production of the scripts is currently underway. The production of the campaign is being funded by LCSPP Gender Mainstreaming work. 2.2.5 Next steps During the next months, the team plans to continue the support MIDIS in the activities described above. A mission at the end of March with the new minister of MIDIS will discuss priorities and new requests in the contexts of the overall package of Bank assistance. In addition, the findings from the poverty note and the analysis on chronic poverty will be disseminated to technical and policy oriented audiences within the Bank and in Peru to local stakeholders (policymakers, government officials, academics, think-tanks, civil society, etc.). Some of the analyses will be also disseminated as part of the Regional Study on Chronic Poverty which plans a study tour, where team members would travel to selected countries in the region to present the study and consult with stakeholders on their views about the questions being asked. 6 Saugato Datta and Sendhil Mullainathan. 2012. "Behavioral Design: A New Approach to Development Policy." CGD Policy Paper 016. Washington DC: Center for Global Development 7 Elizaveta Perova and Renos Vakis, 2013. “Improving Gender and Development Outcomes through Agency: Policy Lessons from three Peruvian Experiences.” Washington DC. 13