The Scaling Community of Practice (CoP) launched an action research initiative on mainstreaming scaling in funder organizations in January 2023. This initiative has three purposes: to inform the CoP members and the wider development community of the current state of support for and operationalization of scaling in a broad range of development funding agencies; to draw lessons for future efforts to mainstream the scaling agenda in the development funding community; and to promote more effective funder support for scaling by stakeholders in developing countries. (For further details about the Mainstreaming Initiative, see the Concept Note on the COP website).
The Mainstreaming Initiative is jointly supported by Agence Française de Développement (AFD) and the Scaling Community of Practice (CoP). The study team consists of Richard Kohl (Lead Consultant and Project Co-Leader), Johannes Linn (Co-Chair of the Scaling CoP and Project Co-Leader), Larry Cooley (Co-Chair of the Scaling CoP), and Ezgi Yilmaz (Junior Consultant). MSI staff provide administrative and communications support, in particular Leah Sly and Gaby Montalvo.
The principal component of this research is a set of case studies of the efforts to mainstream scaling by selected funder organizations. These studies explore the extent and manner in which scaling has been mainstreamed, and the major drivers and obstacles. The case studies also aim to derive lessons to be learned from each donor’s experience, and, where they exist, their plans and/or recommendations for further strengthening the scaling focus.
The present case study focuses on the Inter-American Development Bank’s (IDB) and specifically its sovereign lending window. It was prepared by Koldo Echebarria, former Country Representative and Strategic Planning and Development Effectiveness Manager at IDB, as an in-kind contribution to the Scaling Community of Practice. Johannes Linn, Lawrence Cooley and Richard Kohl provided valuable comments to a first version of this document.
This case study looks at IDB’s institutional standing regarding the mainstreaming of investment operations at scale. Only the IDB sovereign guaranteed window is covered by its purview; it does not extend to IDB Invest or IDB Lab operations. The objective is not to identify programs that have been successful in contributing to delivering results on a scale, but to analyze to what extent institutional incentives are aligned with mainstreaming scaling. In order to achieve this, the case study: (i) first looks at whether a clear mandate for scaling has been established at the institutional, sector, and country strategy levels; (ii) looks whether lending practices adopt scaling as an institutional practice; and (iii) third looks at the institutional factors that influence scaling, providing concrete recommendations for how scaling could be mainstreamed at the IDB. Our analysis leads us to the conclusion that, despite sector and country strategy documents signaling needs, priorities, and policy interventions with a clear potential for scaling, strategy and policy documents do not include a mandate for scaling. At the project level, we see an inconsistent attention to scale. Although need or problem diagnosis and intervention selection are compliant with fundamental scaling requirements, project sizing and scope is based on a financial envelope that is not calculated according to scaling objectives. Not even when projects are included in a long-term credit line do efforts exist to measure up to scale-relevant magnitudes. Projects, however, indicate that the IDB appears to have well-established incentives to pursue cross-national horizontal scaling, drawing extensively on successful interventions in other countries of the region. In order to make scaling feasible and sustainable, the IDB would need leadership and strategic direction at the institutional level, preferably including the commitment to selective outcome targets and extending partnership agreements similar to the one it recently signed with the World Bank. Along with reducing sector fragmentation and adjusting lending instruments and project cycles to a scaling logic, the Bank would also need to work with implementation structures that are aligned with scaling pathways.
Table of Contents
This case study takes a preliminary look at the IDB status in mainstreaming scaling in investment operations. The goal is to determine the degree to which scaling is given priority by IDB strategies and policies and implemented in the allocation of investment lending resources through specific projects. This will make it possible to pinpoint the variables that affect scaling in IDB operations and make preliminary suggestions for progress. The concepts presented in the paper are based on a limited desk review of IDB strategies and a small sample of projects recently approved. To come to stronger recommendations, a more thorough analysis is required, so conclusions should be viewed tentatively.
The paper takes an ambitious definition of scaling. It is not just the expansion or replication of a successful program to reach a greater number of beneficiaries. Scaling is both the goal to reach and sustain a significant impact and the process to get to that point. From a goal perspective, scale requires that two conditions are met: (i) first, that coverage reaches a relevant portion, relative to the size of the need, demand, or issue being addressed; (ii) second, that financial and institutional conditions are in place for a sustainable, lasting impact. Achieving scale entails a long-term program commitment with a sizable output that is supported by sufficient funding and institutional frameworks suitable for the delivery of outputs. From a process perspective, scaling is the succession of steps needed for getting there, by trying out an intervention at a pilot stage, monitoring and evaluating the results and laying the groundwork necessary to expand the intervention to the significant coverage required to reach scale, including the political and fiscal commitment, the policy requirements consistent with scaling, the implementation arrangements for delivery and the monitoring and evaluation mechanisms to confirm impact.
Mainstreaming scaling in donor organizations means the adoption of scaling as a recognizable pattern in development operations, reflected in institutional priorities, allocation of resources, intervention design and implementation and accountability frameworks. Mainstreaming scaling requires political leadership, organizational vision, and a set of explicit and implicit incentives that drive the organization towards achieving significant and lasting impact commensurate to the needs or problems. It goes beyond piloting successful innovative projects, adapting successful interventions to new sectors or projects or even to achieving a high rate of satisfactory results in particular projects or programs. It also goes beyond having showcased successful scaling interventions. These are all necessary but not sufficient conditions for mainstreaming scaling.
Scaling could be considered an implicit requirement for development to take place. To ensure progress, economic, social, and political changes must occur on a scale appropriate for the size of the issues at hand. Change must be significant and sustainable. For capital to produce structural change in the economy, a certain threshold of investment must be reached; for the accumulation of human capital, successive stages must see greater coverage of health, education, and other programs; and for productivity to make a sustained contribution to growth, it must gradually move beyond a small number of large corporations and reach a big portion of small and mid-size businesses, which generate the majority of jobs. When nations are unable to create the pathways that enable changes to reach the required critical mass for sustained growth, they fail to develop. Low- and middle-income traps are failures at scaling change brought on by a variety of factors that constrain productivity growth.
If scaling can be seen as an implicit requirement of successful development, it should be seen also as part of the mission of multilateral development banks. Their original goal was to increase productive capital investment, supplementing insufficient internal savings with external resources. They were instrumental to the “big push” theory of development that could not be understood without a large-scale transfer of capital resources. With agencies moving to see a much greater variety of needs as relevant for economic and social development with limited resources and the “right policies first” approach to development, scale has gradually been displaced as a central priority. This has encouraged interventions that are designed as test runs for novel ideas or as partial solutions to needs or problems, leaving scale out of the priorities. Development banks have been portrayed as merely acting as catalysts for the coordinated action of public and private actors in the development process. The problem is that it is not clear how other actors are being “catalyzed” to affect impact at a scale. The concept of the “knowledge bank,” where the intangible contribution is just as important as the financial one or even more so, also has contributed to downplay the scale logic, even if knowledge is key to any scaling purpose.
As the world becomes aware of the growing storm of several crises occurring simultaneously, things are beginning to change. Global challenges like climate change and recent shocks like the pandemic and the war in Ukraine highlight how fragile past progress in eradicating poverty has been. Fragile and conflict-torn nations are regarded as a complex humanitarian and development priority as well as a danger to global security. The Sustainable Development Goals (SDGs) will not be accomplished on schedule, and only a significant infusion of resources with scaling taken seriously, could change the course. The multilateral development banks are once more viewed as the most practical tool the international community has available to address various crises. To have the greatest impact, they must flex their financial muscles and revamp their “business model”. Scale is making a comeback as a crucial component in the definition of success.
The problem is that this will require a wide turn in the operational reality of multilateral development banks. The project-centered approach, for example, while initially adequate for delivering infrastructure investments, is not necessarily well suited to scaling up initiatives in sectors where the accumulation of capital is not the main driver of scale. Whether the goal is the accumulation of human capital, increasing small firm productivity, the protection of property rights, adapting the watersheds to climate change or the guarantee of gender equality, scale can only be achieved through the right combination of policies, institutional capacities, and sustainable resource allocation. The issue at hand is how multilateral development banks’ business models should evolve to make the best possible use of their available knowledge and financial resources to maximize impact at the required scale.
The Inter-American Development Bank as the subject of the case-study
In the light of these challenges, this case looks at one of the regional multilateral development banks. The Inter-American Development Bank (IDB) was founded in 1959 to operate in Latin America and the Caribbean (LAC) and is regarded as the oldest regional development bank. The IDB is composed of 48 member countries of which 26 are borrowers including all LAC countries except for Cuba and the smallest Caribbean island-states. Developing member countries hold more than 50 percent of the shares, leading some to call it a borrower-driven Bank. However, the United States still retains a 25 percent share that gives veto power in the most important decisions. The IDB Group comprises the IDB, the main window for sovereign-guaranteed operations, IDB Invest (the commercial name of the Inter-American Investment Corporation), which supports private and state-owned enterprises, and IDB Lab, the group’s innovation laboratory.
The IDB was the first development bank with a mission of economic and social development, and to work towards simultaneous national and regional development goals. The IDB group is the region’s primary source of development financing, followed by the World Bank Group (WB) and the Corporacion Andina de Fomento (CAF). It is an institution finely tuned to the region’s political economy, with most staff originating from borrowing countries and highly sensitive to its needs and circumstances. The IDB has demonstrated leadership, supported significant policy transformations, and implemented innovative programs in the areas of social policies, institutional reforms, social infrastructure, or regional integration. One of the strengths of the institution is its convening power for regional policy dialogue that allows for policy transfer and replication. A good illustration of the IDB’s importance in LAC is the financing of social investment funds in the 1990s, which was followed by participation in conditional cash transfer programs, which began in Mexico (Progresa-Oportunidades) and spread to many other countries.
This case study looks at IDB’s institutional standing regarding the mainstreaming of investment operations at scale. Only the IDB sovereign guaranteed window is covered by its purview. The paper does not cover IDB-Invest or IDB-Lab practices except for the degree of integration of the three windows for scaling purposes in country strategies. The analysis that follows is the result of a desk review of IDB strategy, policy, and project documents as well as a few interviews with staff and consultants familiar with the Bank’s priorities and procedures.
The objective is not to identify programs that have been successful in contributing to delivering results on a scale, but to analyze to what extent institutional incentives are aligned with mainstreaming scaling. Some scaling in IDB projects occurs, but as for many other donors, could be ad hoc, opportunistic, or dependent on the initiative of project teams or government partners. To examine whether scaling is an institutional explicit or implicit priority, the following issues will be considered: (i) first, to analyze whether an explicit mandate has been formulated for scaling at an institutional, sector, and country level, strategy and policy documents will be examined; (ii) second, lending practices will be reviewed to determine the degree to which scaling is an institutional practice in project design and execution, with or without an explicit mandate; and, (iii) third, institutional factors that influence scaling will be identified, leading to specific suggestions for how to encourage scaling to be mainstreamed at the IDB.
It is crucial to note that the IDB underwent a leadership change earlier this year, and the new President has emphasized the necessity of prioritizing investments in a few areas and the significance of bolstering the institution’s development effectiveness. He has brought up the need to scale up operations for greater impact in several speeches and interventions. In a partnership agreement recently signed with the World Bank a commitment is made to jointly scale up in certain sectors (like digital education) and regions (the Caribbean and the Amazonia). It is still and open question how this ambition will be implemented operationally and whether scale will mean just an aggregation of resources and other capacities or an effort to further integrate means and ends. A new institutional strategy is being prepared as this document is being written and will be presented to the Board at the end of the year.
Scaling is not explicitly presented in strategy documents as a priority.
The IDB has an elaborate strategy framework that consists of institutional strategy, sector frameworks, and country strategies. They are management-prepared directives with a horizon of 4-6 years that have been formally approved by the Board of Executive Directors. After examining challenges in the region, sector, or country, reviewing lessons learned, and evaluating comparative advantages and positioning, their goal is to set organizational priorities. Priorities typically represent needs that are broadly defined, leaving a lot of room for interpretation and specification. These documents do not address lending, budgetary, or other funding and capacity needs in relation to priorities, which means they play a more vision or perspective setting guidance than specific action commitments. As a result, a gap appears between strategy priorities and actual day-to-day interventions, which is meant to be filled by managerial and technical discretion in programming, other resource allocation decisions and project design and implementation. In the end, actual performance that results from the aggregation of these individual decisions reveals real strategy, no matter what intended strategy is supposed to be.
The current IDB Institutional Strategy was adopted in July 2019. It is presented as the second update of the institutional strategy that was approved with the 2010 Capital Increase. Presenting development challenges in Latin America and the Caribbean and establishing overall institutional priorities take up the first 28 pages of the document’s 38 total pages. Two areas of special focus, “promoting technology and innovation” and “resource mobilization,” are presented as being “key to increase the IDB Group’s value for money and achieve the greatest development impact with existing resources.” The document makes very little mention of how this will be accomplished. When addressing the aggregated impact of the group three windows, it refers to internal coordination tools, such as corporate identity, country strategies, or internal service level agreements, but it says nothing about how to maximize impact through joint scaling i.e., how the three windows could concentrate efforts to achieve specific outcomes or outputs at a significant magnitude.
A Corporate Results Framework (CRF), which is referred to as “a set of metrics to measure progress towards the accomplishment of the strategic objectives,” follows the update of the Institutional Strategy. It has three levels of indicators, the first of which tracks the region’s progress toward the Sustainable Development Goals (SDGs), the second of which provides information on the output contributions made by IDB programs to the achievement of the SDGs. The final one includes organizational and corporate indicators, such as the percentage of lending that goes to various priority areas or the anticipated level of resource mobilization. It is crucial to note that targets have only been set at the third level, which means that neither a financial commitment to regional SDGs’ progress nor to the size of the IDB output contribution has been made.
We can quickly see that there is no correlation between scaling priorities at the regional level and the output figures. For instance, we can see from the most recent progress report that, over the course of two years, slightly more than 12 million people have benefited from anti-poverty projects. For a more precise sense of scale, the number of poor people in LAC is approximately 200 million, with 82 million of them experiencing extreme poverty. However, it is impossible to establish any meaningful scale connection due to the lack of information on the lending resources allocated to each output, disaggregation by country or type of intervention. The same is true for the other output indicators, which also noted sizable differences between beneficiaries and potential target populations. The intended number of health beneficiaries is 80 million, but only 5.8 million students or 0.8 million farmers benefit from education projects or agricultural development initiatives. These numbers don’t reflect a proportion with the overall target population. The data comes from aggregating project data of interventions that may have very different scope and objectives, making it impossible to make any determination of sufficient or insufficient scale. The lack of output targets shows the soft nature of these indicators that do not make the necessary connection between guidance and accountability for the former to be followed. This looks as when the circles are drawn after the bullet has hit the wall.
To have a significant impact, the IDB must consider what its aspiration is in terms of the percentage of a need or problem that it hopes to address, as well as what other factors are essential to achieving it. According to the literature on commercial innovation, the point of coverage to make a difference is approximately 15% of the potential market or issue. This figure can vary according to the specific need or problem being addressed. The point is that a commitment to an outcome magnitude consistent with scaling at an institutional level looks essential for it to become a true priority. Institutional outcome targets could be established, for instance in relation to the SDGs or the National Contributions to the Paris Climate Agreement.
The corporate priorities are the focus of the third level of indicators. Most indicators are centered on the proportion of lending and operations that must be allocated to various priority areas and nations in accordance with the shareholder mandate stated in the Capital Increase. It could be argued that the need to concentrate lending on priority areas can be a powerful incentive to scale up. However, the general wording of the sector priorities allows for ample flexibility in the definition of what counts within each one of them. Only the commitment to allocate at least 30 percent of resources to small countries could be seen as a hard incentive for scaling in that specific context. The IDB sovereign guarantee window comfortably meets the lending targets, according to data presented in the CRF.
There are two objectives for development effectiveness: active projects with satisfactory performance classification must reach 80%, whereas projects with satisfactory results at completion are only expected to reach 70%. The only goal for country strategies is that 90 percent of projects adhere to them. There is no indicator for scaling (or innovation in scaling) that would be aggregated at the corporate level. A point will be made later that basing development effectiveness on success rates of individual projects may be a disincentive to working at scale. Targets for knowledge products that are based on blog readership, book downloads, or a net promoter score, have little bearing on commitments of providing pertinent, actionable advice to maximize impact at scale.
The second tool for guidance that we consider is sector frameworks. Twenty-two sector frameworks and four action plans have been approved. The large number of sectors anticipates a pattern of institutional effort segregation that we will later see reflected in the lending portfolio. Five of these documents—covering agriculture, skills development, housing and urban development, health, and transportation—have been the subject of a desk review. Most of the documents are recent, having received approval within the last four years. They have a similar format and are about fifty pages long. They begin with an evaluation of the regional challenges in each sector, then present the evidence gathered regarding the effectiveness of policies and programs, analyze the lessons learned from IDB operations, and conclude with the sector-specific lines of action. The documents provide a thorough analysis of the sector’s situation in LAC and are also very instructive regarding the policy interventions that have been shown to be successful. The section on lessons learned is based on project analysis and interviews, but it does not include specific empirical data on the operational and country effectiveness of IDB sector interventions or detailed portfolio data with lending operations and their type. They also do not include any assessment of the Bank internal resource capacities to make a difference in the sector.
Many of the documents deal with scaling needs related to regional challenges. In some cases, they are adamant to support universal coverage of basic needs or insist on the need to substantially increase quantity or quality in policies and programs. They are also consistent with the achievement of SDGs in the region and focus on the most relevant gaps. The documents also present, with ample support of evidence, a catalogue of policy interventions that have proven effectiveness in achieving results on a significant scale. However, they are very timid when addressing scaling. A few strategies mention the pilot and innovation initiatives of the IDB Lab or sector departments, but do not follow up with scaling strategies.
The final section presents lines of action, which are a lengthy list of things to do in each sector, supported by evidence and lessons learned. The problem is that they are not connected with resources needed or available, making it impossible to prioritize among them or provide an indication of the best suited for scaling in each context. Scalability assessment criteria could have been used to prioritize the actions identified and provide an incentive to select the most relevant operations. In summary, the documents are excellent state-of-the-art papers that are very helpful for gaining a broad understanding of challenges, identifying potential policy options, and guiding technical design of operations but they are far from indicating a path for resource allocation, which would be essential for mainstreaming scaling at a sector level.
The guidance framework for operational programming and resource allocation to a particular country in a time frame is established in country strategy documents. To ensure alignment with national policy and resource allocation priorities, country strategies’ timeframes are tied to governmental mandates. All strategies have the same format, beginning with a thorough examination of the country’s development challenges and their recent evolution. The IDB’s work in the nation during the prior strategy is reviewed with an account of the outcomes and lessons learned, as well as an acknowledgment of the evaluation office’s recommendations. The country strategy defines priority areas, projects the lending envelope for the period, and includes several considerations for implementation as well as a risk assessment in the actionable part. Country strategies have an annex with a results matrix with goals, expected outcome results, and indicators.
Country strategies offer closer action guidance compared to sector frameworks, but they still have some significant limitations. Three country strategies that correspond to various scaling contexts have been examined for this analysis. The first is Bolivia, a relatively small nation that receives concessional funding from the IDB, followed by Peru, a medium-sized, upper middle-income developing nation, and then Mexico, a large middle-income nation with easy access to financing and where the IDB is supposed to have a less significant role.
The three strategy documents have some interesting things in common: (i) first, they are well-informed and analytically sound documents on the state of each nation’s economy and its challenges. Discussion of social and political challenges is noticeably less extensive; (ii) second, the documents contain a lengthy and detailed section on results that discusses outputs or outcomes attained in relation to country needs but avoids an explicit analysis of performance in comparison to goals planned or the scale of the problem or need; (iii) third, priorities are identified in a general way, frequently lacking a connection with funding requirements, resources available from various sources, and the IDB financing that could be adequate for each priority. This decision is postponed until programming and the design of a particular project; (iv) fourth, the lack of specific output and outcome targets in the results frameworks, limited to defining outcome objectives and follow-up indicators, weakens the accountability for results. No strategy discusses scaling in relation to achieving sector outputs or outcomes; (v) the three IDB windows are presented as separate efforts contributing to a common priority. There is no comprehensive viewpoint on how the combined efforts of the parts can produce outputs and outcomes at the required scale. In other words, scaling across multiple windows is at the whim of mutual adaptation in response to new opportunities and not an anticipated planned effort; (vi) country strategies are the result of a country policy dialogue based on knowledge products that address development challenges and offer recommendations. According to the documents examined, scaling and scaling strategies do not appear to have been discussed or, at the very least, to have informed the conclusions that guide the development of the country strategies; (vii) coordination with other international donors is mentioned, but there is no express commitment to a partnership aiming to scale up operations. Going alone will make scaling very unlikely, at least in countries that receive substantial funding from other donors and are beneficiaries of concessional resources.
Although it doesn’t directly address it, the Bolivia Country Strategy is the one that comes the closest to scaling concerns. The IDB plays the largest financial role in this nation as the top multilateral lender and owner of 30 percent of the external debt. With 20 investment loans approved for a total of $2,369 million in investment loans during the previous strategy and 25 expected outcome objectives, the portfolio still appears to be fragmented. When the results of the Bank’s interventions are presented, output and outcome reach is connected to overall needs, showing the significance of several operations in terms of scale. Results and outputs in the social sectors, as well as in infrastructure or institutional reform, largely correspond to country needs seen on a scale. With additional bank and country commitment to reducing sector dispersion, further scaling progress appears possible in Bolivia.
The Peru Country Strategy is the most disjointed, exhibiting less implicit scale logic and, despite being a larger country, more operational fragmentation. The Bank approved 21 investment projects totaling $1,336 million, with 16 outcome objectives that were listed in the results matrix. The discussion of the outcomes of the previous strategy avoids making a connection to needs at scale and the definition of priorities is vague. Achieving scale in relation to certain needs or problems doesn’t look to be a priority in Peru. The focus seems to be more on demonstration projects intensive in knowledge value added that serve as a test for policy innovations.
Finally, Mexico, LAC’s second-largest economy where the IDB has a small share of total debt, presents a country strategy with a less fragmented picture. The $6,320 million in approvals had been split up into 24 investment loans, with a total of 16 outcomes goals to be met. Results are discussed taking into consideration national stated priorities and needs. Priorities are also worded very broadly, but there is an emphasis placed on integrating the Bank’s projects into potent national initiatives created to have an impact on scale and led by strong national institutions. It can be discussed whether the strategy argues for this to make resource implementation easier or to achieve maximum impact, but both objectives can reinforce each other. In fact, it is not a coincidence that Mexico has led several programs that have demonstrated impact at scale, with the IDB being a supporting partner.
Development Effectiveness Framework
A Development Effectiveness Framework (DEF) was approved by the IDB in 2008 with standards and metrics to be followed by operations and country strategies to demonstrate effectiveness. Standards are verified at entry by a Development Effectiveness Matrix (DEM), a compliance checklist that must be validated for each operation and country strategy by the central development effectiveness department. The DEF was an effort to improve what shareholders at the time believed to be a poorly performing institution when it came to gauging the success of interventions. Only 10% of the projects examined at the time, according to the IDB Evaluation Office, could demonstrate results at completion. However, the DEF does not emphasize scale or sustainability as a requirement for effectiveness. A project’s or a country’s strategy’s relevance is assessed based on the empirical justification of needs and the evidence supporting the potential success of the intervention. Effectiveness of performance is related to the accomplishment of goals and outcomes specified by the intervention itself, without consideration of a desirable scale to maximize impact. Scaling potential is not a reporting variable in monitoring and completion reports.
The IDB Evaluation Office evaluates project completion reports yearly in accordance with the standards agreed by multilateral development banks, considering relevance, effectiveness, efficiency, and sustainability. Achieving scale or at least demonstrating scalability are not dimensions included in the review. According to the last report published, only 53 percent of sovereign guaranteed IDB projects were rated positively for effectiveness, compared with the 70 percent target in the CRF. Many projects show performance monitoring inconsistencies due to changes in project objectives from the original design that have not been validated. The results, which are below other multilateral development banks, pose a challenge that the new IDB management has mentioned as one of its top priorities.
By reforming the DEF, it might be possible to include scale considerations in project planning and execution. Adopting a concept of effectiveness that goes beyond strictly defined project results will be necessary to achieve this. There is evidence having as an objective that most projects are at least marginally successful may act as a deterrent to achieving scale, which is a riskier proposition. It might be preferable to have more projects fail because of implementing innovations, if there are several sizable projects that are simultaneously successful and have a significant impact on scale. This will entail defining developing effectiveness based on overall outcome targets met on a portfolio-wide basis, instead of relying on project-by-project success rates.
We can draw the conclusion that no attempt has been made thus far at an institutional level to define mainstreaming scaling as a priority based on the review above of strategy and policy documents. Innovation is mentioned as a priority in relation to new technologies or policy initiatives, but no commitment is made to provide scaling pathways for successful innovations. Scaling could be argued to be left as an operational concern for project design and implementation. However, the outcome at the operational level will most likely be very uneven in the absence of clear scaling priorities conditioning project selectivity and financial sizing.
A sample of large investment operations’ patterns show an inconsistent focus on scale.
A small sample of the operations approved in 2022 will be examined to identify scaling patterns. In 2022, the IDB approved 96 operations totaling $12,711 million in financing. The approvals included 24 policy-based loans totaling $5,769 million, 24 investment projects totaling $6,392 million, and two special development loans totaling $550 million. A small sample of operations approved in 2022 have been chosen for a desk review to analyze how scaling concerns are dealt with at the project level. They are 7 operations out of the 72 investment loans approved (9.7 percent) that amount to $1,450 million in lending (22.5 percent of total investment lending). The emphasis has not been to search for projects that show results at a scale but to what extent scale is mainstreamed in operations.
Three simple criteria have been followed in the selection of the sample: (i) first, only investment loans have been chosen for analysis. Policy operations could be relevant to a scaling strategy removing barriers through policy reforms, but they would need to be complemented by investment programs. We will look at any connections between the investment loans reviewed and previous if any policy loans. We will look to ; (ii) second, relatively large operations have been chosen in the belief that they would provide more room for scaling requirements to be taken into account in project design; and, (iii) the operations represent a variety of countries and sectors to obtain a relatively wide perspective of scaling concerns across the board. The analysis is presented as follows with the main conclusions reached.
Scale-relevant metrics are used to identify problems or needs to be addressed.
Each operation under consideration is supported by a thorough empirical analysis of the issues and demands it addresses. The issues are carefully investigated and presented with a wealth of metrics. All projects include information on the scope of issues and needs addressed in both absolute (the total population in need of or without a service) and relative numbers (the percentage of units of analysis) terms. Several instances include the following: (i) the Peru Financing Program for Women cites an estimate from the International Finance Corporation showing that 30,000 women owners of small and medium-sized businesses are partially or fully constrained; (ii) the Bolivia Water Resources Project estimates that only 500,000 ha, or 21 percent of the country’s irrigable land, are covered by irrigation; (iii) the operation to universalize early childhood education in Argentina is based on a coverage diagnostics at different ages ; (iv) the Mexico Urban Improvement Mexico Urban Improvement project includes an Urban Social Gap Index developed by the Mexican Government that establishes the level of urban and social gaps equal or above the national average; and (v) the Dominican Republic sanitation program starts by defining the problem in terms of percentage of urban population connected to sanitation services and percentage of receives treatment.
Projects use a range of national and international studies as well as national statistics to support their analysis. Most projects refer to Bank knowledge products that support the diagnosis and offer policy options. Many of them are regional analyses that enable cross-country comparisons of needs or issues. Various studies conducted by other multilateral development banks, international organizations, and academic institutions are also cited. The design of the project is based on a strong body of knowledge that, if applied to a scaling approach, could be very helpful.
Need or problem diagnosis is not followed by an assessment of resources needed to cover for scaling gaps. Only the Peru Financing Program for Women cites a study indicating a $481 million funding gap for women-owned businesses. The Bolivian irrigation program refers to the sizeable financing needs in the water sector but does not give precise numbers. Although the Argentina program is a part of the country’s national universalization effort, it does not include data on the total funding requirements to close the gaps. The operation in the Dominican Republic is a component of a strategy to guarantee universal coverage of sanitation facilities but does not provide an estimate of total funding needs. In the case of Haiti, there is no estimate of the volume of resources required to provide an expanded safety net for every vulnerable population throughout the nation.
Projects consist of interventions with the potential to be scaled up but lack a scaling strategy.
Every operation consists of policy interventions that are backed by solid evidence on their beneficial effects in resolving the issue or meeting the demand. Many of them are predicated on prior experience and accomplishments made in the same country. The Dominican Republic project mentions the approval of 170 operations in the sector with a total financing of $28 billion between 2007 and 2021 and successful scale-related results (like an increase in coverage from 11 to 52 percent in the Santiago metropolitan area). Also, operations in Bolivia, Mexico, Argentina, Panama, and Haiti are part of a series of interventions in the same sector with closely related goals. There seems to be no doubt that most interventions have the potential to cover needs on a scale but rarely discuss scaling as a project strategy or jointly with results achieved in past projects and potential future interventions. There is a clear preference to consider each project on its own and to avoid overreaching with scaling commitments that are not seen as part of the mandate.
Several projects also profit from the extensive Bank experience in these operations in other countries, like cash transfers, irrigation, digital government or water and sanitation. There are a few cases mentioning the replication of successful interventions undertaken in other countries, like the projects in Peru, Chile, or Panama. Horizontal scaling across countries seems to be a well internalized practice in project design, benefiting from knowledge exchange and dissemination at the regional level and the use of active policy networks. The IDB seems to have well-established incentives to pursue this kind of scaling.
There is no connection made between the investment projects analyzed and past or future development policy operations regarding scaling joint strategies. As we mentioned above, policy loans cannot deliver scale alone but can contribute to scaling ensuring policy and institutional reforms that are essential for moving a program up to a sizeable magnitude. Project documents also refer to technical cooperation grants in support of either knowledge generation or project implementation support. However, a connection with project scaling strategies has not been found.
Most projects focus on achieving partial positive results on the limited scope they choose as relevant for the project. This can refer to a particular geographic region, or a volume of outputs to be financed (such as loans in Peru, urban works in Mexico, or renovated childcare facilities in Argentina). Most projects include as an outcome an absolute number of beneficiaries, but data are not provided to pinpoint the program’s precise contribution to closing the coverage gap. This happens even when the programs’ titles express a desire for universal coverage, as in Argentina or the Dominican Republic. The safety net for Haiti is the closest to have been planned with scaling up logic in mind. In the chosen areas of intervention, goals are set to reach 18.5 percent of the vulnerable population and there is an explicit institutional development strategy to allow for further scaling. Panama Digital also presents a scaled-up objective based on a prior project, aiming to digitize up to 30 per cent of all catalogued public administrative transactions.
Several operations are thought to contribute to a national plan or program designed with scaling objectives. In Bolivia, Mexico, the Dominican Republic, or Argentina, projects are framed as a contribution to larger government initiatives. This arrangement might be perfect to support a scaling perspective in coordination with other national or international resources committed to the same effort. However, none of the interventions under review discuss the scaling logic of the national plan or program or the role of the Bank project in this regard. It appears that the Bank would prefer to limit its involvement to the financial support it offers and the intended outcomes of the operation, instead of aiming at the bigger picture. This is the case even when the operation is part of a series of Bank interventions to address the same need and within the same national program.
Project sizing is not connected with requirements for scaling.
A relevant scale in relation to the problem is not what determines the size of the project. None of the projects under consideration explains why the financial envelope was provided in relation to the need or evaluates the future additional resources that would be required to gradually increase coverage, whether using resources from the Bank or from domestic sources. It appears that a country programming exercise matches supply and demand for resources using a primarily financial approach and then distributes resources among sectors and types of loan instruments. The financial sizing of projects looks to be a condition outside the control of the project design team, that must adapt to the envelope given. In this context, scaling up might be seen out of scope.
The project’s anticipated outputs and outcomes are completely pertinent to and consistent with the need or issue raised. In most projects they are not, however, defined in relation to the scale of the need or problem being addressed, even when the project achieves a reasonable level of coverage. For instance, the Peru Program for Women Entrepreneurs seeks to assist 5,144 of the 30,000 financially strapped women. A significant portion of the estimated 5 million people who are food insecure—375,000—are anticipated to benefit from the project in Haiti. The program in Chile plans to increase lending to small and medium-sized businesses in the targeted areas by more than 10%. In other instances, like in Bolivia, the goal—to irrigate 13,881 has out of a possible 2,000,000 has of irrigable land—is relatively modest in comparison to the size of the issue. In most projects the number of project beneficiaries is evaluated rather than the program’s reach in relation to overall needs.
The projects do not include substantial counterparty national resources or scaling-relevant third-party allocations. Simply put, counterparty resources are absent from most projects. Occasionally, the counterparty is extremely small, as in the case of Argentina’s program ($750,000 compared to $200 million of lending). It is connected to a very specific project component in the projects that have a larger portion of counterparty resources. Out of a $74.4 million project, the Panama Digital project includes $14.4 million in local funding to create a hub for digital infrastructure. Out of $190 million, only the Dominican Republic sanitation program has a significant national counterparty of $50 million, which is used to partially finance one of the project’s planned infrastructure works. Projects in Haiti and Bolivia mention the participation of other development agencies in the same sector, but do not go into detail about the funding or the outcomes. There is not a joint scaling approach considering resources from other development partners that contribute to the same needs or problem. This lack of significant counterparty resources being included in the project not only affects the magnitude and impact of the intervention but also compromises sustainability as there is no provision of government funding to substitute for IDB funding.
Many programs make use of long-term investment instruments that commit larger sums of money over ten years. This is the case with the Regional Productive Development Program of Chile, the Bolivian Irrigation Program, and the Mexico Urban Improvement Program, which are the first operations of credit lines that more than triple the initial project’s size. Argentina’s program for early childhood development is the second operation of the credit line. However, all project results projections are made without considering the overall impact of the entire financial effort over the course of the period. The Argentine operation does not show the combined aggregate effects of the two projects or the evolution of magnitudes in relation to the size of the need. It appears that the primary purpose of the credit line modality is to program bank lending resources and expedite the processing of subsequent projects and it has not been thought to maximize impact at a large scale. Conditions to approve the subsequent operations of the credit line are based on reaching disbursement targets.
Scaling requirements are not considered in project implementation, monitoring, and evaluation arrangements.
Implementation requirements reflect exclusively the needs of the project and do not expand on scaling up strategies. Their main concerns are having the ability to execute financially and physically as well as adhering to fiduciary and administrative requirements set forth by the Bank. Even if the project is to be followed by another one or if multiple donors provide funding to the same sector and use the same project executing unit, as in Bolivia, they do not take scaling requirements into account or present a pathway for scaling. Pooling resources within the same unit appears to serve just as an administrative goal for administrative harmonization but not as a scaling logic. They are not viewed as a component of a plan to address the issue or need on a large scale, but rather as independent efforts.
Many of the projects are carried out by government organizations with the capacity to expand operations. This is the case with COFIDE, a top second-tier financial institution in Peru that finances small and medium-sized businesses through a network of financial intermediaries and exhibits significant capillarity. The same is true of CORFO in Chile, where the initiative depends on the country’s already-existing Medium and Small Size Credit Program. The execution-related government departments in Mexico, Argentina, or INAPA in the Dominican Republic are all reputable organizations that manage much bigger programs. The project in Haiti is a component of a larger initiative to create a social safety net, which is supported by the same social registry in the Ministry of Social Affairs and carried out by the same executing agency. Some of the programs invest in institutional strengthening of the implementing agencies, like in Bolivia, Dominican Republic, Haiti, or Panama, but most of these efforts are not linked to specific scaling strategies.
Only in relation to specific project outcomes and financial investments is sustainability discussed. When infrastructure is involved, maintenance provisions are mentioned. Only the project in Chile links sustainability with the future availability of budgetary appropriations so the project objectives can be followed beyond the project execution timeline. No project specifically links scalability of project results to sustainability. Measures related to potential scaling are not included in the monitoring and evaluation arrangements, which concentrate on the results identified in the projects. Only the project in Bolivia chooses a quasi-experimental evaluation, and most evaluations consist of before and after assessments of project results.
Factors influencing the IDB’s lack of systematic attention to mainstreaming scaling.
The lack of systematic IDB mainstreaming scale up focus can be attributed to several factors that are highlighted from the previous overview of strategies and operations. They also provide insight into the levers that could be useful for intensifying the focus on scaling in IDB operations. In this final section, these factors are reviewed with some preliminary suggestions.
Lack of effective strategic guidance on scaling
IDB strategies offer very little in the way of operational decision-making guidance. They intentionally omit details from priority definitions to give decision-makers more leeway in project selection and design. It is not surprising that projects perfectly align with country strategy priorities or that country strategies demonstrate complete compliance with institutional priorities. Strategy results frameworks do not provide a strong accountability framework on outputs and outcomes, missing an opportunity to signal scaling priorities.
In this context, it is not surprising that none of the reviewed strategy and policy documents list scaling as an operational priority. At the institutional or national level, metrics of results frameworks do not reflect a scaling logic to maximize impact and sector frameworks do not contain any result commitment on outputs or outcomes. The IDB development effectiveness standards were created more than 15 years ago to fill a gap in the institution’s ability to demonstrate results, but they have not been sufficient to create powerful incentives to reach a high percentage of successful project outcomes. Scaling is not one of the standards used when designing the framework and it is not used by the Evaluation Office to evaluate project effectiveness.
Without a strong top-down signal that specifically identifies incentive changes, significant scaling is unlikely to occur given the fierce competition among sectors for lending resources and the financial incentives on approvals and disbursements that rule the institution. Mainstreaming scaling must be seen first as an exercise in political leadership, setting the right incentives to find their way into resource allocation, programming and project design and implementation. A combination of specific scale commitments on important outputs and outcomes set in the institutional strategy and one mandatory scaling priority by country strategy may be able to steer incentives in the right direction. This will mean setting outcome targets in priority areas that will effectively influence decisions taken at the level of country strategy and programming.
The best candidates for a special scaling effort are the small and low-development nations where IDB financing accounts for a sizable portion of investment resources. To achieve a size that is appropriate and guarantees sustainability, this will require collaboration with other international donors on joint strategies and operations as well as the inclusion of counterparty national resources. To achieve impact on scale, it will also be necessary to combine interventions in policy, investment, and technical cooperation. Country strategies are the vehicle through which the challenges and pathways to scaling can be effectively addressed, followed by consistent programming.
To enable a cooperative approach to scaling when collaborating with other donors, an examination of the present operational policies and procedures will be necessary. This is pertinent to the recently signed partnership agreement with the World Bank, that will need to translate into concrete operational guidelines to be applied. External resources are only included now within the project scope and counted to the achievement of outputs and outcomes, when there is an explicit co-financing agreement with the IDB, meaning that the Bank receives the resources from the donor and executes them as if they were its own. More than 20 years ago the Sector-Wide Approach, which encourages collaborative planning and implementation by donors around a sector plan owned by national counterparts, was advocated to pool resources and replace the project-by-project logic. Although this has been a part of the IDB tool menu for a long time, there hasn’t been much use of them recently. Their lack of practical application is most likely due to the capacity required of the national counterparties and the high cost of coordination requirements among donors.
Sector fragmentation makes it difficult to scale operations.
Without regard to comparative advantages, the IDB has developed a comprehensive catalogue of sector interventions that spread across developing needs and problems. In addition to impeding critical mass and knowledge specialization, this overextension also distributes financial resources among a broad range of operations without regard to needs at scale. Most national and international development agencies are afflicted by what is known as the “mission creep syndrome,” that also prevents a more useful division of labor among them. Sector fragmentation reduces the potential for scaling and encourages projects to be viewed experimentally as one-off interventions.
In investment lending, which has smaller average and median sizes than policy support operations, sector fragmentation is frequently more pronounced. Policy operations are preferred to meet the country’s funding requirements because of the shorter project cycle and the predictability of disbursements. After the policy loans have been set aside, the remaining funds are distributed among the numerous sectors vying for attention, both at the Bank and the country level. Policy and personal networks that are active during programming to generate sector demand at the national level. The allocation of Bank resources is also a political balancing act at a country level, also working towards fragmentation.
The IDB needs to specialize and narrow its focus if it wants to mainstream scaling and maximize impact. The new President has insisted in his speeches on the need of more and better prioritization. This still needs to be translated into actionable priorities supported by sizable lending resources, budgetary allocations, and funds for technical assistance. Prioritization could ideally be established by a selective definition of outcome targets in no more than four or six areas that have the most potential. Sector frameworks should identify the interventions that meet scalability criteria to allow for selectivity in programming. The social sectors, for instance, have historically been an area where IDB comparative advantage has been acknowledged by borrowers will make strong candidates for expansion. They reflect the requirements of a middle-income area that is plagued by significant pockets of poverty and unsettling inequality, and they are also consistent with potential bankable scaling strategies. If political will exists, a lot can be done in this direction.
Programming decisions are primarily influenced by country demand and financial requirements.
The IDB, as a regional institution primarily led by borrowers, is fundamentally a demand driven lender. Countries anticipate that the IDB will respond to their financial needs and policy priorities. This might be advantageous for scaling if supply and demand are balanced by substantial long-term sector commitments and operational planning or where the government itself has long-term scaling plans (Mexico provides examples of large federal programs that the IBD is supporting). In fact, borrowers show preference for large operations that address national priorities and are relatively easy to disburse. A quick glance at the number of projects approved in the same sector and country and in line with national plans or programs would suggest that this is already happening. The same conclusion can be drawn from the use of credit lines that expand for long periods of time.
A closer examination of project design patterns, however, contradicts that conclusion. Short-term borrower and bank financial objectives and other internal incentives drive project design to keep a narrow scope and to avoid scale commitments. Even when projects can be placed in a larger context, because of past and future interventions, projects are primarily driven by their specific one-off goals and steer clear of scale commitments. The lack of a solid rational link between the scale of financial requirements and the allocation of lending resources is a factor that plays against streamlining scaling.
The programming exercise and the country dialogue that comes before it need to concentrate on selectivity and look to achieve more impact at scale for a different picture to emerge. Sector scaling opportunities and programming choices, which appear to be made with a short-term and financial objectives framework, require more integration. The Bank could suggest two or three scaling opportunities in the policy dialogue in line with government priorities and financial goals. A concerted effort by development agencies to focus on scale with a joint accountability framework will be advantageous to scaling in countries that benefit from concessional resources. The country dialogue in other nations will be guided toward scaling opportunities by shared scaling priorities, such as those established in the recently signed partnership between the IDB and the WB in the Amazonia and the Caribbean. Other partnerships with private or public sector entities can also be advantageous to advance scaling.
The lending instruments available today are not well suited for expanding interventions to scale.
The way that lending instruments are deployed compels borrowers to choose between financial and non-financial additionality. Borrowers’ preferred instrument, the policy loan, which disburses quickly and easily but lacks a long-term engagement and resource allocation consistent with scaling up, is used to ensure predictable financial flows. After policy-loans, borrowers’ second choice are investment operations that can disburse funds as quickly and easily as possible, according to the least “hassle cost” factor. Expenditures that can be presented as eligible for Bank funding in policy areas where the country has already made significant appropriations would be preferred, delinking Bank’s financial additionality to scale up and minimizing the potential of the knowledge contribution. In the past, a counterparty proportional contribution to investment projects was required and had to be spent pari passu with bank resources. This was discontinued to speed up execution and facilitate investment project demand but has reduced scaling potential.
Pay for results programs might be a good fit for large-scale success. Their original reasoning was different, though. They were developed based on the theory that incentives for results would be more effective at motivating national actions than the conventional procedural accountability based on spending eligibility and recognition. In this way, pay for results programs lower transaction costs and improve the predictability of disbursement. The issue is that institutional demands placed on borrowers have restricted the use of these instruments. They work best for middle-income countries that exhibit strong institutional capacity and where country systems can be trusted for project execution. Only the loan to Chile among the operations under review pays for results. Unfortunately, result-based lending is not available to the less developed nations which are the ones with the biggest funding gaps to achieve results at a scale. Because they are created based on the achievement of outputs, which are a crucial step to achieving scale, pay for results programs could be useful for scaling. This indicates that a significant range of needs or problems can be directly accommodated by the project design. This will require adequate financial sizing of the program and the design of a pathway consistent with all other scaling requirements.
To mainstream scaling, the IDB would benefit from implementing an impact lending modality that combines large country and bank long-term commitments with predictable disbursement conditions. Outputs and outcomes would need to meet certain scaling conditions and be part of a national program with strong political backing and administrative support. The instrument needs a way to incorporate the resources quickly and easily from counterparties and other international donors needed to achieve and sustain scale. The existing conditional credit line instrument could be adapted to become a scale-oriented mechanism. It needs to be connected to a straightforward disbursement mechanism based on additional budgetary spending and the fulfillment of policy and institutional conditions consistent with a scaling pathway. This could be a powerful inducement to borrowers to partner with the IDB on scaling up programs.
A scaling perspective is absent in the IDB project cycle.
As we’ve already mentioned, mainstreaming scaling is a problem of both magnitude and process. In addition to interventions commensurate to problems or needs, a process that enables successful innovations to gradually scale up and have a significant impact is also necessary. This necessitates a process of stages where interventions go through meeting pre-established criteria at each stage to have more resources available. In this way, innovations that have been demonstrated to be successful have a built-in process incentive to be scaled under certain circumstances. These criteria, which list interventions that are appropriate for scaling and could profit from a streamlined approval and disbursement cycle, can also be included in sector frameworks, and become an incentive for selectivity in programming exercises.
There are no specifications that are consistent with a process perspective of scaling in the IDB project cycle. Even when projects are repeating previously successful operations, which gives them a scaling positive background, they do not appear to profit from a streamlined process or more adaptable financial sizing. Being a part of a conditional credit line offers a process benefit, but it’s tied to successful disbursement rather than scaling success. It’s also crucial to remember that the project cycle does not take the existence of IDB Lab as a platform for early scaling into account. IDB Lab projects are mentioned in strategies or programs as a supplement to other projects or larger priorities rather than as an integral part of a scaling cycle.
Implementation arrangements and monitoring and evaluation tools lack scaling perspective.
The appropriate delivery mechanisms must be in place to scale up interventions. This is more demanding than ensuring a one-off successful project implementation. Potential scale and scope economies or diseconomies must be evaluated and projected to cost estimates when scaling is an endeavor. Delivering at higher magnitudes necessitates not only more resources but also new or expanded institutional and policy frameworks. Most IDB projects execute through temporary project implementation units to reduce implementation risks and accelerate results. This is unlikely to result in a sustainable framework to scale up delivery. Building long-term institutional and financial capacity is essential to scale up interventions and projects should be designed to support this, including through their implementation modalities (by relying on existing government institutions, instead of project implementation units).
Scaling requirements and strategies must be compatible with adequate monitoring and evaluation mechanisms. This means that they examine how results have been achieved in addition to comparing planned and actual figures and whether the conditions for scaling beyond project end are being created by the project as it is implemented. Rigorous impact evaluations have been used by programs that have been successful at scaling, allowing for adjustments in subsequent phases of the process. The IDB has expanded the use of impact evaluations because of the implementation of the development effectiveness framework. More can be done to make monitoring and evaluation tools relevant for scaling. The expected emphasis of the Bank on development effectiveness and the revamping of existing instruments provides an opportunity to stress the scaling dimension.
Finally, it appears that complying with procedures and policies that have developed over time through management and Board decisions is the main concern for project implementation and monitoring. The objective is to make sure that projects are carried out correctly, considering social and environmental needs and adopting fiduciary standards to prevent money from being misused. The problem is that these requirements add to the workload of implementation and discourage strategies that go beyond a narrow view of the project’s scope. Without clear incentives and additional resources, a scaling mandate might be seen as competing with other obligations and priorities that support a more restricted view of project execution.