Economic Advisory Council November 2019 Meeting Minutes

November 1, 2019
8:30 am – 2:00 pm
Millennium Challenge Corporation
7^th Floor Conference Rooms

Agenda

Meeting began at 9:00 am Eastern time.

Introductions, Amendments to By-laws & EAC Chair Selection

MCC CEO Sean Cairncross opened the meeting by welcoming the Economic Advisory Council (EAC) to MCC, noting the diverse academic institutions that they represent and thanking them for their service. MCC’s Vice President for Policy and Evaluation Tom Kelly, following on the remarks of the CEO, noted the important work that MCC does and the importance of the EAC to MCC’s mission.

Following the introduction, the EAC’s Designated Federal Officer, Brian Epley, asked the EAC to adopt an amendment to the EAC’s by-laws to include the role of an EAC Chair to serve for the remainder of the EAC’s term (until October 2020), who would serve as a point of contact for coordinating with MCC between meetings and represent the interest of the EAC members. The amendment was adopted by the EAC members.

Before the meeting, members were asked by email to self-nominate as Chair. MCC selected, and members subsequently and unanimously adopted, Shahrokh Fardoust as the EAC Chair for its 2018-2020 session.

Session I: Cost Recovery and Tariff Policy in MCC Compacts

Session Lead: Emmanuelle Auriol, Toulouse School of Economics, University of Toulouse
Briefing note: Cost Recovery and Tariff Policy in MCC Compacts

MCC’s infrastructure sustainability goals often entail incorporating cost recovery requirements, often via user tariffs. However, cost sustainability can compete with other goals, including equitable distribution and pricing. Members of the EAC offered feedback to MCC practices, ranging from the assumptions underlying the treatment of subsidies when calculating economic rates of return (ERR), cross-subsidization of infrastructure among different user classes or segments, and the underlying requirement that infrastructure achieve full sustainability via user tariffs. Additional takeaways focused on the implicit prioritization of infrastructure investments in the public eye and the quality-price tradeoff.

Balancing Sustainability with Equity

MCC’s current goal of ensuring cost recovery does not account for the beneficiaries of a project, and as a result, efforts to recover costs frequently fall on segments of users who are least able to pay, i.e. the poor. As such, while ex ante ERRs may suggest surpassing a 10 percent ERR, satisfying an important investment criteria, in reality such projects often fail to achieve cost recovery. For this reason, members of the council made a variety of recommendations for consideration. 

• Identify beneficiaries at the beginning of the constraints analysis. This not only sharpens the analysis on potential poverty effects, but also trains MCC’s attention on the feasibility of cost recovery when weighing infrastructure investment plans.
• Allow ERRs to separately account for funds spent from government’s general treasury. This potentially allows otherwise sub-threshold projects to reach the 10 percent ERR.
• Subsidies received from general funds may vary with the nature of the project. The most “basic” services may merit greater subsidies to compensate for lower tariffs. Particularly when the beneficiaries are poor. For example, water has externalities (better health and nutrition, especially for children; reduction in communicable diseases, etc.) which are not present at the household level for electricity, justifying subsidies. If households perceive water to be more vital than electricity, water treatment infrastructure costs may justify lower cost recovery via user tariffs.  This points to the importance of distinguishing different needs/social benefits across sectors and services.
• Absent subsidies from general funds, cross-subsidizing a utility’s expenses among different segments and classes of users (e.g. rich and poor, industrial and residential) may be a viable strategy for achieving equity while meeting the hurdle rate. Block pricing schemes have been used for this purpose in the past, but poor estimates of demand at different price levels have often led to low cost recoveries.
• Experience with block targeting of subsidies is weak and likely to fail. There are problems with exclusion and inclusion.  Giving cash to the poor may be a better way to subsidize (or paying for connections, which bring them on grid).
• Consider service provider guarantees:  e.g. in housing, where there is an agreement with tenants and if the provider breaks the agreement then the tenants don’t have to pay, and there is a fee or sanction for the service providers
• Communication strategies are important for tariff adjustments to ensure the rationale, opportunity costs, and commitment to quality improvements are understood by the public. 

Quality and Prices

EAC members stressed that a significant barrier to increases in tariffs is the perception that services are low quality or that services should be delivered cost free to the poor and vulnerable. Often, governments are reluctant to raise tariffs for fear of political resistance. While some of the challenge may be addressed with better communications, the quality and cost trade-off remains very salient and vary widely across countries. 

Users view service quality as an important element in their willingness to pay for utilities and infrastructure. However, absent any assurances that quality will rise (e.g. more reliable electricity), resistance to tariff increases will be strong by all users. Service providers must strike a credible bargain with users guaranteeing higher quality service in exchange for (and before implementing) higher prices. Given this, EAC members noted that MCC should consider contracting mechanisms which motivate and link service quality improvements, as well as equity considerations, to tariff increases in situations where quality has been a major constraint to consumer willingness to pay.

Session 2: Using Beneficiary Analysis to inform decision making at MCC

Session Lead: Alan Gelb, Senior Fellow, Center for Global Development
Briefing note: Using Analysis of Beneficiaries to Inform Decision Making at MCC

This session asked how MCC should bring beneficiary analysis (BA) into MCC and government deliberation to inform decisions on compact and project choice. BA as currently practiced is residual, it is undertaken at the end of MCC’s decision process. The question posed was.

• How could this analysis be done earlier?
• What are some alternative methodologies to do this analysis?

The EAC began with clarifying questions:

• Has MCC ever rejected a project because of distributional issues? – Not precisely, but there are sectors where MCC has historically not invested due to poor economic linkages, such as in mining.
• Has MCC considered redistributing benefits after choosing the project with the highest growth? An EAC member noted that the World Bank had tried this and didn’t recommend it. Another member commented that the costs of raising revenue and the associated distributional consequences (even if expenditures are progressive, regressive revenue tools may offset this).
• Does MCC conduct growth incidence analysis, or conducted a constraints analysis on the poorest quintile? No, it has not.  
• Does MCC consider partner social country preferences? Yes, there is close collaboration with government in CA preparation. Tunisia is a recent example of explicitly reflecting social concerns of the government.  

Discussion came back to a key question: “How could MCC do a parallel exercise of identifying binding constraints to poverty reduction alongside the current model, which focuses on identifying binding constraints to growth?”  Both can inform sector and project selection, and help bring issues of distribution and sustainability into focus.  Several points were made by EAC members:

• BA as currently practiced at MCC appears redundant. The BA work should be central to sector and project identification if poverty reduction is to remain central to MCC’s mission.  One way to do this is by using distributional weights in the CBA calculations for passing the ERR threshold. However, this is easier to do in theory than in practice; the World Bank experience was not positive. This also ignores general equilibrium effects.
• The question poses a dual optimization problem. The two goals may or may not be compatible depending on the circumstances. MCC should look project by project.  Poverty reduction has a distributional impact, but may not reduce the Gini coefficient.
• Partial versus general equilibrium incidence analysis is key in BA. You don’t want to forego important interventions where the first round effect may not be pro-poor, if prior empirical evidence of strong general equilibrium distributional outcomes are credible.
• MCC could consider introducing weights on BA outcomes, but also consider outcomes other than poverty, such as women, youth, etc. depending on national priorities.
• Tradeoffs between growth and poverty reduction may be an artifice of failing to measure social rate of return due to poverty reduction.
• The World Bank experimented with downstream benefit focus after projects are selected, but found a poverty reduction objective required earlier, upstream identification of interventions.
• Besides purely equity considerations distributional impact may be important for economic sustainability of project outcomes, and should not be neglected (e.g. consider the case of Tunisia). The constraints analysis work from the start should be framed around pro-poor growth and constraints to it. 

Several EAC members agreed that the BA work must come early and be built into the constraints analysis.  One voice of dissent noted that MCC should not commit to just helping those with income below $2/day.  Shifting to a pro-poor growth approach led to discussion of the political economy of MCC’s objective function: is MCC’s authorizing environment prepared to support a pro-poor growth stance? 

The key message in summary was that the BA currently practiced fulfills a reporting function, but cannot guide selection of projects in a meaningful way.  Moving upstream would be key to embedding a poverty focus alongside the growth mission of MCC.  To do this will require bringing distributional and sustainability issues into the growth analytics.  The EAC broadly recommended piloting this approach, with possible technical inputs on the design from the council, on a compact country where distribution issues are a concern and use relevant indicators.   In this context, members of the Council requested that MCC report back on how the recommendations are taken up and if there are results from any piloted work MCC decides to undertake.

Working Lunch | Jump-starting Africa

Speaker: Célestin Monga, Senior Advisor, Development Economics Department, World Bank Group & former Chief Economist, African Development Bank Group
Presentation: Jump Starting Africa

Following a short break for lunch, EAC member Célestin Monga, now a Senior Advisor at the World Bank, presented his work on structural transformation in Africa.

His main points are summarized with the following:

• Sustained growth in Africa is within reach. There are investment opportunities yet, as money deposited by Africans outside Africa earns negative interest rates. Skilled and unskilled labor are also increasingly available and mobile.
• Africa is not attracting or using enough good ideas. Agencies across the continent do not organize enough to learn about critical issues facing African development. Successful development experiences (South Korea, China, Costa Rica and Mauritius, etc) can offer valuable lessons.
• The result is that there has been very little structural transformation in Africa: countries’ exports are still dominated by commodities, as they did 30 years ago. Tax revenues are declining in share of GDP, the youth demographic in need of training has increased, and Africa is trapped into implementing unrealistic agendas.
• Unrealistic Agenda #1: Services should be a priority and countries can skip manufacturing.
  • There is a notion that Africa can skip industrialization and go straight to a post-industrial service-based economy.
  • The problem is that the kind of services which could sustain growth require higher levels of skills and education than typically seen in Africa. The speaker noted that India created 10 million jobs in IT and related industries, but still does not create enough jobs through services.
• Unrealistic Agenda #2: It’s enough to just improve the business environment.
  • There is not a single example of a country that has achieved sustained growth starting with reforms to the business environment: reforming the business environment simply cannot be the only precondition for growth.
  • Improving the business environment is insufficient. Political economy needs to be taken seriously and there is a need to look beyond the project at sectors and industries. What is the private sector doing? Where is the competitive advantage?
  • This leads to a focus on corruption, even as larger issues are ignored. For example, even if we assume that 30% of the budget is lost to corruption, what happened to the other 70%?
• New Structural Economics (NSE) provides a blueprint for “quick wins”. There should be no preconditions and we should avoid picking the wrong model: Burundi is not (yet) Switzerland. The approach for Africa to follow could include:
  • Identifying sectors with latent comparitive advantage;
  • Removing constraints for firms working in these sectors with strategic investments in infrastructure, targeted reforms, skills formation and access to finance;
  • Scaling-up self-discovery by the private sector with temporary incentives; and
  • Transforming SEZs and EPZs into industrial clusters.
• Beyond these, MCC’s focus on projects is too “micro”. Instead the focus should be on sectors and industries. Together these interventions could attract foreign and domestic economies in Africa.

The presentation was followed by questions from members:

• What countries in Africa today are closest to the activist development framework presented? Rwanda and Ethiopia were cited as examples where this approach is working.
• Agriculture is sensitive to climate which is changing: What are the risks due to climate change in Africa and what research needs to be done on this issue? Development in Africa is viewed as a global public good, but despite promises by the global community, little is being done to help. Agriculture has the potential to generate jobs and growth, but we need to look at the situation more realistically.
• Low tax effort has a lot to do with these issues: How would you change the policy framework given that the continent is a resource rich region and resources are not adequately taxed? For any resource rich country, the challenge is to generate and use public resources well in order to create formal sector jobs which keep the young employed and “out of trouble”.

Session 3: Opportunities for Results-Based Financing at MCC

Session Lead: Shantayanan Devarajan, Georgetown University
Results-Based Financing at MCC

Following an introduction by the EAC chair, the Session Lead opened by noting that the idea for Results-Based Financing (RBF) came about as a result of frustration that the recipient country is not doing what’s needed to achieve results. If disbursement is linked to results, then behavior may change.  This also disciplines the donor who may be incentivized to disburse without achieving results. In practice, there have been issues with RBF: lack of data, pressure to disburse, and manipulation of the process, among others.  The donor-country relationship may be better treated as a cooperative relationship than as a principle-agent problem (as the MCC note suggested). In a cooperative relationship the risk is shared, not allocated to the recipient who may not be able to bear the risk.  This is especially important since the recipient government is not a single monolithic entity.  Some notable problems with RBF highlighted by the Session Lead:

• Output indicators are problematic in RBF practice, particularly in data poor environments.
• RBF approaches work best where the evidence for the theory of change, including all assumptions embedded in the theory of change, have been validated.
• Corruption/manipulation with the process: artificially lowering the denominator or boosting the numerator of disbursement-linked indicators.
• Loopholes emerge from pressures on donors to disburse and countries to take the money
• The RBF framework transfer risk to the recipient country, which may not be well equipped to manage the risk. The risk is may be difficult to measure and monitor. 

The EAC discussion raised several points. 

• If a recipient government is ill-equipped to manage the RBF risks, there must be attention to tackling this and improving service delivery systems. But also note that some incentives and risks are necessary.
• The World Bank experience with Program for Results (P4R) provides insights. One challenge there is how to apply safeguards (social, environmental) with RBF instruments.  This may prove a big constraint to MCC.  
• The discussion of RBF is contaminated by pay-for performance issues. The question is do we have a moral hazard problem between the donor and the government? One advantage of MCC is that it’s tied to an economic model and it would be interesting to try pilots of RBF on the model. EAC asked what can be learned from the Sierra Leone and Moroccan experience.
• Several points were made about the specific Sierra Leone and Moroccan cases, but more information was needed to go into details of their design. [Noting the value of greater MCC expertise to help direct discussion at EAC meetings in the future]. Regarding who should bare the risk and whether RBF is appropriate requires much more information on country context.  
• For MCC the best fit may be in using RBF in infrastructure projects with well identified and measured outputs. If the risk is borne by the contractor in the country it may be best to shift monitoring responsibilities to the country. RBF could be used across stages of infrastructure.
• What is driving the demand for P4R programs? Having governments design their own RBF programs is difficult because there are many members of government with multiple levels of stakeholders. Could donors such as MCC, working with the local government, break the political logjam using P4R-like instruments to work within the system? Members of the EAC responded by pointing out that the frameworks for P4R at the World Bank were found to be acceptable based on early evaluations—though it had been pointed out that in many cases more analytical work was needed to underpin the theory of change for specific P4R projects. However, there is also a perception that some P4R programs were getting around safeguards. While the literature focuses on the attractiveness of getting more money, in practice P4R doesn’t actually result in the recipient receiving more money, so that it is an open question why countries choose P4R.
• Important to note that governments are not single entities and priorities plus incentives vary across ministries, parties, and across levels of government (national, provincial, and local). This complicates how contracts can be designed and executed to advance objectives.  Often one government office may be frustrated that another entity is not behaving as they would prefer.  In this space RBF may enhance MCC’s effectiveness.

Members of the EAC also identified a number of potentially critical issues: how much uncertainty is: (1) related to actions that the recipient needs to take and being unsure if they will take it, and (2) related to whether there are gaps in the theory of change. If the majority of uncertainty stems from the potential gaps in the theory of change, then the program is very risky.  MCC should learn more from the piloted RBF program in Sierra Leone.

One final issue raised by members are the implications for MCC’s investment safeguards. While MCC has social and environmental safeguards, there are questions about the degree to which an RBF program opens certain safeguard loopholes.  A number of members noted that, while demand for the World Bank’s P4R mechanism is high, one reason for this high demand may be as regards enforcement of safeguards for environmental or social protection, since in P4R programs safeguards are to follow country policies.  In practice this leads to P4R being restricted to areas where environmental and social risks are lower, and country systems are effectively equivalent to WB standards.  In addition to having environmental safeguards in all its programs, MCC also has a fixed 5-year implementation timeline, which EAC members agreed could be a significant constraint:  at the World Bank, loans can always be extended if disbursement targets are not met, but MCC lacks this flexibility.

Opportunity for Public Comment

Following the conclusion of the final topic, the chairman asked for members of the public present or for those participating remotely to come forward for comment. However, no one came forward for comment.

The meeting adjourned at 1:45pm.