American Enterprise Institute

January 14, 2008

[Edited transcript from audio tapes]

11:30 a.m.	Registration and Lunch
12:00 p.m.	Introduction:	Paul Wolfowitz, AEI
	Keynote:	John J. Danilovich, Millennium Challenge Corporation
12:30	Panelists:	Simeon Djankov, World Bank
		Aart Kraay, World Bank
		Sherri Kraham, Millennium Challenge Corporation
		Guido Schmidt-Traub, United Nations Development Program
	Moderator:	Mauro De Lorenzo, AEI
2:30	Adjournment

Proceedings:

Paul Wolfowitz:  Welcome.  I hope everyone has had a good lunch.  I’m Paul Wolfowitz, a visiting scholar here at AEI, former president of World Bank.  It is my great pleasure to introduce Ambassador John Danilovich as our keynote speaker to open this meeting. 

Just to say a few words of substance about the issues that we are going to be discussing here -- in, I guess, the last 10 or 15 years there has been empirical social science research that has been able to confirm the extraordinary common-sense fact that the effectiveness of foreign aid depends critically on the governance environment in which it is delivered.  It would not seem necessarily that you would have to do empirical research to confirm that when governments are inefficient, aid will be spent poorly; that when governments are corrupt, aid will actually reinforce corruption and help them stay in power. 

Lest we think this is all about criticizing aid, let’s remember that when governance is working, as in, for example, South Korea, one of the most spectacular development success stories probably in history, aid was a positive contributor to building infrastructure and investing in education.

 When societies desire reform, foreign aid can provide critical support and ideas in training and finance.  On the other hand, efforts to buy policy improvements in countries where there is no movement for reform, by contrast, have typically failed.  In countries with sound economic management, foreign aid does not replace private enterprise; in fact, in countries that are committed to reform, aid increases the confidence of the private sector and supports important public services.  On the other hand, in highly-distorted environments, aid crowds out private investment and that helps to explain the small impact of aid in such cases.

 Moreover, simply choosing laudable sectors such as primary health or education cannot ensure in a bad environment that money is going to be well-used.  Aid is financing the entire public sector, and because it is fungible the overall quality of policies and institutions is the key to securing a large return from this finance.  Now, if you think those are simply the views of some former president of the World Bank who is obsessed with issues of governance and anti-corruption, let me assure you they are not.  I was just quoting from an excellent World Bank report called Assessing Aid that was published six or seven years ago.  It is highly worth reading.

 It is easier to write that; it is easier to come to those conclusions than to actually implement them in practice.  Indeed, one of the conclusions that are drawn from this assessment is that aid will be more effective going into countries such as Ghana, for example - they are performing well - than to any number of countries that are performing poorly.  And yet, when I visited Ghana less than a year ago I encountered the term “over-aided” as applied to Ghana.  I said, “What does it mean?  People think they have more aid than they can use?”  “No, no.  It is that they have more than their fair share of aid that it should somehow be distributed around on a per capita basis.”  That really is not fair to the people who can potentially benefit from aid because it means that money that could go to effectively helping people in poor countries that are functioning well may instead go to countries that are not functioning at all.

 The MCC was created with very much this philosophy in mind.  Its mission is to reduce global poverty through the promotion of sustainable economic growth.  It is based on the principle that foreign aid is most effective when it reinforces good governance, economic freedom, and investments in people.  It uses third-party indicators to identify countries with policy environments that will allow this funding to be effective.  These indicators are developed by independent third party institutions that rely on objective publicly-available data.  It is an effort to improve aid agency efficacy for the purpose of catalyzing political and economic reform and promoting effective development.  But actually doing it in practice is a lot tougher than writing about it.

 Our speaker this afternoon has that much more challenging job of making it happen.  He comes with an impressive background as a businessman and private investor with a strong interest in foreign affairs.  He was active in his private sector career in the international shipping business for over two decades, and served as a director of companies in the shipping, property, publishing, and investment fields.  He served on the Board of Directors of the Panama Canal Commission from 1991 to 1996 and chaired the Commission’s transition committee prior to the transfer of the canal to the Panamanians. 

Prior to his becoming CEO for the Millennium Challenge Corporation in November of 2005, John Danilovich served as American ambassador to the Republic of Costa Rica and as American ambassador to Brazil.  He comes with remarkable qualifications for a remarkably challenging job.  I guess that is in your title right there.  Well, please join me in welcoming Ambassador John Danilovich.

 John J. Danilovich:  Thank you very much, Paul.  It is good to see you again and I’m very happy to be here with all of you at the American Enterprise Institute.  It is a good way to start the New Year amongst friends.  I particularly want to acknowledge Mauro De Lorenzo and the AEI team for organizing today’s program.  Thank you all very much. 

I welcome the opportunity to talk to you about MCC’s indicator-based competition for awarding development assistance to reduce poverty through economic growth in our partner countries throughout the world.  The panel, including the MCC’s managing director for development policy Sherri Kraham, will lead what promises to be a very thought-provoking discussion on how indicators are creating a shift from conditionality to competition.  I would like to use this occasion to take stock of the Millennium Challenge Corporation’s pioneering role in this transition. 

But first I want to share with you a story that, for me, sums up the underlying power of awarding development assistance to countries through an objective, competitive, indicator-based selection process.  Last year, I visited Guatemala and had lunch with President Berger and spoke with him about the reforms underway in his country.  Motivated by its interest to qualify for the MCC, Guatemala had taken significant steps to improve its policy performance on our eligibility indicators.  They formed a working group to carefully examine all MCC indicators.  Their technical teams consulted extensively with many of the institutions that supply the indicators that we use.  They also travel through El Salvador to learn more about what that government has done to improve its performance on MCC’s eligibility criteria.

 Based on their findings, the Guatemalan government created an action plan to improve its performance on each and every MCC indicator.  They spotlighted current reforms that they believe directly relate to improve indicator performance scores, such as a $54-million increase in the budget of the Supreme Electoral Tribunal; a thorough updating of the electrical registry; approval of regulations for political party financing; passage of an executive decree on access to public information; and major reductions in the time and cost required to starting a business, registering property, and exporting and importing.  They then outlined additional steps to be taken over the coming years, including stiffer penalties for those who commit crimes against journalists; the creation of an International Commission against Impunity to help investigate and prosecute illegal, clandestine groups that threaten human rights and the rule of law; plans to reform the electoral law; approval of a consultation law for indigenous people, and the implementation of a policy for the protection of human rights defenders.

 With the presidential elections underway, FUNDESA, a Guatemalan civic organization asked each of the leading candidates to publicly signal whether they were willing to do what was necessary to improve Guatemala’s chances for becoming eligible for an MCC grant.  It is a striking validation of MCC’s incentive effect that even presidential candidates now feel compelled to express their commitment to the reforms necessary for MCC assistance.  We hope that these reforms currently underway in Guatemala will bring the country closer to qualifying for an MCC grant.  While Guatemala is a terrific example, it is only one example of how MCC’s methodology is a catalyst for change.

 Through the competition created by our indicator-based selection process, we are changing country behavior for the better, giving reformers the political space and the incentive to implement difficult policy changes.  Such long-term policy changes are what make development sustainable and transformative in the lives of the world’s poor.   Some 15 years ago, however, the indicators we now use to measure policy performance either did not exist or were just emerging. 

When Congress created the Millennium Challenge Corporation in 2004, MCC pushed the envelope by making indicator-based competition the centerpiece of our model and our unique way for choosing partner countries.  We award aid to countries that perform well on a set of 17 indicators taken from non-US government sources such as the World Bank, Freedom House, and Heritage Foundation that measure a country’s commitment to ruling justly, investing in the health and education of their people and promoting economic freedom.  This assessment process sets the MCC apart. 

We remain committed to this approach for the most fundamental of reasons:  It is the best way to identify the best performers best suited for making the best use of US taxpayer dollars.  There is no silver bullet for development, but the closest thing to it is the political will exercised by a country’s leadership.  The right policies are essential for economies to prosper and for the private sector to invest, and this is what ultimately matters.  Private sector investment is what will make economies grow and will break the cycle of poverty. 

Development experts know that private enterprise reduces poverty and increases tax revenues in the long term.  And that is why modernizing infrastructure, fighting corruption, streamlining business development, and expanding market-based agricultural opportunities play such a huge role in MCC’s programs. 

The right policies are also essential for democratic practices to take root.  They create free societies were people demand voice and transparency and accountability in and from their governments.  This serves the interest of these countries and prosperous peaceful democracies are a core principle of America’s foreign policy and security objectives.

The HELP Commission describes how the most effective development assistance supports democratic principles, good governance, country-led development, and economic growth, all of which define MCC’s approach and all of which started with the right policies.  Some say MCC works only with the winners.  Though the countries with whom we partner may have relatively good policies in place, this in no way means that they are no longer poor or about to escape poverty on their own, or do not need help.

Rather, these countries want to loosen poverty’s grip by putting the best policies in place to do so; it is the “by-your-bootstraps mindset” at its best.  Countries themselves freely determine if they are to benefit from our assistance by how they perform on our indicators.  With $5.5 billion now committed in MCC grants to 16 countries, is it not common sense and the best use of taxpayers’ dollars to target our foreign aid to those countries where it is the most likely to do the most good? 

Yet let me clear.  While we are working with developing countries that are the best performers in their peer group with regards to improving their policy environments, serious work remains to be done.  MCC partner countries are taking the right steps, and we are encouraging and strengthening their ongoing reform efforts.

 This is why our work is both rewarding and challenging.  In every instance, making indicator-based allocation operational at MCC requires hard work and taking risks.  On the one hand, the rewards are real.  MCC is enabling a global conversation about policy reform.  All the development assistance in the world will not matter if the policies are not right and in place to effectively use that assistance.  MCC’s period of investment might be limited to the scope of a compact or a threshold program, but the reforms we inspire in these countries take on a life well beyond that.  That is why the emphasis on getting the policies right is so intense and so important. 

Tying MCC assistance to policy performance drives home this point.  That is why we see ambassadors, ministers, and other government officials visiting our offices here to learn more about what indicators their countries need to improve on to qualify for MCC assistance.  Like Guatemala, at least a dozen countries have created interministerial committees and presidential commissions to develop reform strategies to improve policy performance and address MCC eligibility criteria.  That is why we see countries reforming to make their way into the MCC club and to stay in it, in the phenomenon called the “MCC Effect.” 

The Dominican Republic, for one, implemented reforms to improve its investment climate and strengthen anticorruption policies and institutions to qualify for MCC assistance.  Mali and Benin are undertaking significant land reforms, which are necessary to promote secure land rights, access to credit, investments, and increased productivity.  Madagascar reduced the minimum capital requirement for new businesses by 80 percent in 2006 and saw a 26 percent increase in new business registrations.  Similarly, El Salvador reduced the number of days required to start a business from a 115 days to 26 days, which resulted in El Salvador being named as amongst the top-10 performers worldwide in the World Bank’s 2007 World Business Report.

 Steve Radelet at the Center for Global Development characterizes the MCC Effect as the major success story of the MCC.  He says, “A strong MCC effect to pass specified quantitative indicators has created the incentives for potential recipients to more carefully track the data and introduce the policy changes needed to meet the requirements.”  We agree with this assessment, and what we see in our partner countries confirms it.

 We also see the success of our indicator-based approach in our Threshold Program.  The Threshold Program is designed to work with countries to improve their performance on those specific indicators that they are close to passing.  Nearly $400 million has been awarded to threshold programs in 18 countries aimed at improving their policies.  As a result, we see 132 new girl-friendly schools being built in Burkina Faso.  We see Philippine President Arroyo, in an unprecedented move, matching MCC’s $20-million dollar threshold grant to fight corruption with an equal amount of money.  The Threshold Program has given the Philippines renewed vigor in this fight, and anticorruption investigations and dismissals have stepped up significantly.  We have seen the days and costs to start a business drop in Paraguay and Zambia.  We have seen an increase in investigative reporting to expose corruption in Malawi and Tanzania. 

On the other hand, as much as the rewards of allocating aid based on the indicator performance are real, the challenges are just as concrete.  I see this in two ways.  First, the reality of how reforms take place in partner countries is challenging.  The reform process is often complex as partner countries initiate and consolidate reforms.  How do indicators capture the nature of reform?  What do we do when countries drop in performance?  How do we build incentives for ongoing reforms?  We are looking at all of these ways to address these challenges at the MCC, and Sherri Kraham, who will be in the panel later today, will share our thinking with you during that period. 

In an ever changing global environment, I would underscore that our model is also dynamic, and we are committed to making it more relevant and more responsive.  In many partner countries the indicators have become a tool for citizens to hold their own governments accountable, becoming a hot topic for domestic political discourse.  This heightened visibility drives attention to the indicators and motivates sustainable policy change.

 Second, the need for better, most up-to-date data is another challenge.  Because of the increased demand created by the MCC and other donors, we are pushing the development community to provide us with more accurate data.  While this has already happened to a certain extent, we have further to go.  I have an excellent staff; smart, mission-driven people come to work for us at the MCC.  However, their work is contingent upon the work of others.  No matter how exacting our analysis is, we must depend on external sources to supply objective, transparent data on the 17 indicators we use to evaluate some 135 countries.  We rely on the development field - many of you in this room - to assist us by providing us with the best possible data.

 In short, there are challenges to applying an indicator-based selection system.  Data lags make it problematic to assess country performance in real time.  As countries move from one income category to another, the competition intensifies and it is challenging to monitor performance during the transition. 

At MCC, the future is about looking at all these challenges as opportunities to make our indicator-based model for a worthy development assistance better.  We are 100 percent committed to this approach, and we will continue to look at ways to make it even more effective.  We are without question supporting and accelerating the efforts of real reformers.  MCC compacts and threshold countries are signed with government leaders but they reflect a partnership with coalitions, business associations, women’s groups, and all facets of Civil Society in partner countries committed to meaningful reforms.  Our use of indicators elevates the intensity of competition and helps us to make smart investments like any strategic investor in those countries committed to their own policy reform agendas in order to deliver the best possible return on our investment for US tax dollars.

 And, as we learn from other donors who are interested in our approach, we find ways to improve our model.  Some donors are considering using indicators similar to ours to determine which countries might receive their assistance and are putting incentives in place to encourage policy improvement.  We believe this is a good step forward.  We believe the future of indicator-based assistance will come from this fruitful interaction with others in the development community.  Your perspectives bring dynamism to our model and help us think through complex issues.  We welcome your ongoing ideas and input and challenge you to challenge us.

 Can indicator-based competition make foreign aid work?  That is the question we are here to debate.  MCC wants to prove that the answer is a resounding yes.   Our experience has proved that the aid based on indicator performance is working and can motivate policy change.  If non-performing countries look to their neighbors and see substantial resources pouring in as a result of good governance and a genuine commitment to reform, it is compelling them to reevaluate their own lagging performance. 

We are promoting local capacity building, strengthening institutions, and jumpstarting critical thinking about the policies necessary to assure sustainability by insisting that our partner countries design and implement their own development programs.  Our focus on country ownership reinforces the good policies we demand in the first place.  By adapting and applying age-old lessons on the importance of incentives, accountability, and competition, we believe that MCC’s indicator-based approach is an effective and efficient tool for the delivery of US foreign assistance. 

The bedrock of how MCC chooses partners will continue to be an objective, transparent, and competitive process.  With your perspectives, we can further refine our model and lead the way in awarding development assistance based on good performance on policy indicators.  This is what we can best offer our partner countries, and this is what Americans should rightfully expect from their government as the prudent steward and wise investor of American foreign aid.  I want to thank you very much for your interest in the Millennium Challenge Corporation, and I would be very happy to answer any questions, which you may have.  Thank you very much.

 Mauro De Lorenzo:  Thank you.  Yes?

 Ed Corcoran:  You have a big emphasis on -- Ed Corcoran from GlobalSecurity.org.  When we look at economic development historically, nations that have experienced economic development generally have had population increases pretty much wipe it out.  A lot of the societies you deal with are no better off than hunter-gatherer societies 100,000 years ago.  A lot of the policies that you are working on - reducing maternal mortality, infant mortality, the impact of diseases - all of these will also help to drive population growth.  What are we doing to sort of balance that out?

 John J. Danilovich:  Interesting question.  Our countries are actually a little bit more advanced than hunter-gatherers, and many significantly so.  Of course, they are poor; they are lower-income countries; they have an annual per capita income of less than $1750 dollars a year; they are barely making it.  But they are struggling to do so.  They are trying to make the best of it that they possibly can to address the questions which they have with regards to sustainable economic growth and development.

 Our aid program is different than other aid programs because we are into long-term sustainable economic growth and development.  Our programs, whether they be in transportation, agriculture, irrigation or any other programs related to human development, all have an economic rate of return.  We are hoping that our programs will soon show in the short space of four years of our existence -- we are now in a cycle where in 2008, 2009, and 2010 -- our programs will begin to have results in the economies in which they are functioning. 

It remains to be seen whether or not we can anticipate a population explosion in the countries that you have referred to and the problems that you have referred to.  If that does happen to be the case, we and the US Government and the world at large are going to have to address these questions.  I think population control or population concern is one not only in underdeveloped but also in the developing and the developed world as well.  So, this is a sort of universal global problem, not necessarily a problem that is directly related our specific countries.  Yes?

 Vinod Busjeet:  [Off microphone] Vinod Busjeet, Embassy of Mauritius.  You have used the term sustainable development.

 John J. Danilovich:  Yes.

 Vinod Busjeet:  Yeah.  You have used the term sustainable development.  In certain circles, there word “sustainable” means environmentally sustainable.

 John J. Danilovich:  Yes.

 Vinod Busjeet:  Others, it means financially and economically sustainable.

 John J. Danilovich:  Yes.

 Vinod Busjeet:  And President Zoellick of the World Bank has come up with the term “inclusive sustainability.”  Could you explain what MCC means by “sustainable development?”

 John J. Danilovich:  We mean both of those things.  You mentioned the environment, and I think that is interesting.  We have two indicators that specifically address environmental concerns and ecological performance - a natural resources management index and a land access index.  These are new indicators.  They have just been brought on in the past year.  We had a year of transition into these indicators so that our participating countries would be able to adjust their legislative agenda and requirements and laws so as to comply with our environmental requirements with regards to development.

 The sustainability of the environment is a key part to the second part, whichever comes first, of economic development.  We see the two going hand-in-hand; you cannot have economic development which degrades the environment.  And we strongly believe that both of these areas - the growth of the economy and the sustainability of the environment - are related to each other very directly.  That is why we have two indicators amongst our 17 that specifically address the environmental question. So certainly, sustainability is, to use Bob Zoellick’s term, an inclusive matter.  We look at it, certainly, as an economic sustainability as well as environmental sustainability.  Yes?

 Ernest H. Preeg:  Ernie Preeg, Manufacturer’s Alliance.  A quick comment and a question, sir.  In my government career - I did serve as chief economist at USAID in a developing country - I struggled with this good governance all my life and what I want to say is that I am a very strong supporter of MCC --

 John J. Danilovich:  Thank you.

 Ernest H. Preeg:  -- in concept and practice.  My question is about one country close to us - Haiti.

 John J. Danilovich:  Yes.

 Ernest H. Preeg:  I also have some experience; I was American ambassador there.  These days, I’m chairman of the Haiti Democracy Project.  So, I have two questions.  One, as I understand it Haiti is the biggest aid recipient in the hemisphere, and yet they are at the bottom of the pile for indicators, particularly corruption; it is the most corrupt in the hemisphere. 

So my questions are:  Are you having any serious dialogue with the senior levels of the Haitian Government, the Préval Government, two years now, saying the indicators are vital if they are going to succeed and if they are going to get aid?  Secondly, as long as they are still at the bottom and corruption and everything else, should they continue to be the number one aid recipient in the hemisphere?

 John J. Danilovich:  I cannot actually specifically address the fact that they are a large aid recipient.  But with regards to the MCC, having been also our ambassador in Central America and Costa Rica and also in Brazil, I must admit to a soft spot for the Western Hemisphere, and I am very happy with the engagement we have with our three Central American countries, Nicaragua, Honduras, and El Salvador.

 As I mentioned, I traveled to Guatemala and the Dominican Republic last year to encourage them to move forward on the indicator performance as they are, relatively speaking, close to qualifying; they are not there yet and it will take a little bit longer and a greater sustained effort on the part of their governments.  Haiti is just very far away from any type of qualification and certainly is not eligible for MCC funding in the foreseeable future. 

By the way, I would like to point out that we have other threshold programs in the Western Hemisphere that are working well.  In Paraguay, specifically, we have a $35-million program; a very small program in Guyana, and we are just discussing with Peru their threshold proposal as well.  But Haiti is not in MCC and we will not for the foreseeable future be an MCC country.

 Raquel Gomes:  Hi, thanks.  Raquel Gomes from Oxfam America.  I am a big fan of the MCC.  I am just worried about possible effects that can perhaps dampen the MCC Effect, and have two in mind.  One is we observe, or as you know better than I, as more foreign aid has been allocated to other US Government agencies that, perhaps, have more of an interesting short-term security as opposed to long-term development, what does that mean then for the MCC Effect?   And second question - China.  What does that mean for the MCC Effect?

John J. Danilovich:  Interesting.  Our programs, I think, are really well-structured over a five-year period of time.  Our compacts have that duration, and they are set up for that reason, at least at this stage, because we believe the programs that we are implementing - transportation, health, education, agriculture, irrigation - will take that period of time to come to fruition. 

It may well be as this program evolves that we will need longer than a five-year period to achieve those goals.  But in terms of sustainability, we require that a country adheres to our 17 indicators throughout the life of the program.  It is essential for them to do so not only upon initial eligibility but upon signing of our compacts, and it is absolutely critical for them to have, again, been compliant with our indicators at the end of the program so that they can have a subsequent consecutive program.  So we look at our programs, unlike the others, as being one which actually sustains good government performance, good governance and also sustains economic growth.  These policy changes that countries put in place are not simply to address a one or two-year or three or five-year cycle of a specific program, but will be in place not only for the benefit of their own people and for the MCC program, but for the benefit of the sustained development of those countries.  How their programs do it and how they have done it in the past have also achieved certain results.

We believe that our program is actually a tougher program, and it is an accountable program.  It is a program that holds countries responsible and accountable for their development, and we believe the structure which has both the government compliance, the good governance aspect of compliance, as well as, if you can call it, compliance with the implementation, that those two streams will act for the benefit of long-term sustainability.

With regards to China, I think that is a subject which we are all very much concerned about these days.  Certainly, I am, with regards to the countries that we deal in, not so much in Central America and not so much in the Caucasus - we have Georgia and Armenia in the Caucasus - but specifically in Africa where there has been significant Chinese engagement not only on a bilateral basis but also on a multilateral basis.  I’m sure you have all read recently their purchase of a significant shareholding in the largest African bank and their continued ongoing commitment to countries with regards to extraction of various natural resources and minerals.  They have no concern with regards to human rights, with regards to sustainable economic development; they are primarily concerned about feeding the Chinese machine, the Chinese all-consuming capacity for the development of their economy.

I certainly believe that this is for the long-term detriment of Africa, unless they revise their engagement with China; there are indications that they are, indeed, thinking along those lines.  It may well be that the purchase of the significant shareholding in Standard Bank offers, perhaps, a change in their commitment to Africa on a more committed basis. But it is certainly one that I think is -- the Chinese investment throughout the world is one of growing concern to all of us, and it certainly is with regards to what we want to foster in countries, which is good government, good governance, sustained economic growth.  We do not see them as contributing to that process in the countries that we are involved in.  Thank you.

Mauro De Lorenzo:  Ambassador, thank you very much.

John J. Danilovich:  Thanks, Mauro.  Thanks very much.

Mauro De Lorenzo:  This is going to transition us straight away to the panel part of our discussions, so I’ll ask the panelist to come join me here.

Welcome again.  I’m Mauro De Lorenzo.  I’m a resident fellow here at AEI.  I’ll just say a few words of introduction before introducing Aart Kraay, our first panelist, and then I’ll introduce the other panelists just before they speak.  The inspiration for this event comes from the need to examine some concrete ways of moving beyond, of transcending this unproductive dichotomy that has consumed the aid and poverty reduction debate over the past few years between more-aid and no-aid camps. 

It forces aid defenders into a position where they are unable to acknowledge some of the damage aid can cost, and at the same time it forces aid critics into a position where they are unable to talk about some of the ways in which the U.S. government and other donors can actually play a key role and an important role in poverty reduction and the stimulation of enterprise abroad.

It also is in some sense an ill-formed [sounds like] question because, really, the locus of action is the governments of poor countries - what they do and the relationship between them and their own private sectors and legislatures.  It is really less about what we do and or do not do, although that is important.  But the real locus needs to be on political systems and economic frameworks abroad.

It also obscures or makes it hard to have discussions about things we can do, ways we can work to mitigate some of the damage that, traditionally, giving can cause in the economies of recipient countries and, also, some damage it can do to their democratic politics.  You can look particularly at cases where the executive branch of countries receive large aid transfers from abroad would circumvent parliaments, and which also at the same time, reduce the power of those parliaments potentially, and also remove some of the incentives that any government has to raise revenue domestically. 

At the same time, aid is not going to go away.  I think that the new frontier for aid skeptics is to find innovative ways to build some of the key elements of the skeptical critique of foreign aid into the very structure of aid giving itself - ideas about competition, free market-led growth, accountability, responsibility and incentives; ideas which, by the way, are becoming less and less controversial and more and more normalized in aid debates across the political spectrum.  I think it is time to start taking advantage of this new consensus.  I do not know if it is quite at the level of consensus yet but, nevertheless -- that is how we got to the subject of indicators, and the increased prominence of indicators, and increased science of their effectiveness. 

They have always played a role in development.  The World Bank has always used indicators in one way or another as one of many elements in its aid allocation process.  Most donors, in some way, have tried to measure the effects of their interventions.  There have been many surveys, such as the ones from the Cato Institute and the Heritage Foundation, about economic freedom.  There is Freedom House’s annual surveys of political freedom, Transparency International’s corruption indicators, and many World Bank products that measure various key things in international development.

In recent years, these measures have begun, I think, playing a more prominent role with a number of objectives.  One is, for example, to pressure donors to increase aid levels and target their spending more effectively.  The use of them also reacts to some pressures to make sure that aid is effective and not being wasted.

Finally, you have, I think, the most interesting part; maybe it was not the driving factor but seeing that incentives can actually play a role in getting governments to reform their policies and focus more on some of the key determinants of development.  This is, maybe, unexpected but I think it is, perhaps, one of the most significant outcomes and it is one I would like to explore here today.

So the key question here is:  As we have less and less confidence in traditional forms of conditionality to change policy and politics in poor countries, can the competition induced by publicly available indicators - particularly when those indicators are tied to aid programming - can this become a new paradigm for the way in which [indiscernible] try and effect positive change abroad and the way we program our own scarce aid monies?

An enormous amount of the intellectual work surrounding the Millennium Development Goals concerned around designing the indicators that would measure progress towards them.  The main purpose being, I think, to create a powerful tool that could be used to induce donors to meet the rate commitments.  The most apparently stunning use of indicators to change the behavior of governments, I think, is the World Bank’s Doing Business series; the creator of it is here with us today.

One of the most spectacular performers on these indicators is the Republic of Georgia which I had the pleasure of spending some time in last month.  And when you go and see Mr. Bendukidze, who is the state minister in charge of reforms coordination, one of the driving forces behind Georgia’s success, as you enter his office, you are greeted with a wall full of covers of the Doing Business report and news releases related to Doing Business, and Georgia’s success in those indicators.

Many of these efforts came together, I think, with the creation of the MCC in 2003.  We have just heard from Ambassador Danilovich.  Sherri Kraham is here with us who is responsible for much of this work in the MCC.  The 17 indicators are all collected from independent non-U.S. government sources.  There is a paper from 2006 from Harvard which shows that the MCC Effect is real and starting to be seen.  You could ask how you can see this so quickly after two years, but it is something which should be looked at very closely.

Anyhow, the locus of what we are trying to do or what I’m interested in trying to do, and I hope this event highlights, is that we have to move the locus of the development discussion from us to them, from agencies to the governments that are working to improve conditions, and how they can support or thwart their own private sectors, and, also, while at the same time, giving us a powerful and a key analytical tool for deciding how to allocate some of our aid funds.  I’ll turn right away to Aart Kraay from the World Bank, a thought leader there on indicators, in particular, on the World Bank’s governance indicators.

Aart Kraay:  Thank you very much.  It is a pleasure to be here to participate in this panel and discussion.  I had actually prepared some slides, and there are hard copies outside.  But it turns out that I would then be the only person to spoil a perfectly good PowerPoint-free lunch.  So I’ll just make some of the main points that are in those slides, and if people are interested there are those hard copies around and I imagine they will be posted as well.

We also left out front copies of a recent paper that I have done with my long-time collaborator in this area, Danny Kaufmann, on governance indicators in particular.  Although they are only part of the selection criteria for the MCC, they tend to be the most politically charged for obvious reasons and the ones that attract an awful lot of the discussion.  I would like to say a little bit more about that in my comments.

I actually want to try to be a model panelist and, actually, answer the questions that were put to the panel in the announcement of the event.  So the way that I read that, there were three questions.  One was, are indicators a good idea for aid allocation?  The second one is does indicator-based aid create incentives for reform?  Then the third one that I’ll spend most of the time talking about that so far actually has not had a lot of discussion is the issue of what other things should we be measuring?  Or are we measuring the right things?  Should we be measuring things differently?

On the first two points, I can be quite brief and, particularly, since my former boss Paul Wolfowitz has already made the main point about why it is that indicator-based aid allocation is such a good idea, and it is really this very straightforward point:  We can debate endlessly the quality of the empirical evidence but it is simply a question of both recognizing common sense as well as the more technical but also an important point regarding the fungibility of aid.  That as aid donors we should certainly care about the quality of policies in institutions in the countries in which we are operating.

Of course, I do not think anybody in this room can claim ownership of this insight since it has been around for a long time.  In fact, one point that is useful to remind ourselves of is that the idea of indicator-based data has actually been around for an awfully long time.  Something that often gets lost in the shuffle is that the World Bank has been allocating concessional loans from the International Development Association, IDA loans, to low-income countries for over 20 years based on the World Bank’s own indicators of policies in institutions, the Country Policy and Institutional Assessments.  There is an extra weight on the governance components of the CPIA, and there is very big money involved; well over US$100 billion cumulatively over the last 20 years has been allocated using a formula like this.

The cynics in the audience - and I would like to count myself as one of them most of the time - might reasonably ask yourself, “Well, this is well and good.  The World Bank says that it allocates aid based on the quality of policies in institutions.  But do they, really?”  In fact, and this may again come as a surprise to the cynics, the stated -- it is called the Performance-Based Allocation System, which gives a very strong weight to the quality of policies in institutions, in fact, actually, is followed in practice.  So when you look at the historical data, you look at historical CPIA scores that feed into these things, it actually is followed.

So this is something that the Bank has been doing for quite some time.  I think it is also very consistent with this idea that indicator-based aid really is a good idea.  But right away you have to realize that there are a lot of caveats to this.  One is that we should be much more humble, and this actually feeds into the third point I want to make about what kind of things we ought to be measuring.  We should be much more humble about what kind of indicators should be used for aid allocation because one might superficially think that they are sort of a recipe.  There are 17 indicators; you stick them into a hopper, turn the handle, and out comes economic growth.  Of course, we know the world is a much more complicated place than that, and we need to be very careful in recognizing that we are unlikely to hit on exactly the right mix of indicators.

I think all the indicators the MCC uses are eminently sensible on their own.  The CPIA itself for all its worth - some of which I’ll talk about in a minute - is eminently sensible as well, but do we know exactly the recipe from policies to growth?  No, of course, we do not.  Because we do not know that, we also do not know with a great deal of certainty exactly what indicators we should be looking at.

On the second question: Does indicator-based aid create incentives for policy reform?  I want to make that question a little bit separate from the broader question of whether indicator-based aid is a good idea.  I think we should also be a little bit more cautious about the potential incentive effect created by indicator-based aid.  The thought experiment that I thought was useful in thinking about this is that indicator-based aid is being held up as a paradigm shift from old-fashioned conditionality to new ways of indicator-based aid.  But when you think about it, at some level of abstraction the distinction between the two is very straightforward.

On the one hand, conditionality was ex ante indicators:  We want you to do these things, and we will give you aid, and then you do these things.  So we have -- sorry, I’m getting my ex post and my ex ante mixed up.  The aid first and then the policy reform later is one crude way of thinking about conditionality.  In indicator-based aid, you can think crudely as the other way around:  Meet the indicators first and then we will give you aid.

But I think there is also a huge body of common sense and scholarly research that reminds us that the first order of thing in terms of driving policy reforms in the developing world is domestic country politics.  While I think there are arguments for switching the timing of the indicators and the aid, it is hard for me to understand whether this really is the first order of difference, and that we should expect that indicator-based aid is going to have a much more dramatic effect than the old-fashioned, if you like, conditionality of aid.

An important caveat though - and this is one I personally feel particularly obliged to make - is that I think a very important role of indicator-based aid in creating incentives for reforms is that indicator-based aid is a very powerful tool for strengthening the advocacy role of indicators.  The reason why I think personally this is important is because I have been involved for many years in producing the worldwide governance indicators with Danny Kaufmann.  I have noticed a very large increase in the interest that is paid to these governance indicators that we produce following their use by the MCC as part of their selection criteria. 

So on our own, we were doing a certain amount of advocacy work.  Once this advocacy gets tied to real dollars, people pay more attention.  I think that is actually a very powerful channel through which indicator-based aid can help to spark domestic debates over quality of governance and improvements in policies and institutions.

The last thing that I want to talk about, and I’ll go in much less detail than is in the paper and the slides that are out front, is this question of what kind of indicators we ought to be using.  I want to make five points briefly here.  The first is to recognize that different tools are appropriate for different jobs.  Here, there is this old saw that when you hold a hammer, every problem looks like a nail.  That applies very much in the discussions of indicators and aid.  So the main distinction I want to make here is that there is a big difference between the broad cross-country indicators - cover a lot of countries, cover a lot of years, hopefully, as well - that one can reasonably use to allocate aid across countries.

Indicators like the MCC indicators, indicators like the CPIA - these are probably the best tools we are likely to get for this first cut question of which countries should donors be active in.  But these indicators are not by any stretch of the imagination the right sort of indicators that one should be using for ultimately evaluating program success because at the end of the day, the individual projects that aid finance tend to be much more focused, much more compact and have much clearer outcomes of interest associated with them.

So we should be very careful not to pretend that these very broad - and often including our own work - crude cross-country indicators get used to micromanage aid efforts, to micromanage the assessment of whether aid is being effective.  I think that does a great disservice to the huge amount of careful work that many aid donors also do in terms of designing projects in a careful way so that you can do serious and rigorous evaluation of them after the fact.  And there is really no substitute for this country level and project-specific serious evaluation in order to know whether aid is working well.

The second major point that I want to make is that - and this also is in the vein of humility about indicators - I think that we need to recognize wherever possible, and as explicitly as possible, the fact that the indicators that we use are, in 99 percent of cases, really just proxies for the things that we care about.

Simeon’s wonderful work on days to start a business is a fantastic measure of how long the statutory requirements are to start a business.  But there are, of course, many other things that you need to look at if you are interested in assessing the entire regulatory environment.

If you are interested in corruption you might have a survey question that asks of a number of firms something about their experience with corruption in procurement.  That is not a complete indicator for corruption throughout the public sector.  What this means is that whenever you look at a particular indicator, you have to recognize that there are some uncertainties, some imprecisions, some margins of error associated with it.

One of the things that I feel quite strongly about - and this has been reflected in our work on the governance indicators over the years - is that we are the only producers of indicators in the Worldwide Governance Indicators project to explicitly and transparently acknowledge this imprecision.  Whenever we report country estimates for control of corruption or voice an accountability or rule of law, we always accompany them with a range of likely values, and we think that this just captures a reality in the difficulty of measuring governance, measuring other aspects of the policy environment that matter for aid effectiveness.  Just because they are not reported with other indicators does not mean that that imprecision is not there.

The third point I want to make has to do with, in this whole discussion of what types of indicators to use, being really careful to avoid artificial and, I think, often counterproductive distinctions between different types of indicators.  I think it is much more useful to think about different kinds of indicators as being complementary and working in tandem with each other.

I was struck, for example, that both Ambassador Danilovich and Paul Wolfowitz mentioned explicitly the use of objective indicators in allocating aid.  I think that is a distinction that we should be careful not to overemphasize.  I think virtually all of the types of indicators, particularly in the area of governance that we use, involve, inevitably and to the good, some degree of subjective judgment.  Simply knowing whether various laws or rules objectively and factually exist is not enough to know whether those laws are in fact implemented, whether they are enforced on the ground, and so on and so forth.

A telling example of this comes from the excellent work that our friends at Global Integrity here in Washington do on measuring many, many details of the legal environment across countries.  They document, among other things, that in every one of the 42 countries that they assess, bribery is illegal, and in all but three countries, there is a legal right to vote.  Yet, these countries span the gamut from some pretty nasty authoritarian and very corrupt places to the countries like the U.S. which we think of as being at the other end of those scales.  We need to be very careful.  Just because something is narrowly and objectively measured does not necessarily mean it is the right thing to be looking at.

This also feeds into these discussions of actionable indicators; actionable indicators that are within the control of policy makers and the emphasis that is placed on those.  I think we need to be careful not to over exaggerate that simply because not everything that we can measure, not every lever the governments can pull is a useful one.  We need a lot of humility in recognizing that the links from the different types of policy interventions to improvements in governance that ultimately feed into development outcomes are very complicated.

Just two other quick points to make and then I’ll stop and pass on to the rest of the panel.  One other issue to worry about with indicators is to be very, very careful in being aware of the incentives that are faced by the producers of indicators.  I think a very strong point of the MCC is precisely that they got it right in this respect by relying on outside data sources.  I think that is a very strong point of the MCC.

I think, by contrast, we should be much more careful about indicators in which governments who are being assessed play an important role in providing the assessments.  An example of this are the public expenditure and financial accountability indicators that the World Bank and other donors have worked in collaboration with many countries - some 70 countries now - to produce these indicators.  It has been an enormous costly effort.  It has gone on for the last seven or eight years.

The PEFA indicators are touted as one of the very direct nuts-and-bolts, technocratic, specific, actionable indicators.  Then, when you go and look for data on these you realize that of the 70 countries in which these indicators have been constructed, only nine governments have allowed them to see the light of day.  So we need to be careful when the raters, if you like, are too close to the rated.

The final point I want to make - and this also reflects what I think is a bit of a trend in the wrong direction in some circles - is that there is an increasing trend towards constructing ever more refined technocratic indicators of narrow dimensions of bureaucratic performance.  That undoubtedly plays a very valuable role, particularly once you start really engaging countries in the nuts-and-bolts of policy discussion.  Countries do not want to know generalities.  Survey says corruption is high.  They want to know, “Where do we go from here?”  But there is a danger in going too far down this technocratic road that we miss incipient and very large political democratic accountability risks that are very present in countries.

So one of the graphs you will see in the handout is a graph that contrasts the very technocratic CPIA assessments of the World Bank with one of the Worldwide Governance Indicators’ least politically popular indicators measuring political stability and absence of violence.  There are some non-randomly selected countries highlighted on that graph, including Kenya, Bangladesh and Pakistan where we know these countries have -- while they are performing reasonably well on narrow technocratic dimensions of aid effectiveness, if you like, have also had enormous challenges in other dimensions.  So let me stop there and turn over to the next.

Mauro De Lorenzo:  Thank you very much.  We are going to turn now to Guido Schmidt-Traub, a team leader at the Millennium Development Goal support team in UNDP.  He previously was an associate director at the UN Millennium Project directed by Jeffrey Sachs and has been active on these issues for a number of years, I think almost since the creation of this project.  Thank you very much for joining us today.

Guido Schmidt-Traub:  Thank you very much.  Thanks to AEI for inviting the United Nations Development Program to participate in this very important debate.  The problem in front of us, broadly speaking, is of course, a very, very tough one.  Evidence, I think, is clear - and I think there is a broad consensus now - that many of the poorest countries need international assistance in order to achieve development no matter how you define it - measured narrowly by economic growth or looking at broader socioeconomic data.  There is a need for technical support, and in many countries there is a short-fall of investments.  Some of the poorest countries, particularly in Africa, need to invest more in health, education and infrastructure to empower their people so that they can be productive, raise per capita incomes, and become full parts [sounds like] of the global economy.

The problem in front of us is that these investments cannot be financed out of domestic resources alone in some of these poorest countries.  It does not apply to many countries that receive aid today.  Many countries today receive aid even though they do not need it from a narrow development perspective.  But certainly, a lot of countries do and I think there is a fairly broad consensus around that.

This requires us to think about aid, whether we like it or not.  If there was a way to do this through private market forces or loan I’ll be the first one to advocate it.  But I think that is widely seen as, on its own, impossible.

This means that we need to think through how to structure an incentive-compatible partnership between donors and recipient.  The partnership, of course, has to be performance-based.  There has to be clear accountability, and it has to address the needs of all countries from the well-governed but poor countries to the very poorly-governed countries.  What is it that we can do to support development in those places?  This in itself, of course, is a much more complex challenge.  Of course, indicators play a crucial role in that.

I want to identify four dimensions quickly in which I believe indicators can substantially help this process of structuring this partnership and then flag some of the caveats and some of the issues that have come up.  I will complement the presentation from the previous speaker by focusing a little bit more on the Millennium Development Goals, which do not only include quantitative measures for governance but focus on the development outcomes - health, education, environmental sustainability, gender equality and so forth. 

The first dimension is that these indicators, as simple as it may sound, clarify the objectives and actually raise issues to the political level that otherwise might not get there.  One of my favorite examples is I have been in meetings with heads of states of several Latin American countries that were quite stunned to see how shockingly high their maternal mortality ratios were despite relative high levels of incomes.  And they looked at graphs showing where they were relative to other countries of a similar level of income.  This was the first time, I would guess, that some of these heads of states had really thought about the issue of maternal mortality in their countries.

That was a direct outcome of the fact that this is one of the key indicators of the Millennium Development Goals.  That in and of itself, as simple as it may seem, is a key impact that these indicators can have.  And, of course, I think the business indicators that we will hear more of have been hugely influential in raising some of the very practical issues.

It also underscores the depth of the challenge in front of us, and I think it is very different looking at a middle income country that needs to do a little bit better in extending social service provision, let’s say, in health or education, compared to an African country where sometimes you have as little as 40 percent of school-age primary schoolchildren even getting one day in class, let alone finishing their curriculum.  These are qualitatively and quantitatively very different challenges and require a different level and a different type of engagement from the international community.  That is, again, what these indicators map out.  So that is the first issue of clarifying objective and really bringing out the outcome data that needs to be covered.

The second is benchmarking country progress.  And that is something that, in my assessment, we actually do not do enough of.  There is a plethora of and a very welcome explosion in indicators, particularly, cross-country indicators, that can be compared.  But there is not really enough, in my assessment, of some careful benchmarking on what are some of the success stories that have come out?

Malawi has successfully doubled agricultural yield over the past two years.  This is not widely known across all of Africa.  It needs to be because there are important lessons, good and bad, to be learned from that.  I’m focusing on Africa because that is, of course, the constant [sounds like] of where [sounds like] a development crisis is most severe and where the challenges are deepest.  Many African countries have successfully increased education enrollment rates.  Again, that is not actually widely known, not widely enough in the U.S. and many of the Western countries, I would say.  So that is the second broad benefit.

Now, as we are getting into some of the more operational ones, there are two important issues I want to flag.  That is of course, the use of indicators as an accountability framework and then as a mechanism for allocating resources.  On the accountability framework the key question here is to what extent does taxpayers’ money achieve the intended results?  So to what extent does a dollar invested in the education system bring additional children into school, increase the quality or improve the quality of the education and, overall, add to the human capital base of the country, not to the extent to which that dollar invested in education affects a privatization policy in a given country?  These are important issues but they really are quite distinct.  I think we need to be clear about the instrument that we are talking about, which is development and the outcome that it can target, namely, development outcomes.

There is one very obvious problem here, namely, that the link between the inputs and the outcomes that we care about and the countries care about - the economic growth, child survival rates, maternal mortality ratios, and so - there are a few important intermission [sounds like] chains [sounds like] in that step.  One, of course, is an important lag.  Maternal mortality ratios shift only with a lag of several years, so that is actually quite a poor indicator to measure on an annual budget cycle the effectiveness of programs.

Second, we are dealing with a lot of threshold effects that in my assessment is often not appropriately appreciated.  Half a road is not as good as a full road; it is useless because you cannot use it.  Likewise, half a school is not as good as a full school.  Very often, we are dealing with situations where marginal increases in resources will not produce significant results because the investment levels are so far below the thresholds that need to be met in order to achieve results.  That makes it quite difficult to track the impact of a marginal dollar.

The third major issue is, of course, that we are dealing with too many mappings.  To reduce child mortality, you need to invest in a health system; you need to invest in mother’s education, in infrastructure, water and sanitation and so forth.  Then it becomes quite difficult to tease out the impact of each individual one.   What this means is that these broader outcomes - child mortality rates, economic growth and so forth - are not very well-suited to track the effectiveness of development programs and the effectiveness of resource used in an acceptable timeframe, and I would argue that the same, of course, applies to governance indicators.

Unfortunately, here, we have in recent years had a tremendous increase in intermediate input indicators, like the proportion of mothers that have access to emergency obstetrical care, which is the key intervention to reduce maternal mortality.  That is now more widely available.  Those are the indicators that need to be tracked.

My sense is that there are now a large number of programs out - and the MCC is one of the most important ones - that really do this very well.  It demonstrates how a tight accountability framework can be designed, and how it can be used to keep that program on track and to make this partnership between donors and recipients possible.  It is not easy.  It is very hard; we are working in some of the most difficult poorest places in the world.  But it can be done.  I think the success of the World Bank and many other donors underscores this amply.

The final point on the resource allocation framework -- of course, it is absolutely unambiguous that good policies produce better development outcomes, and that aid is more used effectively when policies are good.  But in thinking about using indicators as a basis for allocating resources, there are a couple of, I think, important issues to underscore.

One is, “What to do about the poorest and most desperate places?”  It is one thing to ask yourself how to most effectively assign the marginal or the incremental dollar in tax revenues.  Then, of course, a purely competitive program like the MCC, I think, plays a very important role in that.  That is very different from asking the question of we as an international community, how can we do everything that we can to support even the most precarious countries in the most difficult circumstances, where often governance is very, very poor?  Those are countries that are far from qualifying for the MCC countries.  Haiti was mentioned earlier on.  That is certainly a good example.  It does not mean that the international community should not do anything to help in Haiti.  The question is:  How?  So that is the first caveat.

The second caveat - and a lot of that was already said about the governance indicators that are used as a lot of cut-off criteria - is that they, of course, measure outcomes and they are often quite distinct from the commitment of a government to actually institute the necessary reforms.  In a poor country, the effectiveness of institution transparency will always be lower than in a richer country because you have less resources for an independent media, for a vibrant Civil Society, for the checks and balances that cost money that require staff who are well paid, adequately paid, in order not to fall into the temptation of corruption.

So there really is a two-way relationship between governance and economic development.  It cuts both ways.  That is something that is really important to bear in mind, so ceteris paribus, the poorest countries will have less good development outcomes across the vast majority of indicators, and we need to take this into account.  Steve Radelet wrote a very interesting paper on that a couple of years ago that I would be happy to share with people.

Therefore, it is actually difficult to justify using these governance indicators, many of which are perception based, as a hard cut-off criterion where moving up or down a ranking of several dozens of countries but one or two places makes a difference between qualifying and not qualifying for the MCC as one example.

Again, from a perspective of allocating the money the most effective way, that may be a good way to proceed.  But if we look at the broader challenge at hand, namely, how can the benefits of private enterprise, of development of economic growth spread to all parts of the world.  This is, of course, only looking at an artificially narrow subset of countries and some of the countries that have not qualified for the MCC could just as well have qualified.

A final point is that governance and sound policies are vital the more complex a development program becomes.  If a country is building a health system, that is one of the most complex challenges around, both technically and logistically from an accountability and fiduciary standpoint.  Therefore, you need effective systems to make this possible.  However, to use as an example, there are certain important health interventions which can be administered even in countries with very poor governance ratings, even in conflict zones.  Most famously, UNICEF and others and the U.S. government for many years have with great success administered vaccination programs, malaria control programs even in war zones.  These programs can be delivered even outside governments, even in cases where governance is not very good.  And they can often play a substantial role in building the human capital that is required to ultimately achieve higher economic growth rates.

So in my sense, it is really important to look beyond the relatively narrow set of countries that stand a realistic chance of qualifying for the MCC.  I think the MCC is a really important part of the range of instruments that we have.  I would argue that some of the criteria are too defined and too strict and too narrow a sense.  I think they could be broadened to allow more countries to qualify.  But that will never change the fundamental fact that some of the most difficult countries would just never qualify for this kind of a program.  Yet, I think a lot can and needs to be done to advance development in those countries.  Thank you.

Mauro De Lorenzo:  Thank you very much, Guido.  Sherri Kraham is the managing director for Development Policy at the MCC.  In that role, she is responsible for managing the selection process every year, or the evaluation process that leads to selection.  Before joining MCC she worked at the State Department for a number of years, including a tour working on reconstruction efforts in Iraq.  Sherri, thank you very much for joining us today.

Sherri Kraham:  Well, thank you for this opportunity to talk about MCC’s indicator-based approach and how we are applying it.  MCC is an open and transparent organization, so I will not just tell you about how we are applying it but, also, some of the challenges that we are facing in the hopes that you can help us improve it and think through these issues.  I’ll briefly address three issues:  Introducing MCC’s competitive process; talk about the ways the indicator-based process is helping us achieve our two objectives in the selection process,  and outline some of the challenges that we are facing.

First, the MCC is a partnership.  It is a competition and it is preconditioned.  It was created taking into account many of the lessons learned from policy conditionality.  Our principle that it is based on is policies matter; that having a sound policy environment in place will create the greatest opportunity for economic growth and poverty reduction to take place.  It is not the only instrument of the U.S. foreign assistance family; USAID continues to do their very valid work. MCC is narrowly focused on those countries where we are going to have the greatest impact on growth by identifying them through this selection process.  So that is the first objective is to select the places that MCC is going to have the greatest impact in terms of economic growth and poverty reduction.

The second goal of our selection process is to reward countries that have put these good sound policies in place and, thereby, incentivizing others to adopt similar policies.  In terms of our competition, the mechanics are not that important but I’ll just briefly mention a few things.  We use 17 indicators - that has been said - in three policy areas.  We look at 108 of the poorest countries in the world.  They compete against each other in two income categories, and we measure against the policies that we in the development community have identified are most critical and most linked to economic growth.

We use independent third party and to the greatest degree possible, objective indicators.  For each indicator, countries are ranked from best to worst and then we draw a median in the middle; if you perform above the median, you meet the criteria for that indicator; performing below, you do not.  So for countries that perform above the median on half of the indicators in each category and on the corruption indicator, they technically meet our criteria and can be eligible for several hundred million dollars of grant assistance.

It is a simple rule-driven competition.  We use publicly available data.  It is all available on the web, and I hope you all have seen our scorecards.  If not, they can be found at mcc.gov, and there is a big potential reward.  The rules are clear to the countries and not just to the governments in these countries, but to the people in the countries.  Because of that, we have seen this process and these scorecards mobilize domestic actors in a very significant way to hold their governments accountable.

It is not a system of conditionality, and we have tried to avoid that.  We do not want to put conditions in place on our aid to buy the reforms that we are seeking to stimulate.  We see this more as a screening device.  So it helps us identify the countries that are most likely to use our aid well and countries that should be recognized and rewarded for doing the right things.  But piloting a new approach creates challenges, so I’ll get to that in a few minutes.

In terms of how we are already achieving the first two objectives, the identification of good performers, stepping back from the mechanics, the key concept that we try to remember is that this system was designed to distinguish between countries that perform well and those that do not.  It really for the most part does a pretty good job at comparing a set of countries with different environments against a set of comparable indicators, and then helping us identify the best performers.

Using the indicators help to discipline our assessments, and it largely removes the kind of political bias that undermines the success of many donors by forcing them to work in places that are not committed to sound development and forcing them to take on projects that have a little chance of success. The system helps us objectively assess who is most likely to succeed in a poverty reduction and an economic growth program, and who should be rewarded.  I will not touch too deeply on do we have the right measures and are we measuring the right things, right policies.  We look at this on a continuous basis, on an ongoing basis, and we balance between our indicators in terms of long-term institutional indicators that measure institutional performance over time and highly actionable indicators.  So we have always sought to maintain a balance of things that the countries can do to address performance, both in the short-term but that measure also the depth of a country’s commitment.   So with our indicator framework as the guide but using good analysis and good judgment, the board makes the most informed decisions about country performance. 

The second mandate that the indicator system is helping us fulfill is creating incentives.  We already talked a little bit about the MCC Effect.  I see this as the impact that we are having on country behavior without spending money.  The clearest evidence of the MCC Effect is how countries respond to our scorecards, and this transparent system helps us to create incentives for reforms in a number of ways.  I think this is really the most exciting part of our process.  First, the governments are using the scorecards.  They give governments a sense of where they stand in relation to their peers, to their competitors, in fact.  It helps us foster a sense of competition.  It is striking to hear the heads of state and other ministers refer to the scorecards, ask questions about the scorecards, and we think that these indicators have their attention.

So for example, last week, I had a video conference with the government of Madagascar where three government ministers sat on the other end of the video conference talking about their scorecards and ways that they continue reform across the indicators.  So this has really opened up an interesting dialogue, and we see that we have their attention.  Countries are increasingly using this framework to assess, track and monitor their own performance. 

The Ambassador mentioned the El Salvador example where they have set up this technical committee in the presidential office.  I think we have all interacted with this group.  They have exported their model to other countries that are similarly trying to track performance by taking apart the indicators, dissecting and diagnosing their performance, and setting up various committees to implement and track reform.

Second, the transparency of the process allows for domestic actors to mobilize outside the government to hold their governments accountable for good governance.  We have seen it empower domestic reformers inside and outside of the government. In addition to the powerful incentive of having a significant amount of grant funding, MCC’s seal of approval seems to be something that countries and governments value.  So we have seen Civil Society groups organize themselves, and the press use scorecards to force their government to reform.  As the Ambassador said, we are seeing some governments bring the MCC scorecard into presidential debates.

So I brought a few props, and I think these are really interesting and illustrative examples of -- a few papers, and we have many examples.  This is Guatemala.  This is El Salvador where they are reprinting the data in the newspapers.  I think this is a fantastic example.  This is an example that I have hanging on my wall; it is called an F for Freedom House, and this is where an Armenian newspaper is disputing the assessment that Freedom House gives it on democracy and governance, and goes through the process of analyzing why they think the assessment is wrong.  I just think this is a great example of how actors are engaging.  So I’m going to send this around if anybody is interested in reading them.

Third, the MCC and other parts of the U.S. government have had an expanded dialogue about reform, and this is something that we personally experience but that we are also hearing from our colleagues in terms of USAID and the embassies out in the field.  In my position, I talk to representatives of candidate country governments, and over the years I have spoken to about two-thirds of the hundred or so countries in our candidate list. And we know that these conversations are happening out in the field with our diplomatic posts abroad.

When countries come in for the initial conversation, we generally go through this cycle.  First, they say, “You should not be doing this.  You should not be assessing us in this way.”  Next, they challenge the assessments and say the rankings are wrong.  Then, finally, they get to the point, and we hope they get to the point where they say, “What can we do about this, and what reforms can we take?  What are the opportunities for us?”  That is really where we want to get.

Sometimes, this cycle takes place in the course of a meeting; sometimes over many months, and sometimes, in some cases, it is over many years.  But this is the type of engagement we are most interested in.  Then, moving to the countries’ dialogue with the indicator institutions, I would like to think that we have seen an expansion of that dialogue.  This is one thing that we strongly encourage.  We see ourselves as a facilitator, so when the countries come in we actually refer them back to our colleagues for a more fruitful and technical dialogue about what we can do.  So we have heard this from our colleagues here.  We have also heard this from Freedom House, the Heritage Foundation, WHO, UNESCO. 

So countries are contacting the indicator institutions to learn about how they are being assessed, to ask that they be measured so that country coverage can be expanded, to report data to a greater degree, to review assessments and to bring discrepancies to light, improving data quality.  And they also ask for reform advice.

While the governments often challenge the assessments, especially in the democracy and governance area, I see this as an incredibly useful part of the dialogue.  Freedom House regularly reports countries like Azerbaijan and Cameroon coming in to dispute and improve information about their assessments.  I think this is extremely valuable.

 So now, to the challenges.  The first challenge is the data and how we use them to make decisions.  We are taking a significant departure for the U.S. government from the way that we typically make decisions by putting our trust in data and there are limitations to the data.  This whole approach is new to many people and not everybody is entirely comfortable with this way of making decision.  So there is this tension between experience and data.  One of the points that we constantly hear is that country X could not possibly pass the corruption indicator because I know that corruption in country X is so high.

 One thing to point out is that while there are a lot of experts and expertise among our stakeholders and in the U.S. government, we are looking at relative performance.  So a country meeting our corruption indicator does not mean there is no corruption in the country or that there is extremely low corruption in the country but that it is a relatively better performer than the other countries in that income grouping.  So we never say that there is no corruption; certainly, in our dialogue with our compact partners we certainly continue to talk about ongoing reforms.

 Some of the concerns about this experience versus data challenge is particularly important given the fact that the data are not perfect and that, of course, the indicators and the system itself have some limitations.  The indicators are the primary means that we use to evaluate country performance, but we do not take them at face value.  It is not as simple as red or green, or pass or fail, or 49 and 51 percent.  We analyze the data so that we can understand it better, that we can review the accuracy, we can take into account any shortcomings and any data lags and provide the board with supplemental information so that they have the most robust picture of a country’s actual performance.  So we provide them with information to complement the scores.

 The other point that was raised is that the indicators do not address all of the purposes that we need to use them for.  They are best suited for measuring the good performers and distinguishing them from the low performers.  But since we are using them to make critical policy decisions, I’m going to speak a minute about some of the things that we find particularly challenging.

I would also like to give some perspective on how dramatically this field is evolving.  Five indicators that we used from the World Bank Institute did not even exist 10 years ago.  The IFC Doing Business indicators - we are using four of them - were created in 2004, I believe.  So we have seen impressive impact in the field in just the last few years, and we are going to continue to work with the indicator institutions to strengthen the data.

 The second challenge about the indicators is that they are not suited for all purposes, and I mentioned this just a moment ago, particularly, not for making the fine distinctions in the middle-performing grouping.  So that is where we dig a little bit deeper and make ourselves more knowledgeable about what is actually happening in the countries.  And they are not as useful for short-term monitoring and for monitoring the performance over time.  There are data lags.  There are countries with large margins of error, so that they constrain us in some ways from being able to follow progress as much as we would like.

One of the challenges where we will spend some time this year - the Ambassador mentioned - is to create stronger incentives for countries once they are already in the programs.  So how can we build the incentives back into the system when a country already has received a compact?  There is an enormous incentive for a country to seek to qualify for MCC in terms of hundreds of millions of dollars of grant assistance.  So we are really thinking deeply about how we can maintain those incentives to keep them encouraged to be doing the right thing, even after they are working on their compact program.

 One of the lessons from our experience in policy conditionality is that each country has its own unique set of circumstances and complicated political economy.  So we are trying to refrain from falling into this policy conditionality trap where we are micromanaging reform by identifying all of the things that a country needs to do.  We largely leave it to the countries to identify their reform path, and it is up to them to implement them and stay in the MCC program.  We largely try to focus on the positive incentive effect and try to avoid punitive actions.  And so I welcome your input and your ideas on either financial incentives or recognition incentives so that we can provide further rewards, perhaps, like the IFC’s Doing Business recognition program for exceptional performers. 

Just to close, I will tell you how we are responding to some of these challenges.  In terms of improving the data, we are working very closely with the indicator institutions.  Because we and other donors are using these indicators, we have seen improvements in the data in the last four years: improved accuracy, frequency of collection, expanded country coverage - IFC alone has added 50 countries since the beginning - improved methodology, more comprehensive measures, increased transparency and availability.  We have even seen the creation of new indicators where we have identified a policy gap.

So the data quality issues are really important.  But I think what we are trying to do is not to lose sight of some of the data issues and lose sight of the big picture, which is our objective - to have that dialogue about policy reform and to work with countries to improve their performance.  So I think if we continue to focus on that goal, we can put some of the data challenges in proper perspective.  So we welcome your ideas on how to strengthen the system and the creation of better tools for measuring performance and ideas on how to increase these incentives within our programs.

Mauro De Lorenzo:  Thank you very much.  We will go straight to Simeon Djankov who is the chief economist in the indicators group at the World Bank and the creator of the Doing Business series, which many have made reference to already; perhaps one of the most striking cases of an indicator having a real effect in policy making in poor countries.

 Simeon Djankov:  Thank you.  Since I’m going last, I decided to pick three questions [indiscernible] respond to the initial questions but by now, I think, well-answered.  The questions that I think both our colleagues at the MCC are probably addressing often with governments and other parties as well as my team -- the Doing Business team in the IFC.  And they all relate to the use of indicators for this type of, essentially, policy work, but also advocacy work as both Sherri and Aart mentioned.

So the first question is by focusing on a set of indicators, is this too simplistic?  So is this both in terms of which indicators you choose?  But just as a policy, let us say for the MCC, is it too simplistic to say, “Well, we just focus on 17 indicators and this is how we make the selections?”  And Aart mentioned at the beginning that actually there is some practice in the World Bank for about 20 years of doing something similar, although it was not, certainly, as transparent of doing this.  Actually, overall, in policy development there are many other examples - very successful examples - of how similar, let us say, indicator-driven approaches have actually worked. 

So one example is in the area of trade policy.  If you think of the predecessor of the World Trade Organization, the GATT, it was called the General Agreement for Tariffs and Trade.  For the first 30 years, basically, the only focus of that organization was reducing tariffs.  And if you think of the various rounds - the Tokyo Round, Uruguay Round - before it was all, basically, focusing in one indicator.  So if you think in that term - in terms you would say, “Well, the MCC is actually a very sort of renaissance type of organization because we are focusing on 17 and the WTO focused only on one.” 

Of course, since then, the WTO has started focusing on a few more things but not that many.  They probably, if you push them, would have three indicators or maybe four indicators.  And yet I think in the last 40-50 years, most people would say that in the area of trade policy the world has achieved a lot.  So it is a much easier environment now to trade than it was 40, 30, 20 years ago. 

And the other area that, perhaps, implicitly, but has been driven by very few indicators, is in the area of macroeconomic stability.  So if you think of the IMF, they essentially, have two or three indicators that define all 2000 people who are working at the IMF.  So they basically focus on inflation, budget deficits, exchange rates from time to time, and that is about it.  There is not really much more that people focus in macro-policy and if you again go back in time, you will see that by and large the world has become a much better place now than it was even 20 years or 30 years ago in terms of macro-stability and macro-policy.

So the point is that if you think of examples outside of the area that we are discussing now, there are actually some remarkably successful examples of development policy trade macro where -- either explicitly in the case of the GATT on tariffs or implicitly in the case of the IMF and Central Banks; there has been focused on a very specific small set of indicators and it has been very successful.

The second question which comes more and more is, “Okay, so now you are using a set of indicators.  We are reasonably convinced that having some measurement is a good idea, but you missing some very important things.”  So at the extreme it would say, well, by focusing on the days and cost of starting a business, let us say, which my team produces and MCC uses, you are missing these extremely important other areas.  For example,  let us pick one.  So how the government or what are the government policies with regard to foreign labor, let us say.  And it is certainly an important policy to have. 

Actually, this reminds me of a fairly famous economics paper going about 30 years ago; Paul Milgrom had a paper that did not study but actually hypothesized what a good educational system would look like.  Basically, in his model - he is a theory economist - he said “The educational system is focusing on math and manners.  Mathematical skills are easier to measure while manners are bit more difficult to measure.”  So basically, his model - I do not know why you need a model for this - predicted that if this is the system, you would end up with a bunch of rude geeks, basically, because you would be measuring Math very well and they will be expanding on that.  But manners, perhaps, they would not be that skilled at. 

And of course, in various comments, at least the Doing Business team and maybe MCC must be receiving a lot of this.   Why are you not including this other area?  It is equally important.  Or this other area?  You are too focused.  And there are two points for that.  One is for some things, if for example, the focus is economic growth, maybe Math is actually more important than manners, so that the policy maker - in this case, the U.S. government or the MCC management - can rationally decide some things; math more than others based on the academics evidence, research and so on in, let us say, economic growth.  So you can imagine Microsoft in the early years probably being a bunch of relatively rude geeks, but they certainly contributed a lot to economic growth both in Seattle and all over the world.

But the second point is that, of course, manners can be measured and I started actually my undergraduate education at the University of Vienna.  There, we actually had a course in manners.  We had very little Math, which is why I left, but we were taught how to eat properly and how to actually go to the casino even and so on.  So what we are experiencing here in some sense is the early years of basically measuring the manners of government.  Governance, measures of government, the regulatory measures of government, and so on. 

And as Sherri mentioned, some of this is very young, so the Doing Business team -- our project is about six years old.  The governance indicators have been around for eight, maybe 10 years; Heritage Foundation about a dozen years and so on.  Compare this with the macro-indicators that have been around for 50-60 years and you get an idea of how young our field is.  But it can be measured and that is what we are doing.

And the final question which relates to this past discussion is is it too early then to use it for policy advice since you have just told us that the Doing Business indicators are only six years old.  Is it too early to, basically, depend on them in advising governments or suggesting to governments policy reforms? 

My answer to that is, no, it is not too early.  One of the things that at least we have seen is that just by focusing on it, the indicators themselves are developing much faster and there is much more quality in the indicator.  So for me, this is the MCC Effect, that there were a few concepts of indicators around and some individual advocacy or research work around governance, around regulation and some other areas. 

Having them in the MCC suddenly has drawn the attention of everybody, basically, to them.  Governments are trying to improve.  The opponents of this kind of agenda are trying to find out errors in them, and this actually increased significantly the quality of what we currently measure.  As Sherri mentioned, it also basically brings us to some things that are not measured yet, some of these missing manners, so to speak, that indicators are starting to actually come up in these manners but probably will take much longer time if MCC was not focusing on them. 

And I think this initiative movement, really, is a very recent movement.  So it is only in the last three-four years, it really coincides and is driven primarily by MCC of suddenly having the set of indicators, having a discussion like what we are having today on how to improve them or do they serve and so on, and engaging the governments in reform.

And I will finish just going back to macro and trade indicators with one very interesting piece of statistics on macro indicators that slipped just around Christmas time:  There was a measurement done by the World Bank, a revisit of the price indicators of a number of developing countries, China and India being the most important ones.  So basically there was a lot of work over a year and a half to see what are the right prices, so to speak, in China.  And most of you, probably, have not seen this, have not heard about that, but it turned out that, actually, pricing in both India and China, it turns out, was miscalculated by about 40 percent. 

So it turns out that China now as opposed to a month ago is actually about 40 percent smaller than we thought because we where thinking of this as output, output, output, growth; in fact, prices were increasing quite significantly as well, and the same thing happening in India.  So if you look at the rankings of countries in terms of global leaders, you know, China just dropped from second to third.  So Japan just suddenly surpassed it in this period. 

The point of this example is that there are going to be mistakes and even in price or in this case, inflation indicators, which have been around for 50-60 years, you find mistakes, even huge mistakes for countries like India and China.  But that has not raised the issue of, “Okay, well, then measuring inflation must be a very, very stupid idea.  Let us not do it if we cannot measure China and India.” 

The basis for discussion having a low inflation is a good idea and we should continue with this measurement; it has not shifted.  I think, maybe in another five years we will get in our type of measurement to the level where we would not be having the question, is this too simplistic or not.  Does it detract from other things or not?  Will it have much more the type of conversation of how much does it matter?  So what are the real effects of these kinds of reforms and how can we, the MCC, but the development community overall -- how can we incentivize even for the effects?  Thank you.

Mauro De Lorenzo:  Thank you very, Simeon.  We have a good amount of time for discussion and I think I would like to open it by asking each of you to pick the one indicator - key indicator - which is missing, because it is either too hard or because it has been overlooked for another reason.  What is the one indicator that is missing?  And then we will open it and have a discussion around them.

Simeon Djankov:  Starting in reverse since you are looking at me, I think the one area where a significant high-quality indicator is missing is infrastructure.  And I mentioned infrastructure because a lot of development assistance has been going and will be going even further, I predict, in infrastructure, particularly in Africa but also beyond that.  And the indicators that we currently have are what was mentioned; sort of, you know, kilometers of paved roads.  But whether the road leads anywhere and what sort of pavement is there and is it going to going to be ruined two months after --so they are both outcome oriented but, sort of, grossly, I think under -- or misrepresent what it is. 

Sherri Kraham:  Well, MCC is currently looking for a measure of educational quality.  Well, there are indicators that are used to measure program effectiveness.  We have yet to find a sufficient or adequate indicator that measures cross country over the hundred or so countries that we look at in terms of what are governments achieving in terms of learning outcomes for the investments that we are making.  We are actively searching for this and we know that there are few efforts underway at the World Bank and other places and we strongly encourage them to bring those issues closer to fruition.

Guido Schmidt-Traub:  Thanks.  This work becomes easier the later you speak in the series.  Now, I would fully support the call for better infrastructure indicators.  Transport is woefully under-measured.  We need better -- World Bank has started this but the data coverage is poor of effective access to motorist [sounds like] transport.  And this is also where the incentive effect between those who collect the data and those who use it is not fully resolved.  If I can add one indicator of my own to this, I think in the area of aid effectiveness most of the discussions are too broad.  The link between aid