Hating on the MCC's Corruption Hurdle - Casey Dunning, Charles Kenny, and Alicia Phillips Mandaville (MCC)

Summary

---

Podcast Summary: "Hating on the MCC’s Corruption Hurdle"

Podcast: The CGD Podcast
Host: Center for Global Development (Lawrence MacDonald)
Guests:

• Alicia Phillips Mandaville (Managing Director of Development Policy, Millennium Challenge Corporation—MCC)
• Charles Kenny (Senior Fellow, Center for Global Development—CGD)
• Casey Dunning (Policy Associate, CGD)
  Date: April 14, 2014
  Topic: Evaluating and debating MCC's use of a "hard hurdle" on corruption for foreign assistance eligibility

---

Episode Overview

This episode explores the Millennium Challenge Corporation’s (MCC) approach to combating corruption through its "hard hurdle" on corruption: a key requirement for country eligibility to receive US development aid. The discussion, prompted by a critical CGD paper titled "Hating on the Hurdle," dives deep into the strengths, shortcomings, and potential reforms of the current methodology, and what smarter, more actionable policy could look like. Alicia Phillips Mandaville offers the MCC’s perspective and response, while Charles Kenny and Casey Dunning articulate their critiques and suggested reforms.

---

Key Discussion Points & Insights

1. What is the Millennium Challenge Corporation (MCC)?

• Mandate:
  • Independent US foreign aid agency established to experiment with data-driven, transparent, and accountable foreign assistance investments.
  • Focuses on three core ideas: country policy assessment using third-party data, fostering country ownership in investments, and enforcing greater accountability through measurement and transparency (01:53).
• Background:
  • Born out of desire to move away from politically motivated aid allocations (e.g., “money to our friends, not our enemies”) and toward supporting countries with sound governance (02:54).
  • Has enjoyed bipartisan support in Congress for being "rigorous" and "sound" (03:40).
  • Relatively small in size (300 staff, $900M budget, working with 25 countries as of 2012) but has an outsized influence due to its unique approach (05:33).

2. The Corruption Hard Hurdle: What Is It?

• MCC’s eligibility criteria are based on a “scorecard” of indicators (ruling justly, encouraging economic freedom, investing in people).
• Countries must pass at least 10 indicators, but must always pass two “hard hurdles”: democracy (not covered here) and corruption (06:03).

3. Problems with the Corruption Indicator

• Charles Kenny:
  • MCC uses the World Bank’s Worldwide Governance Indicator (WGI) for corruption—the best available global measure—but “the best isn’t good enough” (06:34).
  • The corruption metric is an amalgamation of perception surveys, self-reported bribe payments, and assessments of laws—resulting in fuzziness and wide margins of error (07:32).
  • “Perceptions of corruption is not a policy.” Instead, it’s “broad, amorphous, corruption-y, perception-y thing.” (21:23)
• Casey Dunning:
  • Evidence linking these corruption indicators with real-world development outcomes is weak; “if we’re looking at the corruption indicator...the connection is not necessarily always there” (09:02).
  • The “hard hurdle” penalizes countries that appear to be improving, simply because the median (comparative) benchmark moves (09:45–10:14).
  • Example: Sierra Leone and Benin lost eligibility mid-process after failing to keep up with relative improvements, despite actual progress (10:14).

4. MCC’s Perspective and Practice

• Alicia Phillips Mandaville:
  • Emphasizes the necessity of using available third-party data; “what we have on the scorecard right now truly does cover every country in the world and we need that” (28:00).
  • Admits the system isn’t perfect—scorecard data is supplemented by other evidence and narrative, but for the key first stage, the hard hurdle remains (11:57, 12:44).
  • Board has discretion to review additional information (transparency initiatives, anti-corruption reforms) but has not overridden a scorecard fail since MCC’s first meeting in 2004 (12:44).
  • On ongoing reforms: There’s recognition that “tracking change year on year is harder and the scorecard wasn't set up to do that” and a need for “measures that are both actionable and measurable and practical” (28:00, 29:25).

5. Policy Recommendations and Tradeoffs

• CGD Recommendations:
  • Charles: Replace broad perception indices with more granular, policy-actionable indicators (vaccination rates, anti-police-bribery stats); perhaps MCC could fund new, better data collection (20:42, 21:31, 22:35).
  • Casey: Given the lack of statistically significant examples of improvement and “hard hurdle” effectiveness, recommends dropping or significantly revising the hard hurdle (24:40–25:09).
• Political Realities:
  • The optics of “dropping the hard hurdle” could be problematic in Congress, risking accusations of being “soft on corruption.” Nuanced reform might focus on differentiating between original eligibility and ongoing eligibility during compact development (26:04, 26:48).
• Mandaville’s Response:
  • Acknowledges limitations, pushes for collective action on better governance data, stresses importance of using independent sources to avoid “tinkering” accusations (23:02, 24:13).
  • Excited about building collaborations among data producers and users, linking this to broader global initiatives on data and governance (29:25).

---

Notable Quotes & Memorable Moments

• On the MCC Approach:
  • “We are the, if you’ve heard of us, we’re the foreign assistance agency that does those things…If you haven’t heard of us, we’re the small foreign assistance agency that you haven’t heard of.”
    —Alicia Phillips Mandaville (01:53)
• On Indicator Flaws:
  • “The best isn’t good enough.”
    —Charles Kenny, on the corruption indicator used (06:34)
  • “Perceptions of corruption is not a policy…that’s a very fuzzy measure of a lot of different things, some of which may be policy actionable…”
    —Charles Kenny (21:23)
• On Real-World Consequences:
  • “There are very real world consequences to this highly technical discussion.”
    —Casey Dunning (10:14)
• On Lack of Policy Actionability:
  • “Only one country in the MCC’s 10 year history moved from statistically failing the corruption indicator to passing…the hard hurdle is not designed to account for it.”
    —Casey Dunning (24:40)
• On Board Discretion and Reform:
  • “The board has discretion, period, full stop…The idea behind a scorecard was can you take data and use it to start your decision making with and can you use it to be accountable…”
    —Alicia Phillips Mandaville (19:06)
• On the Broader Data Revolution:
  • “Trying to work with the folks who actually produce data on governance…there's finally a world in front of us where all those people want to talk about data. And we're going to. So that's like I'm really excited about that.”
    —Alicia Phillips Mandaville (29:25)

---

Important Timestamps

• What is MCC? — 01:53
• History and bipartisan support — 03:40
• Details on the hard hurdle and indicators — 06:03
• Why the corruption indicator is problematic — 06:34–09:02
• Consequences of failing the hard hurdle (Benin, Sierra Leone) — 09:45–10:41
• MCC practice on supplementing data and board discretion — 11:24–13:07
• How board and staff handle technical ‘wiggle room’ — 16:46–19:06
• Critique: Need for actionable, outcome-linked indicators — 20:42–22:35
• Recommendation: Drop or revise the hard hurdle — 24:40–25:09
• Political constraints with Congress — 26:04–26:48
• Mandaville’s wrap-up on moving forward, actionable data — 28:00–30:03

---

Tone and Style

• The conversation is expert, candid, and occasionally playful (“I was outvoted,” “I love that 20 something title, hating on the Hurdle”), but rooted in real technical and policy substance.
• Critiques and defenses are direct but collegial, with an emphasis on data, accountability, and practical governance challenges.

---

Conclusion

The episode offers a thorough, frank look at the difficulties of operationalizing anti-corruption principles in international aid. While MCC’s ambition to use data-driven “hard hurdles” is lauded for its transparency and rigor, its reliance on imperfect, perception-based indicators raises questions of fairness, effectiveness, and even political practicality. Both the CGD panelists and MCC’s Mandaville agree on a shared vision: the need for better, more actionable data to drive smart, effective, and truly accountable development policy.

---

Transcript

[A]

Welcome to the Global Prosperity wonkcast. I'm Lawrence MacDonald. We've got a full studio today. We have three guests. With me from the Millennium Challenge Corporation is Alicia Phillips Mandeville. She's the Managing Director of Development Policy. Alicia, welcome to the show.

[B]

Glad to be here.

[A]

And two regulars. Charles Kenney, senior Fellow here at the center for Global Development, and Casey Dunning, who's a policy associate with our Rethinking US Development Strategy Program. Charles and Casey, welcome to the show.

[C]

Thanks.

[A]

We just had a roundtable here. The good and the great of thinking about US Development aid effectiveness were in the room and it was occasioned, I gather, by a paper by Charles and Casey hating on the Hurdle. I love that 20 something title, hating on the Hurdle. I think that's why it got all the attention was the title. Did this come from you, Casey?

[D]

It was a collective decision to not go lightly on.

[C]

So I was outvoted.

[A]

It's like valgal talk. Anyway, it's called Hating on the Hurdle. So tell us, how do you feel about this hurdle? Reforming the Millennium Challenge Corporation's approach to corruption. And also Alicia really honored us, and I would say honored CGD by then writing a blog post about it. Responsible Development Data Rule number one, Know Thy Data, which is a very interesting discussion of the importance and the constraints of data. We're going to cover all of that, but first, I know I have some listeners who up to this point have no idea what we're talking about. Alicia, what is the Millennium Challenge Corporation and why should we care?

[B]

Great. Certainly. And thanks for having me in today. I'm really happy to sit down and talk about the paper and life in general with folks here. We did debate whether the title of that blog response should be Don't Hate the Player, Hate the game. But we decided to go with the other one instead. So what is mcc? MCC is not a catchy title generation place. It is an independent US foreign assistance agency that was set up to test out kind of three experimental ideas about foreign assistance. One, can you identify places to place development investments by looking at the policies of that country and using third party data to measure them? Two, how can you work with country ownership in a way that informs those investments? And can you test out the impacts of that? And three, how can you hold greater accountability to the project? So can you measure, set out to measure what you're going to do, measure it and then tell people what you did? So you learn from those lessons. So we are the, if you've heard of us, we're the foreign assistance agency that does those things with the United States. If you haven't heard of us, we're the small foreign assistance agency that you haven't heard of.

[A]

And just to traffic and stereotypes here, this is in response to an allocation of U.S. foreign assistance that was not necessarily driven by data and driven by transparent indicators and which might therefore be readily subject to political influence. So especially during the Cold War, money to our friends, no money to our enemies. This was to do away with all of that and say, if you're a good government and you govern justly and encourage economic freedom and fight corruption, then we'll provide assistance. And if you don't do those things, we won't.

[B]

Yeah, I think that's, you know, there's a reason there were millenniums in the title. Right. So it was around that time a lot of lessons coming out of development, some of which you just said in terms of looking for a good policy environment to place investment, place development investments, but a lot of other lessons as well.

[A]

So there are two other broad things I would mention about the MCC as I understand. One is it was a George W. Bush initiative, but it's enjoyed, I think, quite broad support on the Hill. Bipartisan support in an era where very little, very few things have bipartisan support.

[B]

Yeah, no, I think that's something we're generally pleased with. You know, this is our 10 year anniversary, so I mean, the MCC's been operating now for 10 years trying to put into practice some of those lessons. And it's been really reassuring how much that has resonated on both sides of the aisle. There are different things that resonate on different sides of the aisle, but there is. It's very hard for someone to argue that they don't like rigor or they don't like sound policy investments. And so that's been really great.

[A]

And the final little piece of history has been CGD's involvement. Because when I first joined CGD, it's now been 10 years. We had something, in fact, I think I was involved in the early discussions about what to call it, whether it should be the. At that point it was still called Millennium Challenge Account because it was quite new that the entity was being set up. And was it going to be a watch, was it going to be a monitor? We eventually settled on monitor in part because things that are called such and such WATCH seem like they're hypercritical, where it was our intention. I think in CGD this seemed like a pretty good idea. Some of it borrowed from work by Steve Radlett and other colleagues here. And CGD wanted to be supportive, but also a little bit arm's distance and say, when you do something right, we'll praise you and when you could do a little better, we'll say so now, first, maybe it's useful to know, and I think that kudos to Charles and Casey. For those of us who don't know it very well, you give a sense of the scope. There was about $900 million budget in 2012. Since its inception, it supported 25 countries. So as aid agencies go, it's not nothing. It's also not huge.

[B]

Yeah, I mean, I think a lot of people are surprised at the size of the agency because we do work in about 25. We have worked in about 25 countries. But partly they're surprised at the agency being small because the scale of the investments is quite large.

[A]

I learned from your video, you said there are 300 people there.

[B]

Yeah, exactly. There's 300 staff total. By law at MCC, we are not allowed to get bigger than that in terms of full time hires. And so as a result, our incentive to totally rely on the quality work that folks at CGD do is really high. So maybe that's structurally part of the good relationship there.

[A]

And as a way of getting into this. So there are these indicators that fall into these broad categories, ruling justly, encouraging economic freedom, investing in people. Countries that would have a compact, which is the contract that allows the money to flow, need to pass at least 10 of the indicators. There are two hard hurdles. One is democracy. We're not going to talk about that today. The other one is corruption. And Charles, why do we hate the corruption? Hard hurdle?

[C]

Well, first of all, I'd say one of the reasons more people have heard of the MCC than you might think, given how small it is, is it does have an outsize influence. And it has an outsize influence partially because it's doing a lot of stuff that we at CGD and a lot of other people think is great. The whole scorecard methodology of, you know, you are eligible to be considered for a compact if you pass on all of these indicators is just a wonder of transparency and a great thing for aid agencies. The corruption indicator that the MCC has chosen, I think it would be fair to say is the best available general indicator of corruption that comes out every year that is available for all the countries that the MCC might give a compact for. It's just the best isn't good enough.

[A]

And it's a Brookhens world bank project where basically you survey a whole bunch of people and you say which country stinks worst?

[B]

Well, a little more technical.

[C]

So they take a whole bunch of those surveys which ask people which country stinks worse? And they take some surveys that involve asking people, have you paid a bribe recently? And they take some surveys that are about particular bits of law and they kind of mush it all together and they say, what's the common thread between all these indicators? We're going to take that common thread and we're going to rank countries on that thread. Do they score the worst on average across all of these or the best on average across all of these or somewhere in the middle.

[A]

So some of it is perception and some of it are observable characteristics.

[C]

Yes, although observable characteristics also falls into two categories, right? There's observable characteristics which is surveying people and saying have you paid a bribe? And how much? And then there's observable characteristics like laws and regulations. Most of it, I think it'd be fair to say, is perceptions or surveys.

[A]

On self reported activity.

[C]

And you know, it turns out that these perceptions and even actually surveying people and asking them paid a bribe, you get very, very fuzzy, strange answers from people. And there's a large margin of error around these indicators. Now again, the worldwide governance indicators, the particular indicator that the MCC has chosen I think is the best out of all the kind of globally available corruption indicators, partially because it tries to come up with a measure of, well, how fuzzy is this indicator?

[A]

And I think your paper at one point makes the point that you think this is a pretty good indicator, especially for research, but, but maybe not so good for allocating money. Right, Casey, why is that?

[D]

Well, the idea behind it is that the corruption indicator isn't directly tied to strong development outcomes as we found in the paper. The base of evidence says that if we're looking at the corruption indicator as compared to how these countries have performed on economic growth or any of a whole host of issues, the connection is not necessarily always there. And so is it wise to have sole emphasis on this one indicator if there's not a strong evidential link between the indicator and outcomes?

[A]

Did I understand the paper? I know Alicia's like, argh, I'm going to get there, I'm going to come back. But did I understand it correctly in my cursory, admittedly cursory reading of the paper that you also need to, whether or not you pass the hard hurdle is based on your performance relative to the median. So I could be getting better and if other countries were getting better faster than me, even assuming that the indicator wasn't squishy, which it is, I could find myself, having worked for three years on a compact, suddenly not be eligible.

[D]

You're absolutely right. And that's what happened in December in the cases of Sierra Leone and Benin. They were actually developing a compact. And because they fell just under the median, literally just under the median, the MCC board did not reselect them as compact eligible. So there are very real world consequences to this highly technical discussion, Alicia, that.

[A]

Would feel to me like a drive by. You know, I'm working, I'm working, I'm working. And whoops, you know, the median went up a little bit. I didn't really get worse, but sorry, you know, no gold ring.

[B]

Yeah, no. And I think in the discussion we just had, I was saying there aren't any philosophical questions for us. Somebody said, I have a technical question and a philosophical question. And I said, we don't have philosophical questions because people want to look at data and they need to take a decision. And so that's, you know, I can't emphasize Casey's point that there are real life consequences for these, for technical issues. That's definitely true. But there's kind of a flip side to that, which is that, you know, two different things you said. One, you said, we're not going to talk about democracy. And then one of you said that.

[A]

Just because I don't have my head around it.

[B]

No, no, no, no. And one of you said this. The world governance indicators are okay for academic research, but maybe not to allocate money, but there's actually a link there because a good academic researcher wouldn't just use one variable in a piece of research. They would look at this, plus a lot of other things. They'd control for income, they'd control for democracy, they'd control for a number of other factors. And that's essentially what the rest of the scorecard and the supplemental research that MCC staff does, that's what that's supposed to do. So it's.

[A]

So in this case, MCC staff makes a recommendation to the board based not only on this indicator, but on a lot of other evidence.

[B]

So our CEO makes a recommendation to the board every year based, starting from the. Starting from the scorecard, of which the indicator is a hard hurdle and so a key piece. But we do supplement that with a lot of additional research, and it's something we've tried more and more over time to be transparent about. What other data are we looking at? We look at the Extractive Industry Transparency Initiative, we look at open government partnership, we look at transparency international, global integrity. What else is there that has data, has narratives attached to it so that we can say, yes, the number went from 0.0 to 0.04, but here's some information about what actually happened in the country.

[A]

Has that happened with the corruption hard hurdle where a country fell below the median but the other evidence suggested that they were moving in the right direction, where you have recommended or the CEO has recommended to the Board that the compacts proceed.

[B]

So at the point of selection, when the board sits down to decide who will be newly eligible for a compact, we haven't. That has not happened since the very first meeting. In the very first meeting in 2004, MCC's board selected three countries that didn't pass the control of corruption hurdle. And every year since then, we've had to not be able to say that we've never selected one of those countries because since then we've only selected countries that have passed this passed the hurdle.

[A]

So this other narrative stuff in point effect hasn't made a difference since that first year.

[B]

Well, so at the first point of selection, but there are every year changes in the data. And so what Casey's highlighting Benin and Sierra Leone. Benin and Sierra Leone had been previously selected to develop proposals for compacts. And we're working to do that, have been working to do that with MCC this past December, the board had already selected them, new scorecards came out. And so until we sign that compact, until the letters are all written down about what exactly the investment will be, what will MCC do, what will the country do? The Board reapproves that decision that it took. Yes, we decided MCC should work with Benin and Sierra Leone. We will look at the information we have currently available and affirm that that's still the right decision. So this year, because there were, because the scorecard had changed, because these two countries didn't pass control of corruption, that raised a lot of questions and the board wanted to talk about that. So the decision to not to tell MCC remain engaged with these countries, that decision was based on supplementary information. We can dig in and look at the policy issues in the country and talk about the fact that Benin had stood up an anti corruption commission or Sierra Leone had passed a Freedom of Information act and those things didn't suggest motion in the wrong direction. So the board said to us, engage, stay engaged. But the board also said we don't want, we're not going to vote on whether or not to make them eligible for this year. So that means that those two countries don't have access to this year's funds. But MCC has been pretty clearly instructed stay engaged with those countries. So that conversation is still going on.

[A]

Terrific. We're going to take a short break. When we come back, I want to push you a little further in terms of helping me to understand the politics of staff, management, board, Congress, public opinion when it comes to managing corruption. Alicia is wrinkling her brow. This is the Global Prosperity Wonkast from the center for Global Development. My guests today are Alicia Phillips Mandeville of the Millennium Challenge Corporation, Charles Kenney and Casey Dunning of the center for Global Development. We will be back in a moment. Welcome back to the Global Prosperity wonkast. I'm Lawrence McDonald. My guest today, Alicia Phillips Mandeville of Managing Director of Development Policy at the Millennium Challenge Corporation, Charles Kenney and Casey Dunning here at the center for Global Development. We're talking about something called the hard hurdle on corruption. And we've unpacked a little bit how the MCC allocates money based on data based on indicators of performance across a variety of criteria. That one of the two that you have to pass is on corruption. And there's a lot of questions about that indicator. The thing that I've been trying to understand, Alicia, is as I understand it, one suggestion might be since these things tend to fluctuate a little bit without necessarily a close connection with the underlying trend, broadly they're correct, but they go up and down. Something like smoothing the data over three years. Is that something that the CEO could decide to do and inform the board, or would the board board say, no, you can't, or the Congress would say, no, you can't. And is Congress the sort of entity that ought to be deciding the appropriate way to handle statistics? I kind of think maybe not. So, you know, where does sort of technical latitude of the experts extend to and where does it end in something like deciding on this particular indicator and how you use it?

[B]

No fair question. And I think it's hard to answer without getting into the technical weeds of things.

[C]

That's good.

[B]

Bear with me. That's why we call this the wonky Excellent. Okay, good. So the first point is that where do these rules come from? Maybe that's the kind of where does the rule about MCC has this corruption hard hurdle and exactly how is it constructed? And what data do we use? And what's the rule about what's the line above or below which you have to be. So our board sits down every year in December and takes a decision about which countries MCC is going to work with. Looking ahead, where can we make. Where does it make sense for MCC to develop a Compact or threshold 60 days before that? So usually it's in September the board approves the methodology that they're going to use to make that decision. So in not so technical terms, basically, what are the rules of the scorecard for this year? So, and not just what are the rules of the scorecard for this year, but we actually, in that report, lay out what supplementary information might we also be looking at. So this is where we've tried to be more transparent about, like, what else, what else is there?

[A]

So an effort to separate the rules from the players a little bit.

[B]

An effort to actually talk about all of the rules. So everyone thinks the scorecard is it. I mean, not everyone. Most people. I get very frustrated sometimes because my falls are intensely chaotic and busy and I see my family very little. And every so often someone says, I don't understand what you're sending so much time on. You have the scorecards, the people just look at them and they make a decision. And there's way more than that. And the way more than that is part of what we're also trying to unpack in that report. So the rules on the scorecard are where people tend to focus most. And we have this, and they should. They're the start point. They're the first filter. Countries that don't get past that at the point of selection don't get very. Don't get much farther. In fact, don't get any farther, but there's a lot more that happens after that. And so that's where some of the things that, you know, the paper highlights, you know, there's a challenge with statistical noise, or there's a challenge with. It captures perception of corruption. So if there's been a scandal recently, people's perception ratchets up. Even if the country did the right thing. And really, frankly, corruption's probably been reduced. So some of the research into, okay, what's actually happened in the country, but some of that is also research that's not about corruption. Some of it is. If we've worked with a country before, what was our experience in working with them? And what does that tell us about what a second compact investment is?

[A]

So where's the wiggle room? How much wiggle room does the CEO and staff have in the hard hurdle here? Could you just say, you know, CGD is right. We're going to change what we do here.

[B]

I mean, the board has discretion, period, full stop. It's the board's decision and the board's decision is the board's. I learned long ago not to get in front of our board and think what they were going to think because that's their job. I will say one of the the idea behind a scorecard was can you take data and use it to start your decision making with and can you use it to be accountable for your decision making? And this data really serves that function well in that the amount of attention people have put on measures of corruption and whether or not countries are meeting that changes the conversation away from did this country vote with us at the UN or not last year over to is this country doing the things that we think a country needs to do to better facilitate its own economic development? And you know, it's the board's discretion. The board looks at the scorecard, they look at the supplemental information. But that role of the scorecard in terms of putting data front and center in the conversation, something that's objective, something that we can all disagree about, but that's still what the number says over here.

[A]

Charles I'm going to imagine that more than one, maybe several, maybe more than half of the MCC board members are going to listen to this wonkcast and now I'm going to say we're going.

[B]

To play it at the next meeting.

[A]

I'm going to say what's your advice to the board in terms of their instructions to the CEO and staff regarding this indicator? What should the board be asking, what questions should they be asking and where should they be directing the agency to go in terms of rethinking the hard hurdle? We've heard from Alicia. The board can make these decisions. What's your advice to the board?

[C]

So there are some indicators in the scorecard that are just wonderful policy actionable indicators.

[A]

Policy actionable meaning the country that's trying to get the money can change what it's doing. And that was one of the ideas of the mcc, is that countries with would jump through hoops to get the money. So you have to see where the hoop is.

[D]

Yeah.

[C]

And there's this thing called the MCC effect which shows in some cases, actually because the MCC said you need to do this, some countries have done it in order to get a compact so it can work. One of them is measures of vaccination. Right. I think that's a wonderful indicator. Right. Countries have to vaccinate more kids in order to improve their chances of eligibility for an MCC contract. That's saving lives.

[A]

That's brilliant.

[C]

And it's policy actionable. We know how to increase vaccination rates. Perceptions of corruption is not a policy. It's actually, I could take out an.

[A]

Ad in the New York Times that says my country's really clean. That would help.

[C]

Indeed. My worry is that may be one of the most effective ways to move your measure on this. That it's a very, very fuzzy measure of a lot of different things, some of which may be policy actionable in the reasonably short term. I mean, let's imagine it was really easy to root out corruption in the police force. If one of the measures underlying the WDI Worldwide Governance indicators is percentage of people who pay bribes to the police, well, you could go out and try and root out corruption in the police force. That would be a better indicator, right? At least we'd have something that we kind of know which bit of the system we're looking at. We're looking at the police force and what are we trying to do? We're trying to reduce the number of people who are paying bribes. That's sort of a comparatively specific indicator about corruption.

[A]

So are you suggesting the MCC create a whole bunch of its own indicators.

[C]

Gather its own data, these kind of indicators actually what go into the wgi.

[A]

So you could cherry pick the ones that tie to country actionable.

[C]

Now, the complaint is that those indicators aren't produced every year by every country. Luckily, the MCC has the power to spend some of its money on creating data. I believe this might be a great place to spend some money on actually having an indicator of some bit of corruption that you care about that is measurable and policy actionable, rather than this kind of broad, amorphous corruption y perception y thing.

[B]

Here's where the constellation expands to be kind of the whole sky though too, in that there's the part where we have a challenge about what data is available to measure this. And there's some stuff we're trying to do with the people who actually gather and create data on governance and anti corruption and integrity. Folks who use data like us, folks who enable data like foundations, some work we're trying to do collectively to figure out if there's a more collaborative alliance y like approach to thinking about governance data and what's available that is more actionable and is more kind of predictable year on year and that governments can look at and know what they would want need to do. There's that piece of the equation. There's also the function of our scorecard where that data is neither created by us nor the country that's being assessed. And having been at MCC for seven years now, and the number of times that country governments come in and say, I don't agree with this, our ability to say, you know, there may be errors in this data. This is the, this is the contact information for the place that collects the data. And we would love for this data to be more accurate. That's like one, it makes Lisha's life easier, if I can do that. But two, it also, I think it reduces the perception that the US Government is somehow tinkering inside the data. And that's something that's actually really important to us.

[A]

Charles probably wasn't suggesting you would gather it, but that you might pay somebody else to do it. But I guess then does the same problem still arise?

[B]

Yeah, yeah. I mean, if the Supreme Court is still debating the perception of when money moves, where and what influence is held, and I'm not going to have a view on it.

[A]

Casey, I want to bring it back to you because I didn't get to the end of the paper. I know there's, there are a number of recommendations. What would your advice to the board be in terms of making better use of corruption data in particular, different maybe from Charles suggestion?

[B]

Sure.

[D]

Well, just to actually build on Charles, look, our research found that only one country in the MCC's 10 year history moved from statistically failing the corruption indicator to being, to passing the indicator. That country was Georgia. And when the MCC selected it, it was failing the corruption indicator. So clearly in terms of the policy actionable part of this indicator, it's not there. My advice.

[A]

So you just apply sort of a standard definition of statistically significant.

[D]

Indicator. But the hard hurdle is not designed to account for it. So my advice to the board would be, as is in the paper, to drop the hard hurdle. Because there are cases, as with Sierra Leone and Benin, that they were moving in the right direction. They passed the scorecard, all else except for the corruption indicator. It was this hard hurdle rule that was the problem.

[A]

So to the board, okay, so I'm going to take your advice. I'm the CEO. Talk to the board. The board says, fine, we're going to drop the hard hurdle because Casey Dunning said we should. Six months later, there's a congressional hearing. It's time for appropriations again. Congress says you're soft on corruption. Right. You know Congress way better than I do.

[D]

No, you're absolutely, absolutely right. And I think the Hill aspect here is of critical importance because the optics of dropping the hard hurdle would be just that. I think there is maybe a way to thread the needle in saying once a country is newly made eligible, they must obviously pass the corruption indicator. And then is there a difference between a country who gets into the compact game versus a country who's moving along and then all of a sudden has a drive by, as you said? And so maybe we could think a little bit more kind of nuanced about how we use the hard hurdle in conjunction with a country's path along the compact process.

[A]

So I pass it and then you start smoothing the data or something like that. Charles shaking his head. That's probably bad statistics or something.

[C]

Well, no, I think Congress thinks it's being hard on corruption, but this is not a good indicator of corruption. What we need to move towards is indicators that are better measures of particular elements of corruption or measures of real development problems connected with corruption that governments can act upon. That's the entire reason the MCC is around, right? Is to try and influence policies in a better direction. So let's go towards measuring policies that will have an impact on corruption, policies that we can actually measure that will have an impact on corruption. Then we're being tough on corruption at the moment. We're being tough on countries kind of semi randomly based on the fact that they had a bad week and the Wall Street Journal covered their cover their country in a negative way. That's not fair and not sensible.

[A]

Alicia, I want to give you a crack at the same question I've asked both of my colleagues and that is, you know, somebody's going to ask you, presumably possibly the board is going to say so CGD says this corruption hard hurdle is being done the wrong way. You've interacted with them, you attended their roundtable, you've been on the woncast. What's your takeaway advice? What should the MCC do now about the corruption heart hurdle?

[B]

I mean, I think, you know, Charles point just now that the direction we need to move is that there are measures that are both actionable and measurable and practical for countries to follow. That is the direction that not just mcc, but the field of people who are interested in countries having more transparent, less corrupt practices all need to kind of think about. And you know, the what we have on the scorecard right now truly does cover every country in the world and we need that. That's a non starter for us. If it only covers half the countries that we look at, then it can't be a hard hurdle because half of them automatically fail by dint of not having or automatically pass right now. The rules are conservative right now. So you fail if you don't have data. That's a fact and that's a reality. And we live with that reality. We've built all kinds of systems to figure out how to live inside that reality. But looking ahead, and this is what I'm actually really excited about and why I was so happy to come and talk about this today was that, you know, trying to we have been doing this for 10 years and we have 10 years of lessons on what does it mean to use data to inform decision making. And the reality is that we do see the need for something that you can use to track over time, not just a snapshot. Scorecard gives you a snapshot at a point in time and it has served us well in that purpose. But tracking change year on year is harder and the scorecard wasn't set up to do that. And we try not to use it in that way. But in the absence of something else, it's very people default to it very quickly. And so in the next several, in the next week and in the months to come, trying to work with the folks who actually produce data on governance and people who also, like us, consume data on governance and people who are interested in ensuring that there's an infrastructure for there to be a continued supply of data on governance that fits inside not just MCC's needs, but all of the conversation around the data revolution, the conversation around the post 2015 Millennium Development Goals, broader conversations about governance and anti corruption like it's there's finally a world in front of us where all those people want to talk about data. And we're going to. So that's like I'm really excited about that.

[A]

Thank you very much. I think we're going to leave it there. This is the Global Prosperity Wonkcast from the center for Global Development. My guests today have been Casey Dunning, Charles Kenney and Alicia Phillips Mandeville. We've been talking about the Millennium Challenge Corpor Corporation's hard Hurdle. You can find the Wonk cast online on itunes and on Stitcher. Just search for wonkcast or CGD to hear a new show every week. Until next time. I'm Lawrence MacDonald. Thanks for listening.

[B]

Sam.