Loading summary
Transcript64 lines
- [00:05]
A
Welcome to the Global Prosperity wonkast. I'm Lawrence MacDonald and I'm delighted that my guest today is Lant Pritchett, an old friend from my days at the World Bank. He's a professor of the practice of development at Harvard University's Kennedy School and I'm also proud to say, a non resident fellow at the center for Global Development. He's also the chair of our advisory group. Once a year we get very smart and pleasant people to come around the table and tell us how we should do a better job. And we've just finished having that discussion. So, Lant, welcome to the show.
- [00:36]
B
Thank you very much for having me.
- [00:38]
A
I brought you in today in part because I wanted you to unpack something you were talking about today with the advisory group. This idea of capability traps and isomorphic mimicry. Those are some kind of highfalutin words. There is a paper on our website for those who are interested. Capability traps, the mechanisms of persistent implementation failure. What's that all about?
- [01:02]
B
Well, it goes to the heart of what we mean by development. Kind of what we mean by development is a natural process that unfolds over time. So tadpoles become frogs, and the natural state of affairs is for tadpoles to become frogs and for acorns to become oaks. And the idea of development that was prevalent was that these incipient, less developed nation states were, would naturally become more developed nation states and that the tools of development were to accelerate that process. But the underlying process of a tad becoming a frog was the natural process. Now, you know, we're 200 years after Haiti's independence as a free and sovereign nation state. It's still a basket case. It has not developed. And in particular, the component of development I'm interested in is what's the ability of the government to do stuff. You know, you kind of assume that over time, part of the tadpole to froggishness of development was governments would acquire capabilities. They'd be able to do policing, they'd be able to deliver the mail, they'd be able to build roads, they'd be able to educate children. And if you look at Haiti's capability in 2010 versus its capability in 1810 when it became independent, there's just not evidence that, that the process of developing an administrative capability appears to be as natural and inevitable a process as we once thought.
- [02:30]
A
And so why do those. So that's the capability trap. That's the capability trap. You're not very capable now. Never have been, maybe never will be.
- [02:38]
B
Or certainly we calculate in the paper, for instance, we say, look, if we calculate a business as usual trajectory path, how many years will it be before countries acquire the capability of, say, Singapore? So we take Singapore as a sort of prototypical super high capability state. Well, the answer is at their current rates of progress, many of the poorer countries would take literally thousands of years at their current rate of progress to reach Singapore's level of capability. That's not sort of a process of development that's going to unfold in the time that anybody wants these states to acquire capability. So that's the trap is this very slow progress.
- [03:19]
A
So you have some measures of this and there's some empirical. When you say thousands of years, this is not just a sort of philosophical notion. These are based on numbers. Can you tell us what one of those numbers is and explain one of these indicators that you can? How can you measure capability?
- [03:32]
B
Well, say you walk into a forest. The trick is measuring the change in capability, right? And a lot of people say, well, today you can just look around the country and one of the indicators we take is say, bureaucratic quality. Well, you maybe can't measure bureaucratic quality super precisely, but lots of people have come up with rankings that rank countries and their bureaucratic capabilities, say from 1 to 10. So we take those rankings and we take four or five of them just to make sure we're robustly measuring a component of state capability. But those rankings are pretty reassuring that you know it when you see it. You don't have to have super precise definitions.
- [04:09]
A
So bureaucratic capability, what would be. I mean, this is actual.
- [04:11]
B
Deliver the mail, okay, you know, are you able to deliver the mail?
- [04:15]
A
Are you able to deliver. So there's a number of people who actually get mail. So it's grounded in a real world measurement. It's not just perceptions of quality written by some people who look at the country and say, we think this one's corrupt.
- [04:24]
B
It's. I mean, yes and no. It is the perceptions of people, in part because the phenomena are really too precise to be nailed down, but it's the perception of people on some pretty narrow criteria. So for instance, one of the four measures we use is the failed state index that's computed by some people here. Across the street at Carnegie has a subcomponent called the progressive deterioration of Services, in which they articulate sort of what's happening to your ability to deliver services, and then they ask country experts to rate the countries. But again, are these ratings super precise? No, but neither is GDP per capita. But what they all kind of Uniformly turn on and let me get back to this is that Somalia is at the very bottom. Which makes sense because Somalia, we know, doesn't have any state capability, doesn't have a state, and Singapore is pretty near the top, which again accords with our kind of natural reasoning. And most of the rich countries are up near the top and most of the really poor countries are near the bottom. Now, the coming up with current measures, really the trick is, how do we know what the rate of progress has been? So what we do is we use this trick. So suppose you walk into the forest and somebody says, how fast do these plants grow? Well, you might say, well, I can't tell you how fast they grow because all I know is that some are taller, some are short. But all you really need are three pieces of information to know exactly how fast it is a plant grows. One is how old it is, one is how tall it is right now, and one is how tall was it when it started. So if we just assume that the plants in the forest started from the ground and grew up, we know that their initial state was zero. If we can measure how old it is and know how tall it is, we know exactly how fast over its entire lifespan it grew. So we just apply the same trick to countries. We say, when were you independent? What is your capability today? And then some trajectory of the evolution of your capability. Let's assume the fastest your capability could have grown is if your capability when you were independent was as low as it could be. So what we do is we say, Suppose Haiti in 1810 only had the state capability of Somalia. Well, that's pretty low. I don't think it was any lower because Somalia is essentially anarchy, I think. And then Haiti got from where it was. Say we suppose it was at Somalia independence, which again biases us in favor of finding fast, not slow, capability growth. Today it's a two. Somalia is a one. It's only grown one unit in our capability measure that goes from one of Somalia to 10 in 200 years. So then we say, well, if in 200 years it went from one to two, it's going to take nine more of those units to go from two to 10, right? Or eight more of those units. So that's 1600 years. So that would be an illustrative calculation. And it's pretty simple. It's just trying to let's take what data we have and piece together the growth rate. Because what we're really interested in is not just their levels, but are they on a trajectory of improvement and Just the basic fact of the matter is, you know, in 1963, when you had countries with very low levels of capability, you might have said, well, of course we've got low levels of capability. They've just newly emerged from colonialization and the colonialization that didn't develop the capability of the state. But now we're 50 years in to the development process. Even for the newly independent states, for many countries like in Latin America, we're 200 years into a process. So wherever they are today is the result of a very long term trajectory. And we can ask where would that get you if you continued that same trajectory? And that's where we get the thousands of years or at least hundreds of years. Whereas the premise was when these countries were newly independent in the 1960s and development had its burst of sort of optimism that would be able to replicate the development process very fast, we thought, well, these tadpoles are going to turn into frogs in 10 years or 20 years. Well, it turns out we're 50 years in and some of them are still tadpoles.
- [08:25]
A
So I've got to ask, are there some outliers? Because clearly the notion of growth, you pretty much debunk that idea. If you're low capability, you improve only quite slowly, if at all. But there's maybe some societies where they've managed to import ideas or institutions or something. Are there any that grow up fast?
- [08:44]
B
Absolutely.
- [08:45]
A
Are there some fast growing trees in this forest?
- [08:46]
B
There are, but this is one of the important points that we make, which is, yes, there are fast growing trees in the forest, but even those grow at a pretty moderate pace relative to what sort of expectational plans are about how fast they can grow. So take the concrete case of Afghanistan. You know, the US goes into Afghanistan and its current strategy says we're going to build a capable government. Great. How long do you think that's going to take? Well, in Bob Woodward's book, he reports that they asked that question around the table at the White House and everybody goes, I don't know. Well, so what we find is on a scale of Somalia's 1, Singapore's 10, right. Even the fast growing countries in terms of acquiring capabilities grew at a pace of about 0.1 units a year. So it took them a decade to improve by one point.
- [09:43]
A
So it's 100 years.
- [09:44]
B
So it's a hundred if you're lucky. If you're lucky, if you're a fast growing acquirer of state administrative capability, to get from Somalia to Singapore, it's going to take exactly about 100 years. And what do we expect? What does the US current strategy expect?
- [10:00]
A
It's going to take 18 months.
- [10:01]
B
Yeah, 18 months. And so, you know, in part of the, you know, drawing on the calculations of this paper, we can say, look, if in fact you're able to achieve an environment in which the capability of Afghanistan is growing at the rate of the fast growing capability, you know, the fastest states have grown in terms of their capability, in 10 years, they might be Pakistan. So whereas we're imagining that in 10 years they're going to be sort of Silver Spring, we're going to imagine that they're going to have uncorrupted government that's delivering equal services and thriving. All of this. It's just surreal, the disjunction between any grounding in what the empirical realities of the acquisition of state capability have been and the way in which development plans of official organizations often assume that capability can be grounded.
- [10:56]
A
So this brings us in a way to this idea of isomorphic mimicry, right? Exactly. In the development project, the World bank, usaid, a whole bunch of aid agencies would say, we're going to help you reform your judiciary. We're going to help you create your post office. Exactly. They recognized that these institutions were a problem. They wanted to hasten it, so they went about institution building. I've got my fingers up here for the inverted commas, institution building, which always kind of puzzled me, and I guess I wasn't the only one who was puzzled, huh?
- [11:29]
B
Well, yeah, because, I mean, the notion of isomorphic mimicry, we draw from two sources. One is nature. Let's say you're a bug and you don't want birds to eat you. Well, you could become poisonous, but that doesn't really help you until the birds know you're poisonous. So you could get some spots on your back that signal you're poisonous. But once there's a poisonous bu.
- [11:53]
A
I think this happens with the swallowtails.
- [11:55]
B
Yeah, I don't know of that example, but there are lots of examples in nature. Is that then other bugs say, look, I could become poisonous, or I could just paint some spots on my back and look poisonous. So what we. The second source of the isomorphic mimicry is from sociologists of organization who, in looking at the kinds of reforms Organization did found that often organization reforms weren't driven by any real knowledge of what would make the organization more effective, but by a drive to look effective. Right. So what you would do is you'd paint spots on your back to look poisonous. So, and by the way, when I Read this material. It struck me exactly like the World Bank. The World bank has a very difficult time proving it's an effective organization because its mission is so complex, it's got so many different objectives. So periodically, every 10 years, the world bank has a major reorganization to mimic whatever the latest private sector fad is for management organization.
- [12:52]
A
I was going to say this reminds me a lot of management fads. Suddenly everybody's got to have horizontal organization or open office plan or whatever the hell it is.
- [13:00]
B
Exactly. And so these are the sociologists. Organization were actually coming from the organizations in developed countries where they say all of a sudden you get these management consultants say, oh well, these really successful firms that, you know, produce widgets, have open offices, you're producing, you know, you're a law firm, you should have open offices. Complete lunacy. But if you can't document your own effectiveness, you adopt it so that to the shareholders you gain legitimacy.
- [13:29]
A
I actually remember in the World bank during the.com era when they were trying to attract all these best and brightest and so, you know, there was like blue jeans and toys and there were people who come around and give you massages in your chairs and all this nonsense.
- [13:42]
B
I didn't stay long enough at the World Bank.
- [13:43]
A
The bank actually adopted massages in your chair. You could call up and get a.
- [13:47]
B
Massage in your chair while you're. That's embarrassing.
- [13:50]
A
But that's precisely isomorphic mimicry.
- [13:52]
B
Isomorphic mimicry. It's like, well, Google does it and they're super profitable.
- [13:56]
A
If only we had massages, we'd be profitable.
- [13:58]
B
We'd be profitable too. But it just fundamental misunderstanding in some sense of the difference between true functionality and the form that that functionality takes. So the difficulty with the development process is we can convince, often when the donors are powerful, as in Afghanistan or some African states, we can convince them to pass any law we ask them to. But that doesn't create the capability of implementing the law. We can ask them to form an organization called a post office and staff it up. That doesn't mean that it's going to deliver the mail. So what I mean by a technique of failure is we ask ourselves, we first document this fact that the sort of development of capabilities very, very slow and then see how is it that people have managed to engage themselves in a development discourse consistently for 50 years and yet not have actually made any progress? How have they resolved that contradiction? And one of the ways they've done it is they periodically do isomorphic mimicry. They pretend to do reforms that look like the Kinds of reforms that successful people do, but without their underlying core functionalities.
- [15:10]
A
Okay, so I think we get it. There's a capability trap in the cap, improves very slowly, if at all. There is something called isomorphic mimicry, where states that are not acquiring capability acquire the trappings of capability, which confuses the issue. So we come to, okay, professor, what's your solution? And we've had a guest on the show, Paul Romer, who's got a solution, which is to say you can't really fix it. It's too hard, it's too slow. But these things are culturally determined. They're determined by rules, incentives, norms. Some of it's written, some of it is not written as the behavior norms. But if you could set aside some blank land and import the rules, then people coming from societies that are incapability traps will change their behavior. These corrupt, lazy people who come from the interior of China will become industrious Hong Kong citizens who play by the rules. I actually love this idea because I lived in Hong Kong for a long while. But you today said you had another idea besides the charter cities. What is that?
- [16:07]
B
And by the way, these ideas are not at all in contradiction. As many charter cities as Paul Romer can.
- [16:14]
A
He wants a hundred of 10 million.
- [16:15]
B
E giant hundred of 10 million. More power to him. If he succeeds, I'm happy and I'm convinced that he has the diagnosis right, because these systems are self reinforcing. So our sort of, we take isomorphic mimicry in some sense up one level, which is instead of just being a reaction of organizations, it can become in some sense your system. So that there's no independent measures of functionality. Once there's no independent measures of functionality, isomorphic mimicry can persist forever. So for instance, one of the issues I work on a lot is education. Now education is a great example in the development sphere because at the things that the world has measured functionality on, like getting butts in seats, we have been fantastically successful. You know, the enrollment rates around the world, rich countries, poor countries, terrible countries, pretty good countries, they've all seen phenomenal rises in enrollment rates because it was something where you could measure the goal and pursue it. But in terms of learning while kids are in school, it's complete isomorphic mimicry.
- [17:32]
A
So actually I'm reminded in the case of vaccines, it's different then the vaccine is in fact the outcome. You get the kid vaccinated, she doesn't catch the disease.
- [17:40]
B
Exactly.
- [17:40]
A
You get her bum in a seat, she doesn't learn to read. You've Measured the wrong thing.
- [17:44]
B
Exactly. And so the difficult areas for isomorphic mimicry is when it's very difficult to tell from the outside what the true output should be. But then what happens is the system never systematically measures the output, so you can never tell you're failing. So for instance, I mean, India is just a classic example of this. They have NGOs out there doing actual assessments of what children are capable of that find just spectacularly awful levels of learning among many segments of the population. Part of the Indian population is the most educated in the world, but they have sort of more geniuses and more illiterates than any country in the world. So now the government's response to this, well, we need to focus on the quality of schooling. But they define quality strictly in terms of infrastructural measures. So if you have a girl's toilet, you're a quality school. And then if we say, well, the girls toilet doesn't necessarily lead to a quality school, they say, yes it does. We've defined it to be that. So it's circular, right? It's like, yeah, once we have achieved these, we have achieved quality. You say, but what about learning? Well, that's too hard to measure. We'll put that to one side.
- [19:00]
A
And Charles Kenny is saying, we need a development learning goal.
- [19:02]
B
Exactly.
- [19:03]
A
If you started to measure the learning, you could actually maybe make some progress on it.
- [19:06]
B
Exactly. But that's the thing. In an isomorphic mimicry system, organizations can't get ahead by being better at promoting learning because there's no systematic measurement of the progress on learning. So my first solution as a professor is we need the development agencies in the world to focus on. Let's create measurements of progress that are really reflective of underlying values. I'm not talking about measurements of bureaucratic compliance with what we think your system should look like, but really of the outputs and outcomes that countries want. So go to a country and say, what is it that you want your children to be capable of when they finish schooling? And let's measure children's progress towards that goal. So, you know, I'm in favor of replacing the MDG with the mlg, the Millennium Learning Goal, as opposed to a Millennium Development Goal. That's strictly in terms of butts and seats. But we don't have to impose the same goal. We don't want to fall back into what we know what the goal should be. Go to any country, say, look, you have your kids in school for a reason. Let's measure that. Then once we measure that, then we can Distinguish sort of poisonous from non poisonous bugs, we can start saying so.
- [20:16]
A
Then you could do the same thing with water. It's the number of families that have tap water in their house. Then you don't really wor the water ministry. We assume you've got a water ministry or figured out another way to do it.
- [20:26]
B
Exactly. But as it is, if you say, if you come and you say part of the objective of the water ministry is to look like a developed country ministry. So they have an organization chart and they have a civil service schedule and they have all the trappings of the system without its core functionality. And if you come and you say, oh, we want some other organization part of the trap we're in and the reason why it's a capability trap is if you come and propose something that looks substantially different, we'll say, oh no, no, no, no, that's less quality than what we have now.
- [21:01]
A
I can't help but think this comes in a little bit to Nancy Birdswold's proposal cash on delivery aid. You think ahead of time about what you agree with the recipient country. Everybody would like to see kids learn to read, or everybody would like to see everybody have fresh water at home. And then you don't really worry about how they get there.
- [21:20]
B
Exactly.
- [21:20]
A
They may get there in a very different way than what you expected.
- [21:23]
B
Exactly. So that's why I am also a big fan of cash on delivery aid or mechanisms like it. I mean, again, we don't want to be locked into a particular mechanism. But in some sense, what the isomythmic freak does is it closes off space for innovation. You know, the reason a capitalist system works like it does is for a wide range of consumer goods. The consumer's the judge. You know, if somebody comes and says, an ipod, nobody needs an ipod. You know, entrepreneurs say, we'll see, we'll make an ipod. We'll sell it if they buy it. You know, if they buy it. I was right. And whole bunches of people are trying to satisfy demands in a huge space in which innovation gets rewarded by being evaluated on its functionality.
- [22:05]
A
So we have about a minute or two more minutes and I want to change gears because you mentioned something in our hallway conversation that intrigued me. You lived in India for three years. You've worked on India for many years. You talked about some work on the number of Dalit, what used to be called untouchable millionaires or billionaires. And then you referenced, I think Andy Sumner's got it all wrong. Poor Andy's not here to defend himself. But he has been on the show and those of my listeners who've heard more than a few podcasts may have come across this idea that there's a new bottom billion that's trapped in poverty in the rapidly growing middle income countries. And so right or wrong, I think.
- [22:45]
B
His facts are probably right, his example is probably wrong. So his fact that there are, I think, you know, Paul Collier was a genius to rhyme the bottom billion. But it's a complete fiction that the poorest people in the world are concentrated in Africa. There are more poor, absolutely poor people on whatever metric you want in South Asia than there are today. And so what Paul cleverly said was, oh, if we take the people who live in countries that are progressing and take them out of the poverty because.
- [23:19]
A
They will eventually rise out of poverty.
- [23:20]
B
On the premise that they will eventually rise, the tide lifts all boats. But maybe, maybe not. So I'm just saying the fact that you live in India and India as a country is making progress doesn't mean you're making rapid progress and there are lots of people left behind.
- [23:34]
A
So you don't really agree, disagree with Andy about that, about that, but the.
- [23:37]
B
Example he used was Dalit women. But we've just done a study together with Devesh Kapoor and Dalit activists in India showing that in fact the progress in one of the most backward of Indian states of the Dalits, precisely, has been massive precisely during the market reform era. So between is this Bihar in Uttar Pradesh, which is the state, the core sort of north Indian state that's often linked with Bihar as the poorest place on earth. We did a study of Dalits and just in every single dimension of their own social and economic self perceived status, they have made just revolutionary changes. So just for one thing, you know, and we measured from roughly the beginning of the market reforms in 1990-2007, you know, at the beginning of that period, if you ask Dalits, would they be invited, offered sort of snacks if they came into somebody else's home, which in a hospitality culture is important, but, but because they were richly polluted, other people wouldn't accept snacks from them. So only 4% of people said they would sort of have snacks if they were in somebody's home or vice versa. That's 76% today. So these underlying cultural norms are often reinforced by economic stagnation. So our argument is the market is a solvent of many social structures and the social structures of disadvantaged of the Dalits in India, we think have been substantially eroded by the market freedoms to where the Dalits now send a kid off to the city. The kid sends money home to the Dalit family. And when the landlord comes and says, I want you to do this or that, they say, no, thank you, not today, because I now have my economic power to say no.
- [25:20]
A
This is fascinating and I'm reminded of the stories. I think it's Malcolm Gladwell's stories about it's not Tipping Point, it's the other book about the importance of timing. But he's got the stories about Jewish lawyers climbing up at a time of very rapid change and they had sort have been frozen out of a lot of things and suddenly there was change and they were willing to do things or more entrepreneurial and willing to take on jobs at the white shoes firms wouldn't do. And they climbed.
- [25:47]
B
Exactly. And they did climb. And that's one thing that's happening in India is that by the Dalits, by having been the most dispossessed, don't own as much land and hence they're not trapped in the rural areas. So who's doing the seasonal migration and benefiting from the boom in the cities? So there are people trapped in poverty in India. But the interesting thing is it doesn't follow the simple social lines that people who aren't well acquainted with India might think. There's a much more topsy turvy and more interesting social dynamic going on in India than you would see from the outside.
- [26:18]
A
Lant, thanks very much. I'm reminded I would occasionally run into you on the Orange Line metro train and I always say, lant, what are you working on? And I always learned a lot in the three or four stops we run together. So thank you for joining us on the show.
- [26:30]
B
That was fun.
- [26:31]
A
This has been the Global Prosperity Wonkcast from the center for Global Development. My guest today has been Lant Pritchett talking about capability, traps, isomorphic mimicry and Dalet millionaires. You can find us online and on itunes. Just search for wonkcast or CGD if you have itunes. I hope you'll subscribe to hear a new interview every week. Until next time, I'm Lawrence MacDonald. Thank you for listening.
- [27:00]
B
Sa.