Morten Jerven on African Economic Growth
Jun 22 2015

Morten Jerven of Simon Frasier University talks with EconTalk host Russ Roberts about his new book, Africa: Why Economists Get It Wrong. Jerven, who will be joining Noragric at the Norwegian University of Life Sciences this fall, argues that economists have misread the economic history of Africa, ignoring successful episodes of economic growth while trying to explain a perpetual malaise that does not exist. Jerven is critical of many of the attempts to explain growth using econometric techniques and suggests that a richer approach is necessary that is aware of the particular circumstances facing poor countries.

William Easterly on the Tyranny of Experts
William Easterly of New York University and author of The Tyranny of Experts: Economists, Dictators, and the Forgotten Rights of the Poor talks to EconTalk host Russ Roberts about the ideas in his book. Easterly argues that poverty endures in many poor...
Morten Jerven on Measuring African Poverty and Progress
Morten Jerven of Simon Fraser University, author of Poor Numbers, talks with EconTalk host Russ Roberts about the quality of data coming out of Africa on income, growth, and population. Jerven argues that the inconsistency of the numbers and methodology...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Jun 22 2015 at 10:43am

Jerven was a good guest, but he makes the same mistake a lot of other intellectuals make. Namely, he points out problems with current methodologies without offering viable alternatives. After 10 years in social research I can tell you, if you don’t have a viable alternative to put forward you might as well not even open your mouth. People will nod their heads and listen to your criticisms with interest, but in many cases they are already well aware of these problems. They don’t fail to address them out of lack of awareness, they fail to address them because there is no viable alternative. The world needs solutions, not critiques.

Russ Roberts
Jun 22 2015 at 12:46pm


I guess it depends on what you mean by a mistake. Morten Jerven may be making a career mistake by not offering an alternative. (And to be fair to him, his book has some alternative explanations we just didn’t get to in the episode). But It need not be an intellectual mistake. In fact, I would argue that it is an intellectual mistake to pretend to have a solution that is no solution at all or to be overly confident in an explanation that is flawed. One of the great uses of economics is to tell people what doesn’t work. Or to remind people that a purported solution has unintended consequences.

Jun 23 2015 at 9:13am

Thanks for the reply Russ! Perhaps “mistake” is the wrong word, but it is a very widespread problem in social research so its become a pet peeve of mine.

I wouldn’t call “be more like Germany” a pretend solution. If Jerven can’t offer a concrete, actionable alternative, then its the best people have to work with. Telling 1990’s GM to “be more like Toyota” is great advice. The GM of today has learned valuable lessons from Toyota and is producing better more reliable vehicles. Companies improve by copying each other all the time.

Developing nations follow very similar paths in terms of infrastructure development, passing of environmental protection laws, liberalization of trade, increasing labor protections, lower corruption, etc. Yes the particulars of how they go about actually pulling it off are different and unique to the culture and history, but the end results are fairly similar. China has copied from capitalism and its economy has improved dramatically as a result. The EU is an attempt to create among European nations the common currency and free flowing trade conditions the US enjoys between states.

Finally, it sound like what Jerven is arguing for is a more qualitative approach to research methodology in economics. Qualitative research has its place, but in truth it answers a very different set of questions than quantitative methods. They complement one another. Downloading archival data and running regressions may seem divorced from reality and a risky place to make policy recommendations from, and it probably is. But from that big-picture vantage point you will discover things that people “getting their hands dirty” studying a single country would never notice. The best results are likely to come from combining the two, but that can’t happen without mutual respect and collaboration. I don’t know how it is in economics, but in psychology and education there is a strong antipathy and lack of communication between the qual and quant camps.

Brendan Riske
Jun 23 2015 at 9:37am

fantastic talk on a topic close to my heart. I believe Mr. Jerven did an excellent job explaining some of the issues in studying economic growth in Africa. I ahve a few comments:

a: It was addressed, but I have felt one of the major problems with African development is measurement. There is a huge informal sector, which isn’t captured well by statistics. This makes it very hard to see growth occurring. However, the end result,m the enlargement of the economic pie, is clear.

B: Another topic which has been discussed on econtalk is GDP. I am not sure it is a good measure of growth. Many large, “GDP” boosting projects have huge long term costs and actually drag the country down. For example, the Aswan high damn, and numerous hospitals built across the continent who cannot maintain the facilities. Another issue is the huge part of their GDP which comes from extraction. The full costs of extraction, the externalities, are almost never realized in Africa. Thus the long term costs of this growth may exceed the short term gains. Part of this is a governance issue, part of it is the inherent flaws in neoliberalism. GDP growth should be organic, not the result of outside planning. Also, the people of countries need to be made aware and given a state in the long run governance of the region, in order to better use its resources.

c: neoliberalism has been disastrous for africa. in particular, the agricultural sector has been damaged by “free trade”. Agriculture is still the largest sector of the African economy. These agreements forced African economies to open up to western finished goods, but does not force the west to drop agricultural subsidies. It has terrible long term consequences as the agricultural sector never builds up productivity, and the manufacturing sector doesn’t develop because it lacks protect from international markets. This is a major drag on development.

D: never overlook political instability in Africa as an impediment to growth. The biggest factor i Africa’s lackluster development has been political instability. It isn’t just about protection of property rights, their is no incentive to develop in a civil war environment. And many countries have weak central governments. By this I mean they cannot monopolize the use of force in their country. They may not have reach beyond a small geographical area. So regardless of laws and institutions, if they cannot be enforce, they become superfluous. Weak central governments also create conditions which allow for insurgency and civil war. Huge areas of Africa DRC, Sudan, Nigeria (and west Africa in general), Angola, Ethiopia, just to name a few, have experienced massive violence, which must have damped growth. Especially if you aggregate over a continent, and over time.

Thanks for the talk, I hope people can discuss the points I raised.

Jun 23 2015 at 8:48pm

Your initial question about how homogenous Africa actually is reminded me of a question I’ve always wanted to know the answer to. Every economic study I have ever heard of concerning Africa concerns “Sub Saharan Africa.”

That tells me that economists usually think that Africa has 2 parts. Am I wrong? Do most economic studies concerning Africa usually include Morocco, Algeria, Egypt, etc.? I’m going to assume that the 2 part view is common…

So the obvious question is why that divide? Are the countries inside each section considered similar economically? That doesn’t seem to make a whole lot of sense to me. Predominant race seems to fit the split but that open up a whole other host of questions. Do you you have any ideas about this?

Jun 24 2015 at 1:09am


I think it highly likely that the misreporting of African growth is strategic; research is funded by bureaucrats who take a living directing aid to Africa.

Jerven is way off mark in his criticism of the importance of institutions. Transplanting a particular institution out of its context would be counter-productive. Developing institutions in context which protect individuals from physical harm, resource predation, and stiffling (ala Uber in California) is vital.

Jun 24 2015 at 5:16am

I really enjoyed this podcast and disagree with James that failing to postulate a solution is somehow a mistake. I would argue it shows an appropriate degree of humility when faced with a topic of great complexity.

One thing I’d like to add, but which I think is rarely discussed in these debates, is that much of the grand, “big question” type empirical, quantitative (or pseudo-quantitative) research that goes on typically ends up demonstrating what anybody with half a brain and an ounce of common sense would know anyway. I can’t tell you the number of studies I’ve read which come up with conclusions to the effect of “corruption is bad” or “it’s important to have a legal system which enforces contracts” or “a free press is important”. Great, but where’s the beef?

Another added observation, while we’re on the topic of common sense: what happens when you occupy a large area of Africa by force, give it arbitrary borders which take no account of human geography, rule it by decree, jumble up or completely destroy its existing institutions, and 100 years later set it loose into the world as an independent nation more-or-less overnight? It turns out badly. Maybe the correct approach, which is something alluded to by Russ, is to give African countries space to develop in their own way. That may sound defeatist or depressing, but being left to our own devices didn’t work out badly for the Western world.

Walter Clark
Jun 24 2015 at 10:40am

Russ was pessimistic about changes that are more fundamental than institutions. You can’t just tell them “trust more” he said.
But there just might be a short cut to trusting more; eBay. Billions of dollars are traded without the visible hand of justice at work punishing transgressors. Once eBay and other modern transaction environments are allowed, retail spinoffs such as storefronts which do the shipping and eBay paper work, show up. When walking around type customers and storefront based retail see that example, they will seek out “trust more” mechanisms that allows them to compete.

Jun 25 2015 at 10:42am

David, it does show an appropriate degree of humility, you are right. But of Jerven’s goal is to actually influence policy decisions then he needs to put forward alternatives. Academics who just point out problems get ignored, and people with less knowledge and less humility will end up making the decisions.

Jun 28 2015 at 7:08am

Russ, thanks for taking time to discuss “Institutions.” As you know, the concept has vexed me for a long time. I got the impression after my first listen to this podcast that Jerven was not prepared for this particular question. Even so, he springboarded off several things you said and made a suggestion which I think truly revolutionary. Near 33:30, he says that Institutions could be considered solutions to problems, “…like how to do things.”


I have heard other say that private property protection is an institution frequently associated with highly productive economies. What if we asked: “what problems is the institution of private property protection an answer too?” Perhaps if we had the answer to that question then we might have a better idea as to whether it is transferable to some other populations. For example, does the poor country have the same problem that the rich country had when it implemented private property protection? And, if so, is private property protection a better solution than the institutions/solutions already present in the poor country for dealing with that problem? And is the receiving country capable of offsetting the side-effect-problems the private property Institution/solution creates? In other words, using Jerven’s definition of institution may be the answer to the riddle as to why institutions do not seem transferable between different economies. Most “problems” are relative. But what if there are “universal problems”, like maybe survival. Then there should be some universally transferable Institutions. Even if that is true, any problem is likely to have more than one potential solution. So, in theory, multiple institutions/solutions could all lead to productive outcomes and vice versa. Additionally, any solution is likely to carry with it a unique set of problems/side-effects. Each of those side-effects is a problem that is amenable to some kind of solution [=Institution]. That chain of problem->Institution->side effects->Institutions->side effects…could repeat infinitely outward in multiple dimensions. That might explain why even ideal-appearing solutions may have unattractive results in the short term and why even catch-up economies take very long periods of time to reach the technology frontier.

Alan Clift
Jun 28 2015 at 2:55pm

Wouldn’t the timescale be longer for the topic? Use a few centuries, instead of decades.

[broken url removed–Econlib Ed.]

Cowboy Prof
Jun 30 2015 at 12:11am

I am a bit late to this party but I listened to this podcast twice and one section that seemed to puzzle me a few times.

When discussing private property around the 30 minute mark, there was an assertion that if private property rights were promoted a farmer would have an incentive to get a cow and a plow without fear of the state taking it. However, it is further stated that this might be inappropriate since the cow might get sleeping sickness. I fail to see the connection here.

If there is a known and relatively high probability of sleeping sickness in an area, why would a farmer want to make that investment irrespective of private property rights? “Sleeping sickness” killing a cow would be akin to a situation wherein there is an “arbitrary taking” of the cow by a state that does not credibly guarantee private property rights.

In other words, assume there are private property rights and a 75% chance of your cow dying of some disease in the first year. A rational expected utility calculation would probably lead to a decision not to buy a cow even if the state guaranteed property rights. Why, as Jerven seems to say, would private property rights incentivize a farmer to buy a cow in that situation?

Jul 1 2015 at 6:26am

Very happy with the “prairie” metaphor and the acknowledgement of ecology. The” Equator runs though it ” causes thin topsoil and promotes tropical diseases among other local factors that are well recognized.

Just ran across another tropical curiosity: The mitochondria in the human cells, the little energy packets, are also different. In the tropics each mitochondria “tries” not to generate excess heat. In the polar regions mitochondria “try” to produce excess heat. The basic physiology is different. Another factor for institutions to consider. It is always more complicated than that.

Jul 1 2015 at 10:24pm

Bogwood, That bit about mitochondria sounds really interesting. Can you cite your source? I’d like to look more in to it. Thanks.

Jul 6 2015 at 1:14am

One thing I especially appreciated about Jerven’s perspective is that he sees it as a deficiency that so many economists working on Africa don’t work in Africa. I lived and worked in Lesotho for four years, and while living there it irritated me to no end how people abroad could have strong opinions about these locations and cultures they’d never physically encountered. I suppose I felt this way mostly because after I had physically encountered the corner of Africa where I was living, I had a number of ideas I’d previously gleaned from books upended.

Comments are closed.


EconTalk Extra, conversation starters for this podcast episode:

About this week's guest:

About ideas and people mentioned in this podcast episode:Books:


Web Pages and Resources:

      • Economic Institutions. College Economics Topics, Library of Economics and Liberty. What is an economic institution?

Podcast Episodes, Videos, and Blog Entries:



Podcast Episode Highlights
0:33Intro. [Recording date: June 4, 2015.] Russ: What's wrong with the economists' consensus view about Africa and its economic performance? What is that consensus view and why is it wrong? Guest: The most fundamental mistake that mainstream economists have made is that they have been focused for the past 2 decades on explaining something that never happened. Economists have, since the 1990s, been focusing on explaining why economic growth has failed in Africa, or why there has been a chronic failure of growth in Africa. And moreover it has moved away from explaining slow growth to asking why these African nations have failed permanently, as in books such as Why Nations Fail. However, the problem and the puzzle is that, African economists have, if you look at growth data sets that we have available, not only have African economies been growing for two decades, since the mid-1990s, but African economies have also been growing in the 1950s and the 1960s and the 1970s. So, a basic fundamental thing that we are stuck with after two decades of research is that economists can explain why African economies are not growing; the problem we have, which is a positive problem, granted, is that African economies are growing. So, in that sense, economists have gotten it wrong and have fewer lessons than they would like to have for policy, and so forth. Russ: And we're going to get into that in detail. The misperception that you are trying to correct in this book is certainly one that I--not just most economists have, I think most everyday educated people have. They think of Africa as a backwater, economically, as having been "left behind." And is stagnant. And one of the pleasant takeaways from your book is that that describes one period in Africa's economic, recent economic history. But it's certainly not an accurate picture of what's generally going on. One question I had, which I didn't see a lot on in the book--there is some--is how much variation there is within Africa during these periods. So, you think of three different stylized, times that we should think about: sort of pre-1970--1950s to 1970s; and then 1970s to 1990s, and then mid-1990s to the present. So, it's growth, no growth, then growth again. But do those pictures--does that summarize accurately, in any one of those periods what's going on with most or all African nations? Guest: Well, there are different ways of approaching this. You could look at the average growth for a sub-Saharan African GDP (Gross Domestic Product). You can look at the number of countries that are growing above a certain rate. You can look at the percentage of total population in sub-Saharan Africa that are living in a country that is growing. And you can--rather than to rely on annual data, you can average this for 3 years, 5 years, 9 years, and so forth. So, it's not as if--and we are talking about 54 different nations. So we are doomed to generalize a little bit. And it's surely--sure that today many and most African economies are growing, but that does not mean all of them are growing. So, the kind of--so, in a sense what I'm saying is that the stylized fact so far has been that a chronic failure of growth. And that there has been no improvement in GDP per capita to speak of. I seek to replace that with another stylized fact that has its errors and that as well: but I seek to replace it and suggest that, you know, that we should approach African growth as recurring. And that means that most economies, on average, were growing in the 1950s, 1960s, and 1970s. And for most economies, growth has been coming back again in the 1990s until today. So we are approaching another 2 decades of growth. But somehow, the economists' frame of explaining and approaching economic development and growth in Africa have been stuck with this 'lost decade' of the 1980s and 1990s, which have shaped how the stories with sub-Saharan Africa and how what we seek to explain.
5:45Russ: I'm asking, I guess, a different question, which is: Is it meaningful to talk about Africa? Is there an aggregation mistake there? Do they have sufficient problems, advantages, disadvantages, excesses in common that it makes sense to talk about the African economy? The African problem? Or the African track record over certain periods. What do you think about that? Guest: Well, that's a very good question. I think you can ask the same question about Europe. You can ask the same question about Asia. But then Africa, it's been that kind of, whether it makes sense to talk about Africa and African economic development or not, has been a dominant question that Africanists and development all over Africa keep returning to. Africa is not a country. Because of many--lots of stuff that has been written on Africa and African economic development is written from afar. Some people are not on the African continent. There are often then these kind of mistakes, generalizations, and so forth. I think it's fair to say that there is as much variation within Africa as there is, within Africa as there is vis a vis Europe and so forth. So that's very, very important to keep in mind. And I think one of the things I'm also--you might dive straight into that--is that actually this approaching Africa as somehow a unit and that Africa has some particular character flaws or something like that, is indeed a function of how econometric models starting using the 'African content dummy variable'. Russ: Before you do that-and that was my next topic--so let's back up a little bit and remind listeners who are not economists how economists would attack, statistically, the question of what could explain the variations in growth, say, across African nations. Or the variations in the level of income--not the same thing. People mix those up all the time. But many, many, many studies have been done trying to explain differences in growth rate, using both the world as a whole or Africa itself. First, let me back up--back myself up after backing you up. Describe that process, how that works; and then you can talk about what's wrong with it. Guest: So, basically, economic growth is pretty straightforward. It's improved, increased ability to produce, and/or consume. We know that it's a result from combining capital and labor and the rate of technological change determines how fast our ability to do so. Growth, and that's what economic growth is. And then it's tempting to try to not only theorize about what causes growth--but those of course empirically tested. And that's what--with availability of GDP and growth data sets for a lot of countries, together with data on labor stock, capital stock, human capital as measured by education, educational attainment, perhaps government intervention measured by budget surplus, and other types of measures such as whether countries, capitalist or socialist and so forth like that. With these kinds of empirical data economists started to test whether, what would be, not only explaining why some countries are growing slowly and other ones are growing rapidly, but also trying to estimate the coefficients and the relative importance of capital, labor, whether you are a socialist, whether you are a capitalist, and so forth like that. So, the early--the model where I kind of focus in on here is the workhorse model in economic growth testing, as proposed by Robert Barro, first in an article and then in 1991. Russ: Former EconTalk guest. Guest: Yes. Russ: With whom we'll be talking about many. It's an interesting review. Go ahead. Guest: Yeah. And so, he proposed this model where he has economic growth on one side. Importantly, average economic growth, 1960-1995. And then he added kind of the usual suspects: labor force data, capital stock data. And also added stuff like human capital, a national--to what extent currencies were overvalued as measured by black market premiums. And so forth: assassinations per capita to measure stability and so forth. So, add a different kind of variables that we are supposed to quantify and express things that might be important for economic growth. And once, having added all those variables, and there are still a lot of things not explained. So that you also add the variable that is called the 'African dummy variable.' So, a dummy variable is something can have two values--either 0 or 1. And in this sense, in this particular example, the African dummy variable too the value of 1 if the country was on the African continent, and the 0-value if it wasn't. And once they computed all these date into the regression framework, Barro found that the African dummy variable was large; it was negative, and it was significant. And he interpreted that as that so far, there was something, particular characteristics of African countries that caused slow growth, that has not yet been properly quantified or captured. Russ: As you could [?] a different variable. I just say from the non-statisticians in the crowd. There's nothing explicitly pejorative about the phrase 'dummy variable.' That's a standard nomenclature that has two values of 0 and 1 or takes on discrete values, generally, other than continuous values. So, it's a little bit awkward, though, when you talk about the 'African dummy' when we are trying to blame Africa or explain something is missing or lacking in Africa that's showing up in this statistical measure. Guest: Yeah, that's correct. And so then a lot of research then focused on trying to eliminate this value, this variable. And find other characteristics such as ethnic fragmentation-- Russ: corruption-- Guest: Corruption, although ways that could somehow, maybe being close to the tropics-- Russ: Being landlocked. Without a port. Guest: Exactly. Different types of variables were suggested. Indeed in 2005 a review of the economic growth literature thus far, by Johnson, Durlauf[sp.?], and Temple [?] found that there was so far found that there were 145 different explanatory variables that were found to be significant. And which then reckoned as supporting no less than 43 conceptually different economic theories of economic growth. Different explanatory variables that were found to be significant, and which they reckoned as supported no less than 43 different conceptually different theories of economic growth. Which is--
13:28Russ: You could argue one of the great discoveries of economic science. But I suspect many people would conclude the opposite. This is an embarrassment of riches, literally, to have that many things that we think are statistically correlated with growth suggests maybe we don't understand it at all. Guest: And that's the problem. So, that's one way of kind of innovation in this literature is to find some kind of data set for which we have observation for many enough countries that we can run a regression and check if there is a correlation. And then kind of make up a story--ideally make up a story before doing the testing and to find out why this is related or not even related but even also sometimes causally related. So that many assignations may capture civil turmoil in a country and therefore civil turmoil is not good for businessmen and capitalists and so forth and therefore might be a causal factor in why certain countries are growing slower than others. So that's the--and I want to pause here for a second as well to point out that, to return to--this was, in year 2000, each of the economists-- Russ: I lost you there for a second. Say that again? This was in the year 2000? Guest: Yeah. In the year 2000, Magnusen[sp.?,], The Economist, that painted, had a front page showing that the African continent as the hopeless continent and then on the tutorial line as those in Africa have some inherent character flaw that keeps it backward and incapable of development? So that was the editorial line of The Economist in the year 2000. And this, The Economist, did not come up with this frame, this explanation themselves; they were then taking impetus directly from this literature for a decade or so, trying to find out what was the particular African growth problem. Russ: One of your points is that, one, that by 2000 there had already been some growth in Africa that had been ignored in the 1990s. Your other point, though, is that if you are going to average economic growth over a 3-decade or 4-decade period, you are missing all the changes that are going on in between any one year, going on between decades. And as a result, you are measuring nothing. You are just bizarre. Guest: Yeah, no, it's, you know--the research question was: Find something that is correlated with slow average growth from 1960 to the 1990s. Which is a very different research question than to find out what can explain rapid growth in the in 1960s and 1970s and then a decline in the 1980s and 1990s. You would be looking for completely different types of variables. So it's very important here, what type of stylized fact-inadvertently, I would say, you know, Barro was just trying to propose, say, a way of organizing this new statistical material. He went with this average growth thing. But that also happens--there is path dependency in scholarly work as well. So many kept on hammering on that saying '[?]'. So you could have thought that this was the moment--that economic growth was returning and so forth, [?] places. It was also at this point that the World Bank and the IMF (International Monetary Fund) maybe looked back and saw, 'Hmmm, maybe some of our structural adjustment programs have the effort to improve policies and so forth like that, did not have the results we wanted, and so forth.' Could have been a moment of reflection. But instead, the literature leads into what I call the 'Second Generation of Growth Literature,' where it's taken as a stylized fact that African economies have not grown and that there has been a chronic failure of growth. And therefore taking like step that you don't need to explain growth at all. So that's what you were hinting at in the introduction, as well: you get a new type of literature which is not focused on explaining growth, but explaining differences in GDP per capita, measured today. So then you've got this other type of literature which is most famously summarized in the recent book by Acemoglu and Robinson, in Why Nations Fail, also previous guest on your program. Russ: Daron Acemoglu, yeah. And where there is a different kind of search for variables. And this time it was increasingly perceived that one of the problems of this cottage industry of growth regressions was that you could possibly--and I do that in Chapter 1 in the book as well--show how plausible it is to dismiss most of these as just correlations, anomaly correlations. So, some of the basic problems, for instance--there have been arguments about that--but maybe having a high aid dependence is bad for economic growth. Receiving a lot of aid might distort the way you do policy, and might then therefore cause slow growth on average. Russ: By the way, I've never heard this, but I would call that the 'spoiled child model.' It says basically if you keep giving your kid money from a trust fund or your income, the kid doesn't have much incentive to work on his own, and he develops bad habits; and then his career is a failure--his economic career. And that's sort of what this argument, this literature says, right? It says by becoming dependent on foreign aid they lose their incentive to grow their own economy domestically, and they are somehow ruined. Guest: Yeah, that's one type of argument famously put forward by, in France, [?] aid and so forth. And so, but you could equally argue, if you look at the data, it might be the other way around. Maybe countries that are growing slowly get into trouble and therefore they are in need of a lot of aid, and therefore they receive a lot of aid. More pertinently, if you start looking at the timing of these things, you'll see that, you know, one basic kind of rule, not a kind of an absolute rule about cause and effect, is that cause should precede effect. Not the other way around. So, for instance, one of the problems is that the high aid dependencies a phenomena and an observation that derives from the 1990s. So you can't plausibly argue that high aid dependence in the 1990s has caused slow growth in the 1960s and the 1970s and the 1980s. It just doesn't work that way. And this same view-- Russ: You'd have to be very creative. You'd have to argue that nations that look forward to the possibility of getting aid stopped investing, because they figured if we get really poor we'd get more aid. So there is a story, but it's a bad story. I'll agree with you. Guest: Yeah. You could do some repeated games, game theory kind of thing, I suppose. Russ: Rational expectations. Guest: Yeah. But even more so, I think that it a problem as well with more famous types of correlations is that poor countries or slow growing countries have bad institutions. I mean having institutions of low quality. One thing is we can think that institutions is important for economic growth and so forth, but how do you go about defining what good institutions are, and good governance--what is that and how do you go about measuring it? It's an idea that, to start measuring this is a more recent idea. So, in the regression literature they did find that have low governance quality rating was correlating with having slow growth. But once again, these observation were subjective surveys of businessmen taken in selected countries in the 1980s, long after these economies have experience similar economic growth distress. And therefore you would think it was an effect of economic decline and not a cause.
22:28Russ: I want to stop there because I think it's really important. You are arguing that--I actually want to back up. Because we talk about institutions a lot in economics, and that phrase crops up on EconTalk and occasionally listeners will say, 'What does that mean, exactly?' So, we're going to talk about this for a bit. I think it's an interesting concept, the role of institutions in economic performance. But it's a slippery concept. And your point is that not only is it slippery, not only is it hard to measure and create a statistical variable that you can measure to try to isolate its impact, but it can also be an effect rather than a cause. So that's all true, and I think that's great, and I want to add before we forget, because I want to make sure we say it, in all these studies there's a problem that we talked about the last time you were on EconTalk, which is--and you talk about it in the book as well--the quality of data on what we're trying to explain--the GDP, the level of income--is very troubling in terms of its accuracy. But I want to put that to the side--but I wanted to mention it, in case we didn't get to it. But I want to ask you to talk a little bit about Douglass North and Daron Acemoglu and James Robinson and others mean when they talk about institutions and what they mean by "bad institutions" or "good institutions." Guest: Yeah. Institutions were defined most popularly by the Nobel Prize winner Douglass North as the 'rules of the game.' So there are rules that are conducive to economic growth or not--this is the big question empirically. And then Acemoglu and Robinson in Why Nations Fail built on that, and a lot of other scholars build on that basic intuition that it's not only capital and labor that matters; there's also how society is organized also matters. Russ: Can't argue with that. By itself. It's true, at that level. Guest: I think--the intuition there is, I think, so that everyone agrees with it. I think it's very hard to contest it as an intuition. I think the problem is, is to define it and measure it precisely. Jeffrey Sachs, for instance, in a debate on this has a nice insight on this where he says, 'One thing is to look at today and say, find out kind of like doing your horoscope, knowing already what's happened. Saying that you know, look, these countries are successful; and look, they have these institutions. And look, these are the good institutions. But try this mental experiment instead: Go back to 1960 and then come up with a definition of institutions that predicts what's going to happen.' And that, I think would be difficult to do. And moreover, even if those definitions were solid, it is also hard to think of a way of measuring institutions that is appropriate from country to country. And I think that this new literature that I call the second generation of growth literature that focuses on finding that history matters or institution matters suffers from some very, very particular and even paradoxical problems. They say that institutions matter. So that means the way that societies organize matter. Yes. Fine. Good enough. Let's take the next step in that: is then to say, let's find a variable that can measure that. And then you end up running a regression where institutional quality, institution has somehow lost its real definition but is now measured on a scale from 0 to 5, where 5 is perfect institutions and 0 is horrible institutions. And then Germany somehow has 4.8 type of institution, whereas Tanzania has 1.2 institution. And if you are a policy maker you might read that and think, 'Okay, so we need to make our institutions more like the German ones.' Which is completely the opposite of the intuition to begin with. The intuition is that institutions matter; how the society is organized matter; and that building on the work of North, that these institutions are part of the political process. And they are suitable to a particular condition and are not to be exported from here to there. So you end up having an intuition which I agree with but in the measurement process of these institutions you end up throwing out the basic intuition to begin with. Russ: Yeah; it's a fascinating point. And I've started to--it's come up a number of times here and it's something I think about a lot. It comes up sometimes when we talk about education policy--you know, why can't we be more like Finland? And the answer is: There may be things about Finland that are worth emulating, but there may be things about Finland's success--which has been not so successful lately but we'll put that to the side--we can learn from. But it's very possible that the things that make Finland successful can't be imported to another country. And worse, if they were imported, they wouldn't work very well. Which is--it's not just that it's not even realistic. It is realistic. But it's not going to work. And to assume that everything is malleable that way--I think we made a leap in economics, a first leap, that was very good. And then we leaped over the cliff. The first leap was: Everybody responds to incentives. So, there's no really different economics for Africa than for America. Guest: Yeah. Russ: And there's truth to that. I think people in Africa respond to incentives just like people in the United States. But then when you make the next leap, which is, 'And therefore, the such-and-such policy act of 1953 would have worked great in Africa because it had incentives.' And that just doesn't follow. Guest: No.
28:36Russ: In fact, the thing I liked--I think--this is a very small thing, but it might be my favorite thing in your book; I thought it was really utterly fascinating--was your story of Congo. You give the example of Acemoglu and Robinson, I think it was in Why Nations Fail, maybe it was in a research paper, about trying to explain why agricultural productivity in Congo was so low. And one answer is: Well, they don't use the plow. Well, that seems like a terrible mistake. In fact, a lot of people would say the way to improve agricultural productivity is: We need to give more plows. Or more threshers. Or: we've got a great agricultural system; ours is very productive; let's give them ours. Let's transplant ours to theirs. Now, talk about how Acemoglu and Robinson explain why there's not as much plow use in the Congo and what the alternative explanation is. Guest: Yeah. They think that this is about institutions. And they think it's about protection of private property rights. And that because property rights in this sense being, knowing that the farmers knew that if they made the investment of plowing, getting a plow, getting a cow and getting this thing plowed, then harvested it, they would know that, you know, that the state will protect them from getting their crops stolen. Or they would also know that the state would not tax them too much; they will get their fair share, and so forth like that. Then these investment would have taken place and the Congo would have been a richer place. That's the basic idea. Russ: And the converse. So, if there isn't good property rights, if you have bad institutions, then there's no incentive to invest in a plow because you'll just get it stolen. You'll end up poorer than you were before: you'll have saved up money, made an investment, and you'll get nothing for it. Guest: Yeah. So that's the--they tell that story in Why Nations Fail to kind of explain how Congo could have had different outcomes if it didn't have extractive institutions but instead had what they call productive institutions. Russ: And I have to confess, Morten: it's a very appealing argument to me, because I like property rights; I think they are important. But you make a great observation: there is an alternative explanation. Guest: Yeah. I mean, first of all, you know, having properly secured property rights is not the same as having properly secured private property rights. Those things are sometimes conflated in the literature. But leaving that aside, the basic problem they have here is they are looking at a place where land is abundant. So that means that even it, to just to start--when land is abundant, the supply of it is abundant, then that determines the price, what will then be close to zero. Therefore there is no incentive for the state, nor the farmer himself, to work out some agreement how to agree upon the private property rights here. So therefore, just having the private property rights in itself isn't useful. And even if that was taken care of--let's say, that was taken care of, the property rights--then the farmer would have the option to get the cow and plow. If he would do so, then he would then bring a cow into an area where there is sleeping sickness. The sleeping sickness, this parasite, would attack the cow. And the cow would then die and you might also be affecting your whole family and you might also suffer as a result. So, even if the private property rights, and there it is absolutely not the rational technology to introduce the cow to plow. And even if there was, even so, that there would also be the problem of the land being abundant. So why would you plow in a land, where you'd much rather take advantage of the fact that land is abundant and plant around trees and to do, burn down, and so forth, and do the most efficient way of operating in a land-abundant context. Russ: I think you[?] also said that plowing itself wasn't that productive, given the nature of the soil and-- Guest: Yeah. And finally ask: Colonialists did discover in southern Africa, when they did introduce the plow and they did plow, soil fertility in tropical areas is shallow, so that if you do plow and there are rains afterwards, then the land is ruined forever. So, the lesson is, here, that while we think of one technique or technology is proper in one place, it does not travel very well. Another good example made by people studying Asia is the way that they think about technology progress being rational in America and Europe where they introduced tractors to plow and thresher in wheat fields. That's not the same type of technological change or capital to labor ratios that is rational for a rice field in Bali. A tractor would just make a big mess of the rice field. I think when you think about institutions more down-to-earth kind of things, about like how to do things, it suddenly reappeared to you: No, we can't possibly think that there is one way of doing things, and this will somehow work in all contexts.
34:06Russ: Well, this wouldn't be the first time that economists made policy recommendations or tried to understand something and through lack of their local knowledge made a blunder. But I think the tougher question, and I note that was a beautiful example--but the tougher question is, Acemoglu and Robinson, to give them their due: there is something relevant about property rights, in general, right? In this case maybe it wasn't so important that there weren't property rights in Congo. Maybe that isn't what explains the low agricultural outputs of Congo. But in general, it would seem reasonable that property rights would matter in determining growth. And, if you had a dictator or an autocrat or a thug at the top of the political food chain who was exploiting the country's citizens and extracting rents from the country's productivity, one would--I find it plausible--that that would discourage investment and growth generally. At least to start. So, given that that intuition makes sense, is your complaint that--is your criticism of Acemoglu and others who blame institutions, say, is your complaint that their underlying argument is wrong? Or that they apply it inappropriately? Guest: I don't think the underlying argument is necessarily wrong. I do think that most of this intuition makes sense. We know a little bit about, a lot of, people respond to incentives, what you said before is very true. One of the big discoveries in African economic history was to find out that people in poorer countries were not irrational. They were sensible capitalists who make investments according to market opportunities. So I think that that basic intuition, I don't disagree with. What I do disagree quite strongly with is the way that it has been empirically measured. Because as I said, it's quite a different thing to say that institutions matter, to go along and start measuring it on a scale from 1 to 3, because these measures are not very good, and they might also introduce bad behavior in terms of people who are trying to behave according to what will be measured as a good thing, on the institutional capacity and so forth like that. It's also--so I disagree with the way it has been--I think that the argument is more convincing stated as a theory than when it becomes empirical. I also think that they overstated the empirical results, so that [?] misled to think that if you now introduce private property rights in Tanzania, well, a). that it would make sense, and b). it would cause economic growth. And I don't think that having good sets of governance, democracy, protection of private property rights doesn't have to be defended as a cause of the economic growth. You can just define investments or reforms to improve governance quality, democratic rights--in its own right, without justifying it with economic growth. Now it does happen to be the case that the way that literature has gone out very strongly to say the reasons why these economies are poor is that they don't have the right institutions. And you know, so that they implicitly say that, you know, 'If Tanzania changed their institutions, they will be as rich as Germany.' Or 'Sierra Leone will be as rich as United Kingdom if they will have the same institutions.' And I think that is a massive--not only is there a massive reversal of causality in play here, that, you know, I think that rather the development of institutions and economic growth and all of that is a part of complex development where it is very hard to actually depict what is the cause and what is the effect. And you might also be inadvertently pushing for the kind of, like, 'Why doesn't Tanzania become more like Germany?'-type of argument. Or 'Why aren't you Denmark'-type of argument. And that comes from the way in which they analyzes its frame [?]. In the book I talk about the distinction between doing things according to the subtraction approach and doing things as a reciprocal comparison. And by the very nature of the game, here, looking at global data sets of GDP, global data sets of the institutions and so forth, what has been done is that in effect you put rich countries on one side and poor countries on the other; and then take [?] and found that the explanations for why poorer countries are not rich is that they are, have different institutional characteristics than the rich countries. And so that is, I think is, the subtraction approach. And then pretending--very much backed into the Karl Marxian view of economic history--that for a poor country, the history of a rich country presents it's future. Yep. So this is all a ladder that all countries should grow. Whereas actually, if we do believe that history matters and that therefore these institutions matter and therefore there are different ways to getting there, that's actually paradoxical. It means that the type of research that is conducted in this, what I call the second generation of growth literature, this literature that looks for correlations between income today and in historical data sets, should be abandoned in the favor of deeper, more contextual studies of how actually land, land right system works in Tanzania. Rather than just go ahead and assuming that right now Tanzania would be richer if there would be automatically much richer and have high growth, if we managed to introduce a private land typing[?] across Tanzania. There are many reasons why Tanzania doesn't have a private land typing system in place now. The most important one is the same as made in the example of Congo. The land plots are small; the transaction cost of titling the land is bigger than the value of the--over one year crop. And even if that transaction cost was low, it would still be some kind, have to be some kind of institution that could somehow use the land title so that we[?] get credit. Or that there might a state that was able to tax them, and so forth. But right now those conditions are not there, and therefore the rational outcome is not necessarily to introduce the same set of institutions which has worked elsewhere.
41:16Russ: Yeah. I just sort react to that. I think it's--I think I agree with you. I agree with you very deeply about this. I think when you take such a broad brush approach to saying things like, 'Well, it's bad institutions,' or 'It's bad governance,' you can't summarize that in a paragraph or two when you are talking about a particular country. And you can't summarize it in a variable in the right-hand side of a regression. It really has to be a deeper, more thorough study. The problem is, once you head in that direction, you tend to start saying, 'Well, every country is unique. Every country's history and context is unique.' Which is true. But then the question is, as economists, we tend to want to learn general lessons. We have a deep urge to find general lessons that would lead to policy recommendations. Which is probably not always a good urge. But I do want to raise the question that--I'm attracted to the conclusion--so let me try to give a thumbnail version of what we just talked about. If I want to understand why some nations succeed more than others: Sure, there's a correlation between success and failure and various institutions--culture, norms, trust. These are all vague, difficult things to measure. More importantly, they are very difficult things to create. So if I say, 'Well, the reason that the U.S. economy does better than, say, Russia's, is we have this created[?] advantage. It's not that we have a better legal system, and we have a better culture--we are more likely to trust each other. And then you say: Well, what can Russia do then to be more like the United States, or more like Denmark, or just to trust each other more?' Well, there's no policy recommendation that comes out of that. And I think--so one of the things that I have learned in talking and thinking and reading about this is that a lot of the obvious policy recommendations of "more property rights" or "better institutions" are very difficult to implement. Very difficult to export. Very difficult to import. Very difficult to implement. But then where do you go from there? And one answer would be: We don't go anywhere. Because that's all there is. The world is a complicated place. You can't easily solve a nation's problems with some magical reform that you can impose from the outside. And I'm drawn to that, although I have to say, it's somewhat depressing. Because it seems to say, 'Well, you don't have to worry about them because you can't help in any way. Your attempts to help them are going to have unintended consequences.' So you bring in the malaria nets, just to come back to an old EconTalk idea, and, I read recently a lot of people in the countries that are receiving the cheap malaria bed nets use them for fishing. Which is totally rational, because they have problems catching fish; they were near starvation; and they are more worried about feeding their kids than they are about getting malaria. So, when I see those kind of results, it tends to make me extremely cautious in having magic solutions. But is that all we have? Is that where we're left? Is there nothing--no advice we can give? I'm okay with that if that's true. But with your opinion, where does that leave us? Guest: Well, I think, you know, I think Douglass North put it nicely, according to when he, shortly after he received the Nobel Prize, he was asked whether he had any, putting forward if the institutions matters, and so forth, asked whether he had any advice for Russia. This was in the 1990s. And he replied: 'Get a new history.' Russ: Yeah. Be more like Denmark. 'But I'm not Denmark.' Guest: Yeah. Russ: Saying if I want to be--yeah, provisional basketball player, be 6'6". And I'm not. Guest: Exactly. So there is that. And some side issue I should also--there is a reason why I am being greedy[?] with this literature for so long, as well, is that I am fundamentally fascinated by this literature. I think the questions of wealth and power, [?] of nations, and these big broad patterns of global economic history, I think that's interesting to read. And I think it's a fascinating research program to try to unearth search patterns and try different pieces and so forth like that. But I think it needs to be done with some kind of humility. And which I hope--which I think has been lacking. And I think also, I think that--I'm not saying that all of this type of--so fundamentally, what I wanted to do with this book is to, you know, to provide a way in which students and scholars and people working in development and non-economists, perhaps, how they can engage critically with what seems to very robust empirical findings for what matters, for whom, for growth and development in this world and how to read this economic literature critically, I think. Because I think that's important as trying to counterweight some of the very self-assuredness, some of these conclusions about what matters and what does not matter for economic growth. It's also, I think it's not as if finding law-like statements for what works in economics for all countries of all times is the only way to conduct economic research. There is also more, better ways of--one of the things which I am puzzled by in the new book about African economics, studies of African economic growth, is that you do not longer study African economic growth. They have cites that this issue altogether be, one of the earlier literature to African economic history as starting in 1960. Although there's an artificial Year 1, right? There is a longer story to be written. But the main--if the boundary of investigation is set by so far quantified, then you get artificial ways of thinking about economic growth. There is also--there are some, yes, there might be limits to how many historical data sets it's useful to run correlations with, with the income per capita. I think we reached that limit some time ago. But I think also one of the problems here is that in the study of African economic growth and African economic development, there have also been fewer hands at work. Not that many people working material history. Which means that more difficult questions such as doing careful archival historical work to try to understand what, for instance, does Nigeria have a higher tax, do they tax a higher share of their GDP today than they did 10 years ago, compared to 1980, compared to 1960, compared to 1920? These are kind of empirical questions that could be answered, but instead I think the scientific rewards, in terms of the [?] economics discipline seem to be that there are bigger rewards for coming up with fancy models and more law-like statements about why tax system[?] is important for economic growth. Not having really data to back it up, nor being able to say something about how tax varied in Ghana or Nigeria over time. So I think one doesn't have to be only--I think one should be humble about what kind of universal solutions one thinks about. But fundamentally this is also about how we train economists. So, what is it that economists do their Ph.D. on? What are the kind of research findings they come out with? And are they able to tell us something useful about the economies which they purport to study? So I kind of launched this as a way of--kind of summarizing this is that I make call for, let's try to study economies instead of only studying economics. So that means that you might want to--part of that is the weakness of universities in African countries. But a lot of the research is done outside of Africa. But also with the coming of global data sets, growth regressions using global data sets, the distance between the observer--the economist attached to his computer with downloaded data sets--and the observed, the actual economy in question, has ever been increasing. And I think macro questions continue to be important. And sometimes economists get it wrong. And it's not all--we're not exactly sure I think exactly how useful economists are. But they are of some use. And it's unclear to me what use you have an economist who is able to run global regressions about which type of quality of institutions determines 30% of the income differentials between Tanzania and Germany, rather than ones who is actually good at dealing with, you know, what do we actually know about the field and what do we actually know about property rights? When was the last time we had the [?] Survey? Do they have any [?] statistics? And so forth like that. So what I am asking for is, I am re-focused toward the nitty-gritty hard work of the country economists of the previous years.
51:00Russ: Yeah, most economists don't like nitty-gritty. And I'm in that group, too. I understand that. I want to try to give a different version of your critique. The way I would think of it--and you can react to it. One of the lessons that I think of, when I think about these challenges of 'why aren't you more like Denmark,' or 'why aren't you more like the United States,' or whatever is the case, is that it ignores the organic nature of incentives and markets, that processes, institutions that we're trying to talk about here in the most vague and general sense, that they are emergent. We don't fully understand what has to be put in place. One of my favorite examples of this is: How do you build a prairie? Which I once read in a beautiful book on management, a beautiful image of management, which is: Well, we know what's in a prairie. It's a certain set of plants. But if I start with a bare patch of ground outside of O'Hare Airport in Chicago, which used to be a prairie, so I'm going to recreate that, I'm going to do very poorly. Because if I just sort of mix all the ingredients together, the right plants, because even though those are the right plants to make a prairie, I don't understand the process by which the prairie emerged. I don't understand--certain things had to be put in place first. Certain things had to come at the same time. That there's a dynamic, organic nature to a prairie that's also true of an economy and institutions. So, I take that--I'm very drawn to that for personal, Hayekian reasons. It seems to me to be very powerful. And the research agenda that comes out of that vision is one that says we ought to be studying how prairies emerge. We ought to be studying how economies emerge. And to do that, it's not magic. You can't just sit down and run a bunch of regressions. You have to do the nitty-gritty work that you are talking about. The problem for me, when thinking about our profession, is that the rewards, as you say, are not in learning just about how the world works--which is what I'm suggesting we should be doing more of. But the rewards are for trying to make the world better. And to do that, you have to come up with a set of policy implications that could say a country, increase it's income by 20%, merely understanding the history or emergence of institutions for example, people would go, 'Okay, that's nice. That's like history, anthropology. That's like sociology.' But what's the policy implication? And the answer might be, well, there aren't any. Because you can't say 'Get another history,' and that's the way the world is. But our incentives are always going to push us toward trying to say, 'Oh, and therefore, we need to do x, y, z.' Guest: Yeah. Though, you know, I like that [?] and I think that's exactly what I'm driving at. So that's a nice summary. And I think, exactly, to try to refocus African economic history, material history of Africa or African cultures, African economies, would be precisely to recognize that the African continent is not stagnant. Growth is recurring, as I emphasize in the book. And these periods of growth, while you might derive some policy lessons from periods of growth--we might learn more from studying why certain African economies failed and why certain succeeded--then explaining what did not happen in the countries and why they are different rather than to actually study what is emerging. Staying with how to build a prairie, economists have been explaining for a long while why a prairie is not a desert. So, they've been trying to explain why a prairie is not something else. I think that's trying to explain how African economies work rather than to explain why they don't would be the key message from me. I think that it's how--I wouldn't come away--it would be too bold for me to say exactly how, what should be the responsibility of economists to give policy implications. I think of one--in terms of how I think of economics as a discipline is being rewarded, I think there are some problems. For instance, in the sense that if you do write something, and I know this from personal experience, about Tanzania, the likelihood that an article will be published on Tanzania is equal to zero in the Journal of Development Economics. But if you study something which is saying something about Africa as a whole, generalizing, there would be a chance of succeeding. So in a sense, particularly the studies of African economics and so forth, there tends to be rewards toward overly generalizing. And I think there are goals from not only publication--those are like citation and so forth, like that. So then that's a real issue. Another one is also when economists do publish on--as I just said. Branko Milanovic[?] put it nicely, said that it's a paradox to be hearing that history matters and then be confronted with Wikipedia with regressions afterwards. So then that means if really history matters and really institutions matters, then is this still the domain of economists? At least, it should be a shared domain. [?] I think that many of these papers that are being published about how and why history matters, particularly on poor-income countries--if you do publish a paper on the effect of slavery on economic growth today, you might want to consult historians of slavery, not only think about the robustness of these arguments in terms of econometrics but also whether it makes sense as a historical argument. So that's a policy implication for the economics discipline, I think.
57:42Russ: I want to mention before we close that the book is very accessible to a non-economist. You do some very nice analogies and examples to help people understand some of the statistical issues we've just touched on barely in this conversation. And you go into them in a little more detail in ways that a non-economist or a non-statistician or non-econometrician would learn from. I want to put that plug in. The thing I want to mention, I don't want to miss, that I learned from the book, which I found just sort of shocking for me, is two--I think these are actually facts. One of the things you mention in the book is a lot of things we call facts, like how corrupt a nation is because it's a 3.74 on the corruption scale--these are I think close to actual facts, were surprising to me in ways that I wanted to get your reaction to. So, one of them was the remarkable amount of GDP that comes from trade in Africa. I was shocked, and ignorant of that. So, that number is 50-75% over the last few decades. We tend to think of Africa as isolated from the rest of the world, not particularly free-trade oriented. And yet a huge portion of the GDP is coming from exports. It should be emphasized--I think I have this right--that those are exports of raw materials. Mostly. Some production of stuff. But in general the other part that struck me, the other fact I wanted to combine with that is that the very, very, very low population density of Africa. How incredibly un-dense, really how rural Africa is. And it just struck me that one of the flaws with this entire literature--tell me if I'm wrong--is, it's a primarily agricultural--many of these are agricultural economies. They are not urban. There is not much industrialization. And as a result, we are applying lessons and comparisons across countries that just seem absurd. We wanted to increase income in Africa, it's obvious that we could force people, encourage people to move into cities, that measured income would go up. It's not obvious that that would make people's lives better. But a lot of what people are trying to do in improving African wellbeing, which is what really matters, is trying to impose, like we said earlier, models that aren't relevant for these lives. People living very primitive lifestyles in many of these countries, the ones who are near subsistence, they are very primitive. And the idea that we can somehow make them like a farmer in Kansas--if that was your goal, you should push them into the city. But that's not what they want. They can go to the city. They know there are cities. They evidently have aspects of the rural life that they like. So, I was struck by those two facts, the fact that trade is very high; I assume it's not in finished goods because most of the economies we are talking about are not particularly industrialized; they remain very, very rural--I conclude that based on the density numbers. And therefore this idea that they are like other parts that are not as developed is a total misreading of what the reality is. What do you think of that? Guest: Thanks for bringing those issues up. I think that the land density and the population density perspective is, if you wanted to point out something that makes most African economies special and have some special characteristics, there are some exceptions to that rule. Ethiopia has been densely populated; Rwanda and Burundi as well; and they have also very different state structures. The example which I made about the cow and the plow does not apply to Ethiopia. They already have the cow and the plow and they could also foot a proper army which beat the Italians when they tried to invade. So there are different types of states. But land--the typical pattern has been, thus far and remains to be true in many of these cases that we are looking at a place that is relatively land abundant. That means that these technological fixes to agriculture is not going to fly. It's not--some people ask themselves, 'Why hasn't the green revolution been implemented in sub-Saharan Africa yet, while it has been in India?' Most of that is explained by population density and land abundance. A green revolution only makes sense if, well, land is relatively expensive and particularly related to wages and related to fume[?] and so forth. So that's very, very important to keep in mind and which you may misunderstand if you come at this subtraction approach to the comparison and not understand institutions and choices from the perspective of the countries in question. I think also the trade question is also sometimes overlooked. If you try to draw some stylized facts, which I do in the book--I talk about maybe if you go back 150 years ago, 15-20% of the GDP was traded in Western Africa. If you go to the beginning of the 20th century, you are talking 20, 25%. This increases to about 50% in 1950. And then from 1950 to 1975, that share, which is still quite high, is constant at about 50% as a share of GDP. Meanwhile, GDP more than doubles. So both trade and GDP increase very rapidly in the 1960s and the 1970s. Then you have the economic decline, 1970s, 1980s, total GDP is about stagnant for two decades, as measured. And so is the share of export and import of GDP--it remains stagnant at 50%. Whereas in the past two decades, this share of trade in GDP has increased to almost 75% in the highest year just before the financial crisis, 2009. And that's particularly stunning when we know that total GDP has almost tripled in the same period. So we are looking at, not only is Africa not stagnant--there's been a lot of growth--but it also remains not on the margins; it remains deeply integrated in the world economy and has been so for centuries. And exactly how this growth interacts with openness and how institutions respond to whether these states are able to tax more of these rents and so forth, like that, are research questions that are currently slightly overlooked. That doesn't mean as you so correctly said, Russ, that we can just go around thinking that metrics such as GDP means the same thing in every place. So, a 2.5% GDP per capita growth in Germany might tell you something about the efficiency of the labor markets. It might tell you something about the application of technology. If you look above that you might find something about total factor productivity growth and come up with some explanation about why maybe right now Germany is better at implementing good new technologies compared to Finland. Like that. But a 2.5% GDP growth in Equatorial Guinea or Botswana or Tanzania where we have [?] where 75% of GDP is traded on the world economy, when most of GDP growth is determined by quantities sold on the world market and-- Russ: Subject to price variation that you can't control. Guest: Exactly. Yeah. So then to go on and build very sophisticated variables to come up with generalizable statements about what factor productivity was, globally speaking, simply doesn't make sense. You need to analyze these economies in a different way and to think about which part of this economic growth is important. Do I know something about labor markets? Should I use these labor stock data or not? Are they based on something real? So it requires much more fingertip knowledge and understanding of the economy in question than has so far been [?] in the past two decades.
1:06:45Russ: I just want to clarify that when I say that people perhaps want to stay leading a certain kind of lifestyle, I don't want to romanticize people living near subsistence--their children dying young. Obviously people very desperately want to escape some of those conditions. But it's also clear to me that, for various reasons which are perfectly sensible--people who grew up in a certain way are comfortable with that way and are not comfortable radically changing; not everybody is, not everybody can emigrate; as I like to remind people, luggage can solve some problems, just moving is very powerful. But it doesn't solve all problems. And it has costs. So, when we talk about people moving to cities and getting higher income, some things are lost. And they know that; they take that into account. I just think we should be respectful of people's choices as much as we can. Having said that, we desperately sympathize with people who struggle to stay alive and keep their children alive. And I'd like you to close by talking about some of the small things--not grand, doesn't get you, doesn't save the world in one swoop, but the small things that people are talking about that maybe would make a difference at the margin. Meaning a little improvement here, a little improvement there and that might lead to other things emerging alongside that. So, some of the experimental evidence that people are talking about and trying to improve--the use of cash. Any thoughts on those? Guest: I think Lant Pritchett has somewhere written or said that that kind of randomized control trials, the experiments, and so forth are giving very, very precise answers to the wrong questions. So that means that--I think it's sad[?] that those small, carefully conducted studies, certainly fulfills requirements of rigor and control groups and it's tidy. It's neat. There is the problem that--it satisfies the peer review way of how we want economics to be done. It can have it causally clean and so forth. It satisfies that requirement. It's in vogue in Washington; it's in vogue in London; and so forth. The question is whether it does give how many experiments do we have to run before we know that actually it makes sense to give people free lunch at school instead of subsidizing their uniforms or instead of vaccinating them at home. So before we know whether that only applies to one village in Kenya and one village in India, so how many tests do we have to run before we know it applies universally? But it's also--it's sometimes--there is very careful tests conducted about whether fertilizers work or not for a small-scale farmer. And if everything is controlled and fixed for researchers find that it is profitable to use fertilizers. But then the big question is the macro one, which is not answered by these tests. Does that mean that we should build a fertilizer factory? How much can we afford to import? Russ: If everyone uses fertilizer, are the conclusions going to be the same? Guest: Should it be subsidized? Should it be not? And so those things cannot be controlled for. It remains the domain of the very messy macroeconomics to figure out what's feasible and what's not feasible in this. And I think to answer those questions more carefully--I should add that I think one of the great benefits of this tendency towards economists to do randomized control trials as well is that it has recreated the field economist--the one that does field research. So in that sense it is useful. But I think there is still work for the field economist that knows what goes on in the Central Bank, that knows what goes on in the Statistical Office, and understands how decisions are made within the Ministry of Finance. And that's if institutions matter for growth then that's the kind of knowledge you need. Not to know that 30% of the difference between Germany and Tanzania is explained by some abstract way of quantifying the quality of institutions.