William Easterly on Benevolent Autocrats and Growth
May 30 2011

William Easterly of New York University talks with EconTalk host Russ Roberts about the oft-heard claim that poor countries led by autocrats grow faster than poor countries that are democratic. Drawing on a recent paper, "Benevolent Autocrats," Easterly argues that while some autocracies do indeed grow very quickly, a much greater number do not. Yet, the idea that the messiness of democracy is inferior to a dictatorship remains seductive. Easterly gives a number of arguments for the perennial appeal of autocracy as a growth strategy. The conversation closes with a discussion of the limitations of our knowledge about growth and where that leaves policymakers.

RELATED EPISODE
Bruce Bueno de Mesquita on Democracies and Dictatorships
Bruce Bueno de Mesquita of NYU and Stanford University's Hoover Institution talks about the incentives facing dictators and democratic leaders. Both have to face competition from rivals. Both try to please their constituents and cronies to stay in power. He...
EXPLORE MORE
Related EPISODE
William Easterly on Growth, Poverty, and Aid
William Easterly of NYU talks about why some nations escape poverty while others do not, why aid almost always fails to create growth, and what can realistically be done to help the poorest people in the world.
EXPLORE MORE
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.

READER COMMENTS

Jonas Kölker
May 30 2011 at 12:12pm

On bureaucratic incentives: as far as I understand the Henry George theorem, the value of the land rent equals the value of the public good(s).

Therefore, if the tax revenue is land rent (and land rent only), budget deficits or surpluses provide a measure of how big a gain or loss the public goods are to society.

It’s a small step from there to paying people based on their so measured performance. If the top brass is motivated not (mostly) by money but reelection, they should be paid in “artificial votes”. Then again, thinking about campaign financing, they probably are motivated by money.

Comments? Did I get the Henry George theorem wrong? Does the land rent include too much noise (e.g. private externalities)? Does it work to pay people based on broad outcomes if they don’t know how their actions influence their rewards?

David B. Collum
May 30 2011 at 1:28pm

Attributing growth to leaders certainly dovetails with the ideas presented in “Fooled by Randomness” and other treatises on the flaws of statistics.

I thought the suggestion “growth miracles almost never last” was the money line. One day I calculated the result of an inflation adjusted of 3% from the time of Christ–a pretty spectacular couple millennia overall–and found that it would lead you to 10exp28 compounding. I subsequently found that Adam Smith had done this same calculation (assuming 5% not 3%) in the 18th century and described the compounding as producing an “orb of gold” some millions of times the size of the Earth. Rule of thumb: We consume and destroy wealth at the same rate that we produce it; the bulk of what we have accumulate is knowledge.

I don’t know if Hayek is Hayekian or not, but I have suspected for some number of years that Keynes would not be a Keynesian as we currently practice Keynesianism.

W.E. Heasley
May 30 2011 at 1:37pm

Collectivism based upon central planning which leads to authoritarianism resulting in some cases as autocracies: can we state that of Easterly’s eighty nine autocracies, of which only nine are successful, that central planning based on Thomas Sowell’s propositions set forth in Visions of the Anointed, A Conflict of Visions, and Intellectuals and Society, has a temporary success rate of 10.1 percent i.e. (9/89)?

Stated alternatively, the autocracy based on elite intellectuals’ notions delivered through central planning has a 10.1% success rate. Meanwhile the success rate is merely temporary as all autocracies are short lived by historical standards.

David Ellerman
May 31 2011 at 9:32am

Whew! That was close. Managed to support democracy over autocracy as a growth strategy without mentioning new colleague Paul Romer and his Charter Cities program.

Gabriel Rossman
May 31 2011 at 10:37am

The “no such thing as a true autocrat” point reminds me of Ambrose Bierce’s definition of “absolute”:

ABSOLUTE, adj. Independent, irresponsible. An absolute monarchy is one in which the sovereign does as he pleases so long as he pleases the assassins. Not many absolute monarchies are left, most of them having been replaced by limited monarchies, where the sovereign’s power for evil (and for good) is greatly curtailed, and by republics, which are governed by chance.

John Rozewicki
May 31 2011 at 11:32am

On the subject of Japan and autocracy…it’s interesting that Japan was brought up as the only example given in that scenario of “strong leadership” that was not an autocracy. Japan’s industry has long been in bed with the political world dating all the way back to the meiji era with the Zaibatsu. A lot of those Zaibatsu actually still hold sway today in some form through the system of Keiretsu where large conglomerates basically hold each other’s stock in a loop of collusion. If you’ve heard of a Japanese company then chances are they are a subsidiary of the keiretsu.

This kind of collusion inside the business world, inside politics, and then between the two worlds creates actually something akin to economic autocracy in a lot of ways. So I don’t think it’s even functionally an exception in the group of “strong leadership” examples.

By the way, you do an amazing podcast Russ. You manage to walk a fine line between econ jargon and lay understandability. As someone not well-educated in economics in school your podcast has taught me more than any other about lots of different facets of economics. You do a great job.

John Berg
May 31 2011 at 1:03pm

Though someone else alluded to it, the two discussants have no justification for suggesting that probability or statistics, as they used it, apply. We’re back one podcast to the “blind spot.”
Like weather forecasters they can say, “70% chance of growth in the next 24 hours.”

Of course, one might say that a rule for Randomization of Governments was demonstrated in the paper.

On a separate subject but related by the degree of authoritarianism in the current US administration, is there a way to relate the increase in Doctor wait time to the $500 Billion removed from Medicare by Obamacare? One realization of wait time is the way lines are rendered orderly at Amusement parks. The controlled sequencing of customers back and forth within those piped restraints should be capable of quantification.

John Berg

Mariusz
Jun 1 2011 at 10:21am

You guys have me totally confused: we do not know what to do in the short run but we know what to do in the long run (I thought we are all dead in the long run) and at the end of the day the advice is “get local people who know how to play the political game of actually making things happen in this society and know what the relevant institutions and political constraints are. And then have them apply good economics to the society.” How do you do all that? Who are those who will make sure that “good” economics is applied? How do you know that you know the people who know that they know how to do that? I watched “Too Big to Fail” on HBO recently and it was clear (ha, ha) to me that even here in the US the people who may think they know they did not know. Does Bill have any thoughts on on that?
BTW Russ for a change of perspective you may consider interviewing Bernard E. Harcourt, the author of “The Illusion of Free Markets: Punishment and the Myth of Natural Order” http://www.amazon.com/Illusion-Free-Markets-Punishment-Natural/dp/0674057260/ref=sr_1_2?s=books&ie=UTF8&qid=1306937679&sr=1-2

emerich
Jun 1 2011 at 5:34pm

Easterly’s thesis is an excellent corrective to the (Tom) Friedmanite view that autocracies like China can do all the right things and usher in ecnomic Nirvana. But doesn’t his agnosticism about what does and doesn’t work go too far? Can’t we still look around and see what policies work better than others? Look at all the lab experiments of the late 20th century: East vs. West Germany, North vs. South Korea, Hong Kong and Taiwan vs. PRC. Is it illegitimate to draw conclusions from history and observation? Is the difference between good and bad economic performance really so utterly mysterious?

Seth
Jun 2 2011 at 12:15pm

Re: top-down thinking beginning in the womb.

I remember having faith in grade school that some government aptitude test would tell me what occupation I should do. I lost faith in it when I saw the list of possible occupations.

Re: the performance of local governments

I believe one reason local governments do better, even though fewer people are interested in local elections, is the the market feedback of competition.

I have the choice of twelve municipalities within a 10 mile radius of my current home. Go out 20 miles and the number of choices is nearly and order of magnitude higher. If my municipality does a poor job, I can move to another one. It’s easier for me to move than to try to effect the outcome of elections and change things for the better through the political process.

It’s not as easy to move to the next Federal government.

John Berg
Jun 2 2011 at 11:33pm

Perhaps I stated it too off-handedly but is there a way to assign a cost to the wait times increased by the $500 billion taken from Medicare? It is not clear what benefit is provided to gain the monies removed. Though it is not unlike the money taken from Social Security and that eventually must be repaid by taxing those living at that time.

John Berg

nethy
Jun 6 2011 at 5:52am

This is actually an interesting topic. I would prefer to at least be introduced to the arguments by their proponents. When it’s an opponent being interviewed by a skeptic being listened to by someone who instinctively agrees with them (autocrats are bad isn’t exactly a minority opinion) it feels a bit unfair.

Russ, maybe you can get an eloquent of the dutiful autocrat to come play?

Dylan
Jun 7 2011 at 11:01pm

Interestingly, this whole scenario was discussed by Plato. Any country or community body would best be ruled by a benevolent tyrant. However, autocracy can be the worst and the best forms of government but as said here Democracy will typically be moderately good.

Furthermore, look at the military chain of command and we see why autocracy works. No, political banter just information then action.

keatssycamore
Jun 11 2011 at 8:28pm

“The ‘no such thing as a true autocrat’ point reminds me of …”

Funny. I was reminded more of the phrase ‘mature democracies’ from about minute 13 in this podcast. I was concerned at first, then I realized the most mature democracy around is Greece (after all they invented it).

Which is to say, I basically agree with Mr Berg’s initial point about how the ‘blind spot’ podcast applies to this & Prof Collum’s point about the possible (probable?) limits of ‘statistics’ as applied here.

BTW, I looked for a list of all the countries used when I read the actual study at the above link and I couldn’t find one. I think, as graphed/plotted, it’s really hard to know which country is which (not to mention the lack of a key makes some abbreviations tough to fully grok). Perhaps I’m just not seeing it? Or it’s in a footnote reference? Regardless, I don’t think that info would change my mind about the study’s utility (just really wanted to know who the ‘mature democracies’ are).

Thanks for doing the podcast Roberts et. al. I appreciate your efforts.

Jim Feehely
Jun 13 2011 at 1:08am

The problem with all these analyses is the assumption that growth is necessary, even crucial. But it is this obsession with growth that constitutes the crux of all the world’s real problems. And the problem is compounded by mistaking growth in consumerism as a measure of growth in general prosperity. So called ‘mature democracies’ are really ruled by consumerism, not the public good. Rampant consumerism seems to mean that we no longer can identify what is good for us.

Russ, can you get John Ralston Saul to discuss this more competently.

Comments are closed.


DELVE DEEPER

About this week's guest:

About ideas and people mentioned in this podcast:Books:

      • The Myth of Benevolence, Chapter 3 from The Reason of Rules: Constitutional Political Economy, by Geoffrey Brennan and James M. Buchanan. Free on Econlib.

Articles:

Podcasts and Blogs:


AUDIO TRANSCRIPT

 

Time
Podcast Episode Highlights
0:36Intro. [Recording date: May 13, 2011.] Recent paper, provocative, called "Benevolent Autocrats." What is a benevolent autocrat and what does it have to do with economic growth? A benevolent autocrat is any leader in a non-democratic society that gets the credit for high economic growth. I'm thinking of, like, Lee Kuan Yew gets the credit for the Singapore Miracle and Park Chung Hee gets the credit for South Korea's high growth and the Korean Miracle. And there's China. Now, everyone's favorite example is China, even to the extent that people are starting to talk about a Beijing consensus replacing the Washington consensus, in which authoritarian high growth would be the model to follow. The Washington consensus was the idea that property rights, rule of law, open markets yield growth; and now we have another view that says that autocrats yield growth, would be the polite way to say it. Autocrats, experts yield growth. Paper opens with remarkable quote, seen before, but bears repeating, from New York Times economist Thomas Friedman: "One-party autocracy certainly has its drawbacks. But when it is led by a reasonably enlightened group of people, as China is today, it can also have great advantages. That one party can just impose the politically difficult but critically important policies needed to move a society forward in the 21st century." Now, what's your reaction to that? Tom Friedman is almost too good to be true; he's almost the perfect caricature of the arrogant intellectual who thinks his ideas are the key to the success of the society he is living in. The reason he likes autocrats so much is they have no political constraints on following the advice of brilliant intellectuals like Tom Friedman. In theory, right? In theory they have no constraints. That's one of the things I'm going to examine in the paper, actually. Fantasy version of the benevolent autocrat: they have no constraints, they can do whatever they want; and they follow the advice of the great experts and the intellectual elites, personified by Tom Friedman; can do all the great things to move this society forward into the 21st century. We'll talk about the probability that that's true. But I have to mention my favorite part of the quote. Well, it's probably that he spelled "led" correctly--most people spell it "l-e-a-d" like the metal. But the part I find most striking is the first sentence, said in a smoking-jacket tone: "One-party autocracy certainly has its drawbacks." Yes, it does! Concentration camps, the Gulag--those would be two. Famines that kill 10 million people. Some drawbacks. And then the next part, also worth noting: "But when it is led by a reasonably enlightened group of people, as China is today"--you could say as China appears to be today. I have no intimate knowledge, and I don't even think Thomas Friedman has intimate knowledge; and I assume of the relatively enlightenment of the Chinese leaders, he's drawing the conclusion that they are enlightened or are following green policies, which is what the rest of the quote goes on to talk about: wind power, solar power, electric things. Though he's also just praising them for wise economic policy in general, as others have, of course. We could spend the whole podcast on that quote, but let's cast a wider net. He's not alone. As you correctly, beautifully summarize in the paper, it's a widely espoused or at least considered view by at least academics, pundits, and others. What are some of the other views that are similar to Friedman's? I give a number of quotes from all kinds of sources; but I thought one possibly representative quote was from a group that came together to produce the World Bank Growth Commission Report. The World Bank spent $4 million to get the collected wisdom of all of, literally, 200 academics and other think-tanks to give their views on what causes growth, what to do to raise growth. And $4 million turned out to be a pretty high price tag for reaching almost no conclusions whatsoever on how to achieve high growth. They didn't ask enough people. They only had 200. Some people say you need the S&P500; why have the Dow Jones, to balance your portfolio? What about the Russell 1000? But they had 200; they did the best they could. They have these remarkably wishy-washy quotes, such as: whatever causes growth to go up today may not cause it to go up tomorrow; things like that. But the one thing they did seem to feel as a very strong consensus conclusion: "Growth at such a quick pace"--meaning 7% GDP growth, which would be about 5% per capita growth, incredibly high growth--"over such a long period"--and they were talking about periods of 25 years or more--"requires strong political leadership." Decode it: if you look at all of their actual examples, all of them except Japan were actually autocratic. So that was their euphemism for benevolent autocrats--strong political leadership.
7:29By the way: the word benevolent is somewhat uncertain in many of these cases; but it is certainly true that many of the most dramatic examples of high growth rates, as you chronicle in the paper, are from undemocratic states that are autocratic, correct? Correct. Singapore, China, Korea (South Korea), and a handful of others. Japan would be an exception--a democracy that managed to sustain high growth for a while. I say a while--not a long time, but maybe it is a long time, you could say. That looks pretty convincing. All the winners are autocrats. What's your response to that? That taps into another thing the paper talks about a lot, which is that, unfortunately, all of us have these, what are called cognitive biases--we have ways of thinking about evidence that leads us systematically to misunderstand the way the evidence points. And this fact you were just describing is a classic case, that we notice: if you are a success, you are an autocrat, so the secret to success must be to be an autocrat. But that was asking the question the wrong way around. We don't want to know what is the probability you are an autocrat if you are a success. We want to know what to know what is the probability that, if you are an autocrat, you will be a success. Which are not the same thing, but they are so close. They seem like the same thing. Totally the reverse. So, if you are an autocrat, you have not only great successes of the Lee Kuan Yew's, but you also have to factor in the Mugabes. North Korea, not just South Korea. Once you stop to think about it for 10 seconds, you realize that is the correct probability we need to know--what is the chance of success if you are an autocrat. This systematic tendency that all the psychologists and behavioral economics literature have documented that we mix up these probabilities. We think that the first probability--that most of the successes are autocrats--implies that most autocrats are successes. And it's not at all. Proud as I am of this program and the many things listeners have learned, this may be the single most valuable thing that you learn from this entire program, not just this hour. For example, in the paper you have a table, and if you approach the table nakedly and just look at it, you could easily make the mistake again; but I'm going to try to help the reader read the table. Nine out of 10 of the extraordinary growth stories of the last 50 years have been autocracies. Wow; obviously autocracy has a lot to do with growth. But it turns out, of the 124 countries, there are 89 autocracies. Only 9 of them are successful. Growing really fast is hard to do. You might say: Ok, being an autocrat doesn't lead to growth, but the only way you can get a high growth is to be an autocracy. And that would be a legitimate possibility. Right. Doesn't mean it raises your chances even. You need to know how many democracies there are. Very hard to keep those two things straight. So the rest of the paper goes on to demolish that possibility.
12:10Let's talk about that. So, one view, for those of us who are anti-autocratic and skeptical of this romance about autocracy--we have to face the fact that, even though the odds are small, somehow being an autocracy can lead to high rates of growth. It did happen 9 times out of the 89 countries. That's surprisingly large. How did they manage to do it? There's several ways you can think about this. The first thing that is obviously going on here is just that the variance of growth outcomes under autocracy is much higher than it is under democracy. So, if you are a mature democracy, you are pretty much guaranteed that you are going to have a growth rate that's pretty close to 2% per capita, every year, year in and year out. With pretty small variation around that average. If you are an autocracy then the variance is much more wide open. You could be at -2%, or you could be at +6%. So, that's how you get most of the high-growth countries being under autocracy, because you are picking off the very high end of that variable distribution. To put it in more instinctive terms, basically you can think of autocracy as this very risky bet, sort of like, take all your life savings to Las Vegas and bet it on black on the roulette wheel; you could very rich or completely penniless after that bet. It's more like putting it all on 1 on the roulette wheel, or 17; because black is at least 50-50. It's not quite 50-50, is it. Not in Las Vegas. No, I meant in autocracies, in growth outcomes. Many of the high growth outcomes are autocracies and many of the utterly disastrous growth outcomes are autocracies. Are the really disastrous ones roughly equally numerous, or much more numerous? I actually defined catastrophe in a way that you have roughly the same number of big catastrophes as you do big successes, so yes, that turned out to be less than 5% per year declined. Basically sharp negative growth declines are about as likely as big successes. So, that's where you are gambling. The next question is: Is it correct to attribute the high variance to these good and bad leaders? That's the key question that has been left unanswered. So, all that we've talked about up to now was already well known, at least within the economics field. But where I'm trying to push the debate forward is on this last question: were we really justified in attributed high and low growth to good and bad leaders under autocracy? Because if it's just random--then, it's random. But if it's not random, then you say: We look at what the dumb ones did, and you don't do that; and what the smart ones did, and you do that; and then we could have autocracy as a path to growth. How does that work out? We're much more likely to be favorable to the concept of the benevolent autocrat if we do recognize that at least some successes of these 9 success stories, if we do give credit to the autocrats for them. We are going to end up having this fairly rosy view of benevolent autocrats. But what this paper is arguing is that there is really no evidence that they get the credit for that. That they should get the credit. It looks at that several different ways. It turns out that this really high variance growth under autocracy is actually not between leaders. It's not explained by having high growth under some leaders and low growth under other autocratic leaders. Most of the high variance of growth is actually variance within the terms of long-serving autocratic leaders. This is getting a little wonky. This is serious evidence--if the high variance is due to the leaders then we would expect there would be some good ones, who would get the credit for the high growth. The variance would mostly be driven by the difference between the good leaders and the bad leaders. If we just calculate the average for each leader, then high variance of growth under autocracy is going to correspond to the variance of leader growth averages under autocracy. But it doesn't. It turns out that most of the variance--there isn't that much systematic difference between autocratic leaders. Most of the variance is happening within the terms of leaders. In other words, leaders are starting out good and then going bad, or starting bad and then going good; or good in the middle, bad on either end. All kinds of other patterns; harder to say that it's really the leader that explains the good or bad outcome. If growth is fluctuating wildly during their term in office, then on what basis do we attribute the growth to their term in office?
18:25You could argue that they start off stupid but they get smart. They try these steel smelting foundries in everybody's backyard; when that turns out to be a bad idea you stop it and try something different. And then when you hit on the right thing, you get your legs under you and you can just go forward. Is that in the data? No. That's essentially that's a lot of what benevolent autocrat ideas argue: when all else fails, people will fall back on purely circular reasoning, saying, I'm sorry, I just insist on believing in benevolence, and any high growth that occurs, I'm going to give to autocrats, and any negative growth that occurs, that was their learning period. The hypothesis has become so infinitely flexible that it just boils down to pure circular reasoning. But it could be true. I assume they are benevolent because growth is high; I explain high growth because they are benevolent. But it could be true--that we are handicapped in the United States by this foolish, two-term limitation. Yes. And I certainly wanted to go into this in this paper, admitting the possibility that it could be true and giving the benevolent autocrat hypothesis a fair shake. How'd that turn out? Well, I'm convinced by the paper that there's no evidence to support benevolent autocrats, but the part about cognitive biases has the prediction that we are so wedded to these concepts that no matter how much logical argument I throw at you, you are still going to believe in benevolent autocrats. Ironic feature of this whole debate. Saying that cognitive biases help explain why we do believe in benevolent autocrats and many others, like some version of a Hollywood, stereotypical story that we just like to attribute great events to great personalities. That's called leadership attribution bias. We like to have a hero. These biases are so strong that the prediction of this paper, which shows you that benevolent autocrats are false, also predicts that you will continue to believe they are true. There's an amazing line in the paper where you note that over 140 factors have been found significant in different growth models. Right, 145. That's sobering right there. So how did a benevolent autocrat actually know how to raise growth, when the economists are having such a hard time? One of the conclusions of the paper, one way to summarize it, is, since economists struggle to systematically understand what causes growth, it's hard to believe that a benevolent autocrat would be able to implement growth intentionally. Right. What do you think your opponents would say, your intellectual opponents who did some of those regressions and found some of those 145 factors and would happily pass that on to an autocrat and say: I've found the right recipe? Or, who would say that some autocrats just have a feel for it, and you want to try to find one of those? At some point the story becomes circular and non-falsifiable. If you have your favorite autocrat who has presided over high growth, then that one episode has many explanations. Basically impossible to prove on the basis of one episode what caused or did not cause it. So, if someone's really wedded to that concept, impossible to disprove it. All we can do is look at the whole pattern of the evidence. The other important pattern in the evidence I wanted to highlight in the paper is, the good and bad growth rates that happen under autocracy, you also have to take into account that autocracy is very strongly associated with all sorts of other features of economies that also predict extremely variable outcomes. For example? So, autocracies are usually commodity producers, oil producers. Some of the commodity producers are oil producers. Not only are oil prices incredibly volatile, but even output of oil producers is incredibly volatile. So, some of our highest growth rates are oil producers; and other oil producers have catastrophically negative growth rates. Lots of other features that are correlated with autocracy. Low financial development. Lack of diversification of the economy. These are mostly agricultural, low-income economies with no financial ability to diversify risk. So, you have an oscillation between booms, and good harvests and good times, and catastrophic busts with bad times and no financial ability to hedge risk. These are also economies that are prone to political earthquakes like civil wars. If you had a civil war early in the period we are considering and sort of destroy the economy and then there is rapid reconstruction after the war, that predicts high growth. Or if the civil war happened in the latter half of the period and destroyed a previously prosperous economy, that predicts low growth. This fact that autocracies have very good outcomes or very bad outcomes is really concealing the fact that autocracy is just standing in for low-income economies with lots of other features that also predict very variable outcomes.
25:39I'd like to look a little more at the high end, though. Obviously, as you point out--I quote Hayek in the "Fight of the Century" rap video when he talks about the question of whether WWII cured the Great Depression, and he says: Wow, one data point and you're jumping for joy. There is a tendency to take one data point that could be unrepresentative, an outlier, flukey, it's got other factors that are the true causes. But it is fascinating to me that China is thriving as much as it is. I wonder if we could dig a little deeper into that. To me, there are a couple of issues that I think even the defenders of Chinese economic policy have to look at. As you point out, a lot of people claim successes as their own regardless of their ideology. So, market-oriented folks will say: China's growing because they are liberalizing. Friends of autocrats will say: China's growing because they steer the economy from the top down; or, because they play with their currency. Whatever is their favorite policy du jour. But it is interesting that China appears to be growing. I would ask two questions. One is: Is that growth real? Is it true growth? We have an enormous migration of people from the countryside to the city, where previously a lot of economic activity was outside the market and now it's going to be measured by the market--if they are measuring it accurately or as best they can. So, you'd expect there to be strong measured growth even if the actual growth is smaller, because we've ignored a lost household-production that hasn't been counted. You also have the issue that when you migrate from the countryside to the city you have increased the chances for specialization that has nothing to do with governance from the top or the bottom. And then the next question would be: Is it sustainable? Is it really plausible that they are steering things masterfully, given that they might have an utter collapse any day now? What are your thoughts on that? First question, this is another thing that I think undermines the benevolent autocrat idea, that another characteristic of autocratic low-end economies is that they have statistical departments that are either incompetent or dishonest. Corrupt. So, you would expect that statistical departments that are very bad at measuring things are going to have very noisy growth rates. In the case of China, there is obvious political incentive to exaggerate the measured growth rates. I'm not a China hand myself, but I understand there is a strong camp arguing that Chinese growth is exaggerated, some fudging going on. It's not hard to fudge the growth because there is still a large non-market part of the economy, where we have no idea what the stuff being produced is worth as far as fair market value. I'm not a China hand, either. There's a lot of state-owned on enterprises. I think they are building a lot of things. There's still a lot of investments. May just be the growth of crap. But they are putting it on the books at high value, looks like growth. Now, I do not believe that that explains all the Chinese miracle and the rapid growth--that would be going too far. There is really something going on there. What is going on? Lots of different stories you could tell. I think it's a fairly straightforward case of change, that you had this movement from this totalitarian, destructive government, which itself was following decades of political chaos in China; and then you switched to a government that, even if you don't give it any credit for doing anything positive, at the least stopped being horribly destructive. Then you combine that with this remarkable overseas Chinese diaspora that has showed the potential of the Chinese to participate in world trade and achieve huge gains by taking advantage of the high demand being produced by those democratic market economies to sell them things. I think it's just a changes story. You go from a psychotic, horrible, negative government to one that is just not so psychotic and negative, and that gives you high positive growth. And the opportunity to participate in world markets. And then, on the last point, is the growth sustainable? This is the most underappreciated and yet one of the most robust facts in all of empirical growth economics: Growth miracles never last. They will last for a while, some longer than we expect, but in general they don't last. Most of them don't last very long at all, 5 years, 7 years. Very few last 25 years, like China's has now. But all of the studies suggest that growth is a process that, you can get a string of high growth for a while, but it just doesn't last. Very strong mean reversion in growth rates. Those countries that are above the world average are going to move back to the world average sooner or later. Overwhelming statistical tendency in the data that's been documented by every study we have. Think of all the previous growth miracles. Do you remember the Brazilian miracle? Happened from 1967 to about 1975; Brazil was growing at 10% per year. Who remembers the Ivorian miracle--now the Ivory Coast is everyone's favorite byword for a basket case. But in the 1960s and 1970s, the Ivory Coast was a booming economy, created this gleaming city, Abidjan, gleaming skyscrapers sort of analogous to Shanghai. Sad and tragic things happened to the Ivorian miracle; I'm not predicting that for China. But the point is overwhelming evidence that miracles don't last. In fact, all China's predecessors, the East Asian tigers, have gone back to average world growth. When you were talking about Brazil and the Ivory Coast, I'm thinking about Japan, which was the big attractive story of my youth: Japan's growing faster than us; they must be doing something wise that we are not doing; what is that? Government cooperation; we should emulate them. All the same things now being said about China were said about Japan, right up to the very last moment, around 1990 when suddenly Japan stopped growing and reverted to a very mediocre growth rate and has stayed at that since.
34:06We're struggling too. That's sort of a tragic fact in the history of studying economic development--we give way too much importance to very small amounts of data. It had a huge impact on the history of development economics. In its formative years in the 1930s and 1940s, what everyone remembered was rapid growth of the Soviets under the 5-year plans and the Great Depression in the market economies, and so development started off being much more about Soviet central planning than it was about market economics. Something like that in a much tamer, less severe version, is going on now, coming out of the crisis in the West and China's rapid growth. Let's talk a little bit about the cognitive bias. It reminds me a little bit about the neighbor who either runs up an enormous credit card bill, or maybe he's embezzling an enormous amount of money from his firm; but all of a sudden he's putting an addition on his house and he's got a nicer car. And you look at your life and say: I've got to be more like him, without thinking about what that means. It looks good: His car's bigger than mine. His house is nicer than mine. Right. We have another world of experience, which somehow we are able to treat more sensibly than economics, which is sports. If a player starts off the year hitting 500, or even 400, which many players do early in the year, the baseball season, because if you've only had 20 at-bats, then a few players get 10 hits out of 20 at-bats. But nobody can sustain that for the whole season, and we know that. We don't expect anybody to finish out the season batting 500. And yet, somehow we are not able to appreciate that same principle at work in economic growth. That if you take a very small number of observations, some countries are going to be able to record a remarkably high growth rate average over a 7 year period, able to record a 10% average yearly GDP growth. But we don't expect anyone's going to grow for 100 years at 10%. Should be exactly analogous to the baseball hitter. We get it right on sports but we can't get it right on something that matters a lot more--economic growth. Well, I think we have the same problem in sports, actually. The problem we have in sports is we look at the coach--the coach of the Redskins, Mike Shanahan, he's got two Super Bowl rings; he also had a quarterback named Elway who was really good. Is he a genius; he's tough; it worked in Denver and all of a sudden it's not working. Well, he's got different players. We attribute so much to the person at the top. That certainly is the same in sports and economics. There is a certain romance we have; colleague Dan Klein calls it the people's romance--that government in general can help us; but I think it's more than that. I think it's top down. I'm going to throw out a horrible theory here: I think it starts in the womb. In the womb you are kind of relying on a top down process. And it lasts for a long time, huge portion of your life. At current lifespans, 15-20% is spent dependent on others. So, it's very hard for us to think that salvation could come from ourselves or a different process rather than just top down. I think there are a lot of reasons why people tend to favor top down things. We see a tragic problem and we want to know the exact course of action that will solve the problem. The top down people are happy to tell you what that course of action is, even though they have no evidence and a miserable track record. "The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design", favorite Hayek quote; and for those of you out in the pool, we were about 38 minutes into the podcast before it got mentioned. You also have a quote from Kenneth Arrow that's quite similar that I'd never heard before. Do you have that handy? Let me just look it up on the iPad while we continue talking. Arrow--who we think of as not being in the same ideological place as Hayek; of course, Hayek is I think somewhat unfairly treated as being ideologically boxed in, when he really spans a lot of different ideological camps. Sometimes he gets called a conservative; sometimes the father of the welfare state because he once wrote that health insurance might be a good idea. Anything done in the area of entitlements is fine because Hayek said so. Strange conclusion to leap to. The idea of spontaneous order nowadays has much more appeal across the whole political spectrum than it did when Hayek was writing. The internet is giving us a huge example of spontaneous order. There was a wonderful economist who told a story of an investor who went to France in the early days of the Internet and he was selling an Internet-based company, and the French investors wanted to know who was the president of the Internet. And he said: There is no president. It's a bottom up, self-organizing thing. And the French thought he was completely crazy. It was clear they were not going to invest with him if he was peddling such a crazy idea. So, in the end he told them that he was the president of the Internet. Apocryphal story? I think it's in Paul Seabright's book. Reminds me of that visiting Chinese or Russian communist story of who is in charge of whatever it was--can't remember. They probably were both true, actually, instead of being both false. Any luck with that Arrow quote? My iPad is slow, actually. Almost there. Arrow: "The notion that through the workings of an entire system effects may be very different from, and even opposed to, intentions is surely the most important intellectual contribution that economic thought has made to the general understanding of social processes." Great quote. I believe I have said that last part myself; nice to hear that Kenneth Arrow agrees. Makes me happy.
42:06If you were a listener to this podcast and you hadn't heard much else about growth, we're talking to Bill Easterly; he's a leading authority, arguably the leading authority on economic growth. There might be some people who disagree with that, but everybody would put him in the top 10. Maybe the top 5, top 3, even if you don't agree with what he says. And here he is telling us that we really don't have any clue. We've got Bill Easterly saying this; we've got the 200-person Growth Commission of the World Bank saying it: we really don't understand the processes that lead to growth, and we certainly don't understand the role that political institutions play in growth, although we might occasionally want them to be otherwise. And yet, you just finished teaching a class on growth this semester, a graduate course. What do we know, other than that we don't know what we'd like to know? Is there, other than all the main theories are wrong, anything positive we can conclude? I'm looking for a little uplift here. Or knowledge. Happy to supply some uplift. We don't have the luxury of not choosing. We have to choose. There are alternative camps: should we do central planning, free markets, government picking winners, industrial policy, or should we let the market pick the winners? In the end, we have to choose. One of the reasons I like Hayek is his insight--and in a way, Hayek predicted this growth ignorance that is now become so visible in economics, because he said growth is about innovation, about the emergence of what we want when we see it; we don't even know ahead of time what we are going to want, through the independent and competitive efforts of many. Hayek is using our lack of knowledge as the best possible argument for a decentralized system. If there is no central authority that knows what to do to raise the growth rates, then let the decentralized system find the innovations that will lead to economic success. That's Hayek's big insight. We know it very well in the form of the invisible hand for private goods in the free market, but I think it also applies more generally to society and individual liberty. Let's not give a lot of power to any centralized authority because they don't have any expert knowledge on which they can make growth go up or down. Let's leave things to the spontaneous efforts of not only private markets, suppliers, and entrepreneurs, but also social entrepreneurs and political entrepreneurs in a system of liberty in which you get rewarded positively for things that have a social return and you get penalized for things that have a negative social return. This past Sunday, in the New York Times Book Review, Francis Fukuyama reviewed the new edition of the Constitution of Liberty. He made the claim that Hayek is not Hayekian--which is quite a claim. Interesting argument, along the lines you just said: There's trial and error, people do the best they can; they stumble toward stuff because they don't have the cookbook or the manual. The way I read Fukuyama's critique in this section of his review was that: Well, governments can do the same thing. They can lurch forward, and sometimes they lurch forward successfully and find good policies; and sometimes they fail; and they be entrepreneurial. That's how I took his critique of Hayek's defense of market process. And you just disagreed. Why? What he's totally missing is, if we are talking about a centralized government trying to achieve a whole society-wide outcome, then that kind of knowledge simply doesn't exist. But the kind of knowledge that Hayek said does exist is dispersed, local knowledge, which people have the incentive, in a decentralized system of liberty, to apply to solving local problems. And a problem-solving system works when there is no reliable centralized knowledge, but what the society can do, if there's lots of useful localized knowledge, then the system you want to go for is a system that relies on individual liberty and lets people get rewarded for their contributions to society and also penalizes them for anything negative they do to society. Those are the rules of individual liberty. Couldn't we incentivize bureaucrats to be more successful? Good luck with that. I think there is a role for government. That's another caricature that people throw at Hayek and other similar-minded people. Of course there are public goods. And yes, there are feedback and accountability systems in a free society based on individual liberty where any politician who does a really lousy job of supplying public goods has to answer to voters. I think that tends to work somewhat better at a decentralized municipal level more than at the national level. Have you ever lived in Chicago? No. Maybe in a small town. You've got a small data set here. Maybe in suburbs of Chicago. Yes, the democratic feedback mechanisms could make possible some system of innovation and trial and error; and in fact, they have. Some of the great successes in public goods, like cleaning up municipal water supplies--I think those happened because of democratic feedback mechanisms. But it is a much cruder mechanism than the market mechanism. So, you don't want to turn over the whole market economy to the government because you are substituting a very crude, badly calibrated feedback system in democracy for a finely calibrated feedback system in a decentralized free market feedback system.
49:26I was joking a minute ago about Chicago, but let me make a serious comment along those lines and get your reaction. There is a presumption that municipal, local control works more effectively than national control, and that you should therefore decentralize even to the extent you can the provision of public goods to local areas, local authorities, where there is in theory more control. I think there is less control in local areas. I understand the presumption but it seems to me plays into the cognitive bias part. I'm just speculating here. Most people don't pay any attention to their local elections. They don't pay much attention to local outcomes. I live in Montgomery County in Maryland. It's hard to believe, but we have a monopoly on taxi services. You want to drive a cab in Montgomery County, you have to be a member of this company. I know the name, but I'm not going to say it. And this company, strangely enough, gives a lot of money to the head of the county council. Now, I know that. Do most of my fellow Montgomery County citizens know it? I don't think they do. And it might not matter either way. It's a little taste of autocracy. It's more efficient! Strangely enough, they are often late. But my argument is: There's not much glory in local politics, so we don't spend a lot of time fascinated by the county council or the head of the county government. We spend a lot of time fascinated by the President; who is going to run against him. The more top down, the better, because the more power he'll have. But these local guys--I wonder who is paying attention. And the answer is: not too many, so they can get away with a lot. I hear what you are saying and I agree with those examples of democratic failure. But I think there is another bias that we put too much emphasis on the elections component of democracy and we forget all the other things that can go on in a democracy, that people really do care about. I used to live in Montgomery County; I lived in Takoma Park, which is a particularly vigorous, activist community. My personal favorite example is that I one day picked up the phone and called my city Councilwoman, Kathy Porter, who became mayor of Takoma Park; and I said: Kathy, I've got this gigantic pothole outside of my house. And the next day, Takoma Park Public Works came by and filled the pothole. That's the kind of thing that I think we are ignoring when we obsess about elections. I think that anything that citizens really do care a lot about, they are going to find a way in a democratic system to get heard. There is lobbying and there are all these ways to get heard about things you care about, and that's not reflected at all in the fixation on elections that we have with our usual analyses of democracy. That's a great point. In John Mueller's book, Capitalism, Democracy, and Ralph's Pretty Good Grocery, I think the book was written in the 1990s and he was talking about Mexico; and he said: The PRI Party wins every year, every election, so Mexico is a dictatorship, really. It's not much of a democracy. But that's really only on election day. The other 364 days a year, there's a lot of political forces pushing that party in a lot of directions if they want to stay in power. We've accepted through most of this conversation that there is such a thing as an autocrat. But of course, my favorite counterexample is that Adolph Hitler in the middle of the war--a bunch of German housewives protested in Berlin. Either they had Jewish husbands; I can't remember what it was. And they got their way. He didn't just say: Hey, I'm the dictator, I do what I want. Even Stalin and Hitler and Mao couldn't do "whatever" they wanted. They could do a lot of things they wanted. Horrifyingly tragic and destructive, but they had political forces, whether it's in the army, within the populace. There is political capital that those people have to shepherd, even when they are what we call autocrats. I think that's the other big flaw in the benevolent autocrat theory, that we have this incredibly simplistic idea that the autocrats desire high growth and they are totally unconstrained in being able to reach high growth. When, really what is going on under autocracy is strategic interaction. Big game being played between all the players, only one of whom is the autocratic leader. Potential revolutionaries, demonstrators, elite power-holders coming to the game, and the military. And the outcomes are also often unintentional. The difference often between unintentional autocracy and the unintentional market is that there is no mechanism in autocracy to make sure that the unintentional outcome is a good one. So, the unintentional outcome is going to often be disastrous. In the paper you reference the work of Bruce Bueno de Mesquita, your colleague, who we've interviewed about this. He talks about the selectorate, the people who have control over who is going to be the leader; and they obviously wield a lot of power. And once there's a lot of them, you start to get into the question of what they want, and they don't want a single thing. That is an emergent process as well.
56:06A minute ago, when I asked you what we know, you took a Hayekian route. I'm sure I'd enjoy being a student in your class. But if I were a student in someone else's class, perhaps at your university or perhaps at a different one, they get a different story. Maybe less agnostic than you are about most things. Do you think there is a consensus? At the beginning of our conversation you suggested there was a consensus that we don't know anything. You think that is the consensus in the academic field called development economics? Now a lot of romance about experiments, maybe. I confess that I have a bit of a weakness for overdoing this ignorance point, because I get confronted with so many people who have excessive confidence in how much we do know. But, the reality is: Of course, it's not total ignorance. Development economics is just economics. What do we know as economists? We do know a lot of things that work in the long run even though we can't confidently predict that they will give you a 6% growth rate next year. We do know that if you take advantage of good economics, you will do well in the long run. If you realize gains from trade, gains from specialization, comparative advantage--all these principles, these are ways you can get a lot out of a little. What is a system that does that? It's a market system that lets people trade and specialize and realize their own comparative advantage. I'm not really saying economics is not useful. Far from it. What we don't have is the kind of specialized centralized knowledge that would allow a central authority to achieve a given society-wide outcome. But what we do have is a set of principles, which are basically the principles that tell us how individual liberty is going to work out. If we design institutions that allow individual liberty to flourish. The reasons individual liberty leads to prosperity in the long run are all these principles of gains from trade and gains from specialization. All of that that we have learned from kindergarten. Economics does still work. I'm a firm believer in that. But how do we get there from here? Is that the tricky party? Why is the Washington Consensus not a consensus any more? I think there's confusion between principles and application. I think the principles, we're very clear about, and I think there's even a consensus within a lot of economists about those principles even though there are some differing views on the extent of a government action. The reason the Washington Consensus, if you just took it as principle--it's very innocuous. It's hard to imagine you could get many people to disagree with the principles that are in the Washington Consensus. Like: Don't have gigantic government deficits and debt. Don't have super-high inflation. Let markets set the prices. Mostly don't have government interference in market. Rule of law is a good idea. Private property rights. I think the reason it failed was what was implied in those words Washington Consensus--some elite group in Washington that was going to impose these not as principles. They were going to decide the practical application of these principles through the IMF and World Bank loans in Bolivia, Ecuador, Zambia, Zimbabwe, and the Philippines. And then there was this populace backlash. Foreign experts frankly did not know enough about these economies to apply these principles. I think that's why it failed--in the application stage, not as principles. When I think about it, I think about is more as an interaction problem: there's culture, there's these principles, there are institutions; we just apply the principles without the other two--the culture and the institutions. Doesn't work very well. The principles certainly have to include the institutional rules that make markets work. Culture plays a role in supporting those roles. If you have a culture that supports individual responsibility and trustworthiness, that kind of supports the institutional rules that can also be formalized legally in support of contracts that you need to be able to enforce in order to realize gains from trade. That's the underpinning of the principles that you need to make them operational. But it still comes down to: Economists have spent 200 years since Adam Smith kind of honing this body of knowledge, which still is a very useful body of knowledge about what are the principles of prosperity. So, if you were in charge--if you were in Zimbabwe and Mr. Mugabe took a liking to you and thought you were a swell fellow, he thought the cover of your last book was really nice. So, he says: Tell me one thing to do here? This is a serious question. I know I'm joking, but it's serious. Can you do something productive? We've talked about the pretense of knowledge and how little we know. Could you comfortably and confidently make suggestions for that economy at the margin, or would you find that to be an unbearable situation? The first thing I would tell him is don't hire some white guy from NYU who knows nothing about Zimbabwe. Hire some seasoned politicians with something at stake in the game of policy reform to implement the policy. Even as terrible as the Mugabe government is, there is this finance minister actually--meeting at NYU recently--who did bring down the hyperinflation through standard monetarist principles. Stop printing money. Introduce the dollar as the means of exchange and get rid of the domestic currency. That is the trick to getting rid of inflation. I think it worked much better because an insider did it than it would have worked as a coercive outsider coming in and doing it. Would you continue down that path of suggestions? More to suggest? The suggestions are always going to be: get local people who know how to play the political game of actually making things happen in this society and know what the relevant institutions and political constraints are. And then have them apply good economics to the society. I think the act of economic reform, the art of economic reform, is really figuring out a way in which you apply Econ 101 to local conditions, where you have to have someone very knowledgeable of local conditions.