The Illusion of Autocracy

Ideas Untrapped

Welcome to Ideas Untrapped. My guest today is Vincent Geloso who is a professor of economics at George Mason University. He studies economic history, political economy, and the measurement of living standards. In today's episode, we discuss the differences between democracies and dictatorships, and their relative performance in socioeconomic development. The allure of authoritarian governance has grown tremendously due to the economic success of countries like China, Korea, and Singapore - which managed to escape crippling national poverty traps. The contestable nature of democracies and the difficulty many democratic countries have to continue on a path of growth seems to many people as evidence that a benevolent dictatorship is what many countries need. Vincent challenges this notion and explains many seemingly high-performing dictatorships are so because their control of state resources allows them direct investments towards singular objectives - (such as winning Olympic medals or reducing infant mortality) but at the same time, come with a flip side of unseen costs due to their lack of rights and economic freedom. He argues that the benefits of dictatorships are not as great as they may seem and that liberal democracies are better able to decentralize decision-making and handle complex multi-variate problems. He concludes that while democracies may not always be successful in achieving certain objectives, the constraints they place on political power and rulers mean that people are better off in terms of economic freedom, rights, and other measures of welfare.

TRANSCRIPT

Tobi;

You made the point that dictatorships usually optimise, not your words, but they optimise for univariate factors as opposed to multiple factors, which you get in democracy. So, a dictatorship can be extremely high performing on some metric because they can use the top-down power to allocate resources for that particular goal. Can you shed a bit more light on that? How does that mechanism work in reality?

Vincent;

Yeah, I think a great image people are used to is the USSR, and they're thinking about two things the USSR did quite well: putting people in space before the United States and winning medals at Olympics. Now, the regime really wanted to do those two things. [That is], win a considerable number of medals in [the] Olympics and win the space race. Both of them were meant to showcase the regime's tremendous ability. It was a propaganda ploy, but since it was a single objective and they had immense means at their disposal, i. e. the means that coercion allows them, they could reach those targets really well. And it's easy to see the Russians putting Sputnik first in space, the Russians putting Laika first in space. We can see them winning medals. It's easy to see. The part that is harder to see, the unseen, is the fact that Russians were not enjoying rapidly rising living standards, they were not enjoying improvements in medical care that was commensurate with their level of income, they were not enjoying high-quality education. You can pile all the unseens of the ability of the USSR as a dictatorship to allocate so much resources to two issues, [which] meant that it came with a flip side, which is that these resources were not available for people to allocate them in ways that they thought was more valuable. So, the virtue of a liberal democracy, unlike a dictatorship, is that a liberal democracy has multiple sets of preferences to deal with. And in a liberal democracy, it's not just the fact that we vote, but also that people have certain rights that are enshrined and which are not the object of political conversation. I cannot seize your property, and it's not okay for people to vote with me to seize your property. And in these societies, the idea is that under a liberal democracy, you are better able to decentralize decision-making, and people can find ways to deal with the multiple trade-offs much better. Whereas a dictatorship can just decide, I care about this. I am king, I am president, I am first secretary of the party, I decide this and we'll do this regardless of how much you value other things that I value less than you do.

Tobi;

Two things that I want you to shed more light on. Depending on who you talk to or what they are criticizing, people usually selectively pick their dictatorships. If someone is criticizing, say, for example, capitalism, they always point to the Cuban health care system in contrast to the American health care system. How the American system is so terrible, and how capitalism makes everything worse because of the profit motive. And how we can do better by being more like Cuba.

On the other end of that particular spectrum, if you're talking about economic development, critics of democracy like to point to China. China is not a democracy. And look at all the economic growth they've had in the last 40 years, one of the largest reductions in human poverty we’ve ever seen in history. I mean, from these two examples, what are the shortcomings of these arguments?

Vincent;

Let's do Cuba first, then we can do China. So, the Cuban example is really good for the case I'm making. Because the case I'm making is essentially that the good comes with the bad and you can't remove them. So, people will generally say with Cuba, “yes, we know they don't have political rights, they don't have economic freedom, but they do have high-quality health care.” And by this they don't mean actually health care, they mean low infant mortality or high life expectancy at birth. My reply is, it's because they don't have all these other rights and all these other options [that] they can have infant mortality that is so low. That's because the regime involves a gigantic amount of resources to the production of healthcare. Cuba spends more than 10% of its GDP on health care. Only countries that are seven or ten times richer than Cuba spend as much as a proportion of GDP on health care. 1% of their population are doctors. In the United States, it is a third of that, 0.3% of the population are doctors. So, it's a gigantic proportion. But then when you scratch a bit behind, doctors are, for example, members of the army. They are part of the military force. The regime employs them as the first line of supervision. So, the doctors are also meant to report back what the population says on the ground. So, they're basically listening posts for the dictatorship. And in the process, yeah, they provide some health care, but they're providing some health care as a byproduct of providing surveillance.

The other part is that they're using health care here to promote the regime abroad. And that has one really important effect. One of those is that doctors have targets they must meet, otherwise they're penalized. And when I mean targets, I mean targets for infant mortality. [If] they don't meet those targets, the result is they get punished. And so what do you think doctors do? They will alter their behaviour to avoid punishment. So in some situations, they will reclassify what we call early neonatal death. So, babies who die immediately after exiting the womb to seven days after birth, they will reclassify many of those as late fetal deaths. And late fetal deaths are in-utero deaths or delivery of a dead baby so that the baby exits the womb dead. Now, if a mortality rate starts with early neonatal death [and] not late fetal ones, so if you can reclassify one into the other, you're going to deflate the number total. And the reason why we can detect this is that the sources of both types of mortality are the same,[they] are very similar, so that when you compare them across countries, you generally find the same ratio of one to the other. Generally, it hovers between four to one and six to one. Cuba has a ratio of twelve to 17 to one, which is a clear sign of data manipulation. And it's not because the regime does it out of, like, direct intent. They're not trying to do it directly. It'd be too easy to detect. But by changing people's incentives, doctors’ incentives, in that case, that's what they end up with.

There are also other things that doctors are allowed to do in Cuba. One of them is that patients do not have the right to refuse treatment. Neither do they have the right to privacy, which means that doctors can use heavy-handed methods to make sure that they meet their targets. So in Cuba, you have stuff like casa de mata nidad, where mothers who have at-risk pregnancies or at-risk behaviour during pregnancy will be forcibly incarcerated during their pregnancy. There are multiple cases of documented, pressured abortions or literally coerced abortions. So not just pressured, but coerced. Like, the level is that the person wants to keep the infant, the doctor forces an abortion to be made. Sometimes, it is made without the mother's knowledge until it is too late to anything being done. So you end up with basically the infant mortality rate, yes, being low, but yes, being low because of data manipulation and changes in behaviour so that the number doesn't mean the same thing as it does in rich countries. And now the part that's really important in all I'm saying is [that] what people call the benefits for Cuba is relatively small. My point is that, yeah, maybe they could be able to do it. But the problem is that the measures that allow this to happen, to have a low infant mortality rate are also the measures that make Cubans immensely poor. The fact that the regime can deploy such force, use doctors in such a way, employ such extreme measures, it's the reason why Cubans also don't have property rights, don't have strong economic freedom, don't have the liberty to trade with others. Which means that on other dimensions, their lives are worse off. That means that, for example, their incomes are lower than they could be. They have higher maternal mortality. So, mothers die to [a] greater proportion in labour than in other countries or post-labour. There are lower rates of access to clean water than in equally poor countries in Latin America. There are lower levels of geographic mobility within the country, there are lower levels of nutrition because, for example, there are still ration services. So that means that, yes, they have certain amount of calories, but they don't have that much diversity in terms of what they're allowed or are able to eat without resorting to the black market. Pile these on. These are all dimensions of life that Cubans get to not enjoy because the regime has so much power to do that one thing relatively well. Let's assume it's relatively well, but the answer is, well, would you want to make that trade-off? And most people would probably, if given the choice, would not make the choice of having this. So, those who are saying, “look at how great it is,” are being fooled by the nature of what dictatorships are. Dictatorships can solve simple problems really well, but complex multivariate problems, they are not able to do it in any meaningful way.

The other part that is going to be of also importance is when you look at Cuba, before we move on to China, the other part about Cuba that's worth pointing out is, I was assuming in my previous answer that the regime was actually doing relatively well. Even without considering all the criticism, it still looks like it has a low infant mortality rate. But when you actually look at the history of Cuba, Cuba was exceptional in terms of low infant mortality. Before the Castros took over, Cuba already had a very low level of infant mortality even for a poor country. And so with a friend of mine, a coauthor, Jamie Bologna Pavlik, we used an econometric method to see if Cuba has an infant mortality rate that is as low as it would have been had it not been for the revolution. So, ergo, we're trying to find what is the effect of the revolution on infant mortality and we're trying to use other Latin American countries to predict Cuba's health performance. And what we find is that in the first year of the regime's, infant mortality actually went up, so it increased relative to other Latin American countries, but it gradually reverted back to what would be the long-run trend. So that Cuba is no more exceptional today in terms of infant mortality than it was in 1959. That is actually a very depressing statement because it's saying that the regime wasn't even able to make the country more exceptional. So even if it's able to achieve that mission quite well, it's not clear how well they've done it. At the very least, they haven't made things worse in the very long run, they only made things worse in the short run. So when you're doing, like, kind of, a ledger of goods and bads of the regime, all the bad trade-offs I mentioned: lower incomes, higher mortality rates for mothers and maternity, lower rates of access to clean water, lower rates of access to diverse food sources, lower rates of geographic mobility - pile these on, keep piling them on, that's the cost. What I'm saying is what they call the benefits, they're not even as big as it's disclaimed. The benefits are relatively small.

And now with regards to China…

Tobi;

Yeah.

Vincent;

The Chinese case is even worse for people because they have a similar story with GDP. So, in China, a regional bureaucrats have to meet certain targets of economic growth. Now, these same bureaucrats are in charge of producing the data that says whether or not there is economic growth. You can see why there is a who guards the guardian's problem here? The person who guards the guardian is apparently one of the guardians. So you could expect some kind of bad behaviour. And there is an economist, Luis Martinez, out of the University of Chicago. What he did is he say, well, we have one measure that we know is a good reflector of economic growth and it is artificial light intensity at night. Largely because the richer a country is, the more light there will be at night time. And so if you have like 1% growth in income, in real numbers, you should have some form of commensurate increase in light intensity during night time. If the two deviates, it's a sign that the GDP numbers are false, that they're misleading. Because if they deviate, the true number, the always true number will be the light intensity at nighttime. So, when Martinez used the nighttime light to compare GDP in Chinese regions overall and the actual GDP, he found that you can cut the growth rate of China by, maybe, two-fifths, so it is 40% slower than it actually is. So, China is not even as impressive as it is. And the thing is now think about the pandemic, think about how extreme the measures that China deployed to restrain this has been, no liberal democracy would have been able to do that, no free society would have tolerated forcibly walling people into their houses. And there are massive downsides to the communist regime in China. Like, yes, the regime is free to do whatever it wants, but it also means that it can put Uyghur Muslims into concentration camps. It also means that it can wall people into their houses when they do not comply with public health order. It also means that people are under the social credit system where they are being largely surveilled on a daily basis. It also means that the government can allocate massive resources to the act of conquering Taiwan or flexing muscles towards Japan. All things that when you think about it, is that really an improvement in welfare? Obviously, you can say that, oh yeah, they're doing X or Y things really well but here are all the bad things that come with this. And those bad things are on net much worse than the good things.

Tobi;

Now, you keep emphasizing liberal democracy and I want to get at the nuance here because I've seen several results. Either it is from Chile and other countries that say unequivocally that democracies are better for growth than dictatorships, even in the case of Chile, despite all the reforms of Pinochet regime. But what I want to get at is, what exactly about democracies make them better? Because, for example, we can think of Nigeria and Nigeria as a democracy. We've had uninterrupted election cycles for over two decades now, but there's still very weak rule of law. Successive governments still rely on extracting oil rents, basically. And, the degree to which people enjoy rights vary depending on who is in power or their mood on any particular day. And, of course, Nigeria is a democracy. So is it liberal democracy? Is that the key factor?

Vincent;

So, think about it this way.

Tobi;

Yeah.

Vincent;

Think about it this way. Inside the big box of liberal democracy, there is for sure democracy. But the part that makes the box liberal democracy is not only the smaller babushkadal inside that box which is a democracy one, it is the other constraints that we put on the exercise of political power. The true definition of a liberal democracy, at least in my opinion, is that not only are people allowed to vote, but they are restraints on what we can vote on. So, for example, if it's not legitimate for me to steal from you, it is no more legitimate for me to vote with two other people to steal from you. The act of democracy should warrant some acts that are outside the realm of political decision-making. There are also constraints that exist on rulers, so it's not just that there are some rights that are not subject to conversation. There could be also incentives that prevent rulers from abusing the powers they have. That would mean, for example, checks and balances, where there are different chambers that will compete with each other, different regional powers of government that will compete with each other for jurisdiction, and so they will keep each other in balance. It could also be some form of external constraint, because a liberal democracy can also rely on external constraints upon political actors. It could be the fact that people can leave the country, the fact that taxpayers can migrate to another country, puts pressure on politicians to not abuse them. People can move their capital out of the country, [this] creates a pressure on politicians to not try to steal from them, because people will just remove all the productive capital and the ruler will be left with very little to exploit as a result, regardless of whether or not the ruler is elected or not. So the way to think about this is liberal democracy is, you want to have a system where there are rules, incentives, constraints that make it so that we are not betting on a man or a woman, for that matter, being the correct man and woman for the moment. We care about a set of incentives, constraints, and rules that will make sure that even the worst human being possible will feel compelled or compulsed [sic] to do the right thing. So, that's like the old Milton Friedman thing, it’s like “I don't want the right man. I want to have a system that makes sure that even the most horrible person on earth is forced to do the right thing.” That's what a liberal democracy is.

Now, it is a broad definition that I've provided. It is not narrow in any way. It is not specific, largely because I don't think it can be what works. It’s not everywhere the same. The general family to which this belongs is universal. But the way it can work is not the same everywhere. A homogeneous, small, Sweden probably doesn't need as much level of, say, breakdown of provincial versus federal powers. Whereas, from what I understand, Nigeria is a somewhat multinational country, multiethnic country with multiple groups east and west from what I understand the divide is in Nigeria. There, it might be good to have a division inside the country where things that are most homogeneous, you leave to the federal government, to the highest level of power. Then the things that you can delegate to the local level, [it is] better to do it that way. Countries that are incredibly heterogeneous maybe need even more federalism. What is optimal for one place won't work elsewhere. So I couldn't take Belgian institutions and then just dump them in Nigeria. Same as I couldn't just say, well, let's take Swedish institutions and dump them into Canada. But what makes generally Sweden work better in terms of institutions than Nigeria, for example, is the fact that Sweden does fit in that general box of liberal democracy. There are clear constraints, there are restrictions, there are constitutions that are well respected, there's a strong rule of law, and politicians are compelled to not fall prey to their own baser instincts.

Tobi;

A couple of months ago, I had Mark Koyama on the show.

Vincent;

Great guy. He's a colleague of mine.

Tobi;

Yeah. So, we were talking about state capacity. We're talking about his book with Noel Johnson. So I did bring up your paper on state capacity, [in] which, basically, one description that stuck with me is that you never really find a poor, but highly capable state in history…

Vincent;

You mean backwards. A rich society with an incapable state?

Tobi;

Yes, a rich society with an incapable state. Thanks for that. So, I've been trying to disentangle this state capacity thing, I know Bryan Caplan basically dismissed it as a sleight of hand. Right. So, like, how does it work and how is it a necessary ingredient for economic development, so to speak?

Vincent;

I am actually quite respectful of the state capacity literature in one way. So let me do like kind of a quick thing. State capacity says that you want the state to be able to do certain missions. Right, so we're not making judgments as to whether the mission is good. State capacity is about the abilities of the state. The reason why that literature has emerged since the 2000… here's a story of economic thought really briefly: in the 1950s, Samuelson and others show, ‘oh, well, there are market failures’ and then a few years later there are the public choice rebuttals, where the public choice economists say, ‘well, you're kind of wrong. There are also government failures.’ And the state capacity crowd tries to come in between these two and say, ‘yeah, there are market failures and there are government failures. How do we get a state to solve the market failures but not fall into government failures?’ Okay, straightforward, good argument. The part that I'm sceptical of is that the argument of the state capacity crowd is that you will have a lot of rich societies that will have strong states, you will have much fewer societies that have strong states but are very poor (the USSR would be a good example of that), [and] you will have a lot of societies that are poor and have weak state. The thing is that they can't seem to explain why it is under their theory that there are no societies that are relatively weak state but rich. Even though in history we do have many examples of these and they collapse all the time.

The argument that I make with my colleague, Alexander Salter, is that societies that have weak states will fall prey to predation because their neighbours with stronger state will try to capture their wealth by conquest. If they are conquered, they grow immensely poor, they are made poor. Basically, it's a terrible event for them. Or they resist, and if they resist ably, the result from resistance is that they have to build a strong state themselves to resist predation by other rulers. And so in the argument me and Alex build, it boils down to: the state is not necessary for development, but it is inevitable as an outcome. So, the task of political science, of political economy, is understanding if we are going to be stuck with one of them, how do we make it that we get the least terrible one? If it's not necessary, but it is inevitable, then how do we get to one that will maybe do some benefit, or at least, we can get the best kind possible? Well, that's where the liberal democratic answer gets into. [It] is [that] we need to find sets of constraints, rules, incentives that force the politicians to make it too costly for them to engage in predatory behaviour, in redistributive behaviour, and that they concentrate on what you could call productive behaviour. That would be like solving externalities. Like dealing with pollution or producing public goods stuff that markets have a harder time to produce. Getting into that category is the task of what liberal democracies are trying to do. That is a much harder proposition. Daron Acemoglu in his somewhat awful book, The Narrow Corridor, calls it a narrow corridor. (I don't like that book that much. I think it's a horrible piece of literature. He should have kept it at Why Nations Fail, we had everything we needed with Buchanan, and it was much better in the other version. He was a much worse version of that.) So, Parenthesis over on Daron Acemoglu, but his point is still relatively okay. There is a narrow corridor on which we evolve. That is a very narrow equilibrium that we want to stay on to, to avoid veering either into more territorial forms of government or into different types of authoritarian[ism], in a certain way. So the corridor for a liberal democracy is very, very, very, very narrow.

Tobi;

I like that description. The state is not necessary but inevitable. Whereas with the traditional state capacity crowd, the state is often assumed and never justified.

Vincent;

Actually, that's a bit unfair to them. The state capacity crowd, a lot of them are interested in state capacity as a story of the origins of states. That, I think, is a much-valued contribution. However, the issue of whether or not state capacity is linked to growth, I think this is where there's overstretching. My point is “no, there's very little reason to believe that state capacity is related to growth.” State capacity is more the direct or indirect result of growth in the past. So, either you are getting state capacity because you get conquered and you get imposed it by somebody else, or you get state capacity because you want to protect your wealth from other predators.

Tobi;

For the record, I'm not talking about your colleagues. There's this industrial policy school in development economics who are also big on state capacity, who think the state has to do this heavy lifting. They sort of assume the state and not justify it. But I won't let you go without asking you this final question. You recently published a paper - talking about the work of Thomas Piketty, the French economist - with Phillip Magness, I should say. What is your critique of his work? Because so far as I can tell, yes, I read the op-ed in the Wall Street Journal, [but] everybody else is sort of pretending that a critique of Piketty does not exist. And the political coalition around their research, along with [Emmanuel] Saez and [Gabriel] Zucman is moving rapidly apace, whether it is in taxation or other forms of agenda. So, what is your critique? I know there have been others in the past Matthew Rognlie, I'm not sure how to pronounce his last name.

Vincent;

Yeah. Our argument is actually very simple. And to be honest, I don't really care about the political conversation where, [for] the political people who are using Piketty's work, I ignore them. There may be a motivation for doing this work because it tells you the importance of his work, but the person I'm trying to talk to is Piketty himself. And the point we make in the paper is that he [not only] massively overestimates inequality in terms of levels, but he also misses times a lot of changes. In the article that me, Phil, another Phil, and John Moore published together in the Economic Journal, we find that there is a very different timeline of inequality in the United States. The most important part is that unlike Piketty and Saez, who can assign most of, and later Zucman… who can assign most of the changes in inequality to tax policy, we find that actually half the decline in inequality that happens between, say, 1917 and 1960, half of it is because of the Great Depression. And just as good economists, we should not be happy that, okay, the rich are growing poor faster than the poor, but the poor are also growing poor. That is not a decent outcome. So we're minimizing the role of fiscal policy and tax policy in doing inequality, but also the other changes that we find give a very different story of what matters in changing policy rather than being taxes, it has more to do with labour mobility within the United States. With capital mobility within the United States. So poor workers from the south, mostly black Americans, move to richer northern cities where wages are higher. Capital moves from the rich north to the poor south where workers are made more productive. So, the levelling has to do with a very standard force in economics - it's a Solow growth model - capital goes to where the returns are greatest, labour goes where the wages are greatest. Most of the convergence is explained by this, not by tax policy changes. So that's the critique we make of them. And there's a lot of other people who are joining in, Gerald Holtham, David Splinter, a lot of people are actually finding that their numbers don't make much sense and they're actually in violation of a lot of other facts of economic history, even though they're correct in the general idea that inequality fell; fell to 1960 and rose since the 1980. The problem is that all they got right is the shape, but they got wrong the timing, the levels, the extent of the changes. They got most of it wrong. They just got the general shape right. And that's no great feat.

Tobi;

Thank you so much for joining me.

Vincent;

It was a pleasure.

This is a public episode. If you’d like to discuss this with other subscribers or get access to bonus episodes, visit www.ideasuntrapped.com/subscribe