Mechanics of Political power

Part 1: A macro view

I have always been mystified why dictators who brutalize their population, are often in power for such a long period. Especially Mugabe, the previous ruler of Zimbabwe. The way he managed the economy was so incompetent. Why are his followers not deposing him? Why are citizens, who are in the majority not deposing him? How can one person control so many others? Just one bullet or a few drops of rat poison is all it takes. 

Power in general is an interesting thing. It has a weird pull on people, but most would not want to admit it. For example a while back I bought 48 laws of power by Robert Greene. The book was laying around my parents house at one point, and my mother actually started reading in it. She normally never does that. Then my uncle picked it up while he was visiting, he also never did that with any previous books that were laying around. And there are usually a ton of books about a wide variety of subjects laying around. Then a friend borrowed it (who rarely reads non fiction) and I have not seen the book since. 

Power is very illusive as well. One moment one has it, and the next moment it is gone. That is what makes it more fascinating to me. It is hard to wrap your head around it if you really think about it. I have not found a lot of good books that analyzes power in a more systematic way. Mostly books with personal tips on how to get more of it.

George RR Martin illustrates the mystery of it well:

“Oh, I think not,” Varys said, swirling the wine in his cup. “Power is a curious thing, my lord. Perchance you have considered the riddle I posed you that day in the inn?” 

“It has crossed my mind a time or two,” Tyrion admitted. “The king, the priest, the rich man—who lives and who dies? Who will the swordsman obey? It’s a riddle without an answer, or rather, too many answers. All depends on the man with the sword.” 

“And yet he is no one,” Varys said. “He has neither crown nor gold nor favor of the gods, only a piece of pointed steel.” 

“That piece of steel is the power of life and death.” 

“Just so… yet if it is the swordsmen who rule us in truth, who do we pretend our kings hold the power? Why should a strong man with a sword ever obey a child king like Joffrey, or a wine-sodden oaf like his father?” 

“Because these child kings and drunken oafs can call other strong men, with other swords.” 

“Then these other swordsmen have the true power. Or do they?” Varys smiled. “Some say knowledge is power. Some tell us that all power comes from the gods. Others say it derives from law. Yet that day on the steps of Baelor’s Sept, our godly High Septon and the lawful Queen Regent and your ever-so-knowledgeable servant were as powerless as any cobbler or cooper in the crowd. Who truly killed Eddard Stark, do you think? Joffrey, who gave the command? Ser Ilyn Payne, who swung the sword? Or… another?” 

Tyrion cocked his head sideways. “Did you mean to answer your damned riddle, or only to make my head ache worse?” 

Varys smiled. “Here, then. Power resides where men believe it resides. No more and no less.” 

“So power is a mummer’s trick?” 

“A shadow on the wall,” Varys murmured, “yet shadows can kill. And ofttimes a very small man can cast a very large shadow.” 

Tyrion smiled. “Lord Varys, I am growing strangely fond of you. I may kill you yet, but I think I’d feel sad about it.” 

“I will take that as high praise.”

I think saying that power is based on where people believe it is, only scratches the surface of what is really going on. There are clearly some principles which govern why people believe what they do, especially since ones holding on to power tend to often be abusive.

What I have learned so far

One of the first books I read that analyzes political power in a more academic way is The logic of political survival. This book introduces the selectorate theory, and provides a good starting point. It is backed up with a lot of data. It does have some issues though. I will try to make a tl;dr of the book with some of my own thoughts added in. The same author also wrote The Dictators handbook, I have not read that book.

The way the author defines it, in any given population you have the general population (N) which is the total population, the selectorate (S)  a subset of N from which from which the leader and winning coalition is selected, the marginalized (N - S), the winning coalition (W), a challenger (C) and the leader (L). Then there are public goods and private goods. Public goods being things like infrastructure and national healthcare. And private goods being bribes and government and foreign aid money being stolen and given to close supporters.

So for example in Russia, since they technically do have elections and anyone can go into politics (although very rigged), the selectorate (S) is the same as the number of residents (N). But the winning coalition (W) is very small and consists mostly of Putin, the oligarchs, siloviki and silogarchs and some people serving under them. With Putin of course being the leader (L). And probably Navalny being the most credible challenger (C). Public goods are rather lacking in Russia and probably a majority of assets are being stolen and go to Putin and his cronies. So in reality Russia is closer to a autocracy. (From now on I will just use the letters instead of words as that is quicker). 

The selectorate theory argues that this kind of system is generally the most stable, where leaders last the longest once affinities are established. This is because ratio of W/S is small. If W/S is small it means that the probability that someone in S enters W is small. Generally in autocracies, W is about 10% of N. So this gives L more leverage, as being kicked out of W means a big loss. Since the wealth difference between someone outside of W and inside W is large. Also because W is very small, it reduces the total amount of resources that L needs to keep his coalition together, increasing the amount of rewards that people in W get. Having a wealth advantage also allows members of W to suppress anyone outside of W more easily.

The reason L is relatively secure from power grabs from within W is because for C within W to challenge L, they would have to offer more resources than L is currently giving out, since C would have to do it probabilistically while L is already doing it definitely. And cost of failure is likely to be great. What would have to happen as well is that W would have to be thinned out after C deposes L if L is already optimally rewarding everyone within W to give the new W their increased reward. So this generally only occurs when L is very sub optimally rewarding everyone within W.

This is why generally public goods are very limited in autocracies, because there is little incentive to spend money on anyone outside of W.  Statistically, autocratic leaders are far more often toppled by people from within W than from anyone outside of W. So the biggest threat comes from within W. So they are incentivized to spend as much on W as possible, and don't have much of an incentive to spend much money on anyone outside that group.

A military Junta or monarchy would be a system where S is actually small too, with W even smaller. Since leaders can only be selected by bloodline or from a select group of officers. This generally gives L less leverage. As the ratio of W/S is much higher. So L is not as rich as in a system with a small W/S ratio. This is why they are also more unstable. A system that selects by bloodline is more likely to become unstable over time as the selection for competent leaders, maintaining the current political system, is much smaller than for an autocracy. Since in an autocracy, a leader can be selected from the general population, so a much larger talent pool. This could explain why there are so few kingdoms left standing today, as pressures on autocratic leaders have grown due to modern technology and urbanization (a more concentrated population is harder to control?). But that is speculation on my part. 

In a true democracy W/S is very high as well. But since the cost of rewarding supporters is very expensive, since W is large and S is also large, the difference between public and private goods become very small. So W is less loyal to L (since there is less to lose). And since every person actually has a vote, it would be difficult to reward your own faction with a lot of private goods and stay into power. So this system incentivizes leaders to spend a lot of public goods instead of private goods. The data seems to back this up (with some issues I will come back later on). 

This is also why there are few authoritarian governments with a moderate W/S size. Usually the loyalty norm between leaders and W becomes too weak. So this would be an unstable system that would either revert back to a small W system with a new leader who would thin out W and increase rewards to remaining members of W, or towards a democracy. 

Democratic leaders tend to not last very long though, since the loyalty norm is much weaker because cost of being outside of W is much much lower compared to an authoritarian system both for L and other members of W. And being C outside of W is far easier and has a much lower cost of failure as well. 

The reason why a democracy tends to stay a democracy is because to increase rewards and power for L, W has to be decreased. And getting into W has to be made more difficult. Generally in a democracy everyone in N (which is same as S) has more resources. So there is more to lose and people in S who will not be part of W have a greater ability to stop this. Plus since W is generally more than half of S, there are not a lot of resources around to reward each person in W and loyalty of people in W would be low (since resources are spread out over large amount of people). So to go to a low W/S ratio you would need support of a lot of people, who would then need to be back stabbed and kicked out of W after supporting L. Which is difficult. The data seems to back up the fact that actual democracies tend to stay democracies. With some well known exceptions of course.

Now revolutions in authoritarian systems are generally initialized by people outside W and/or outside S. They have a large incentive to dislodge the leader and increase the potential W and S as much as possible. This is because they need all the help they can get. Even the leader in a Junta for example, is incentivized to increase S as well and not increase W of course. Since it gives him more leverage. This can be seen in Venezuela recently where Maduro greatly increased the number of generals in the army (source:reuters).

Some revolutions end in democracy and some end in autocracy. In a small W and large S system, it is easy to find a lot of people who would be interested in increasing W and basically go towards a democratic system. Since the majority of the population is outside of W. And if L is not careful, they might not have anything to lose. So nothing to lose plus everything to gain is a dangerous combination for L. And to dislodge L and his winning coalition W, C needs a lot of supporters. Because W generally has all the resources (like more advanced weapons). So C needs everyone he can get. And the only way to do that is to promise democracy. Since promising private resources would only barely offer more than public resources (as S - W, or the disenfranchised are a large majority of the population). So might as well promise public resources then to get as many people as possible. 

But as C deposes L, he actually becomes L. If he has no significant opposition and access to all the resources, his incentive changes. As rewards would be great if he thins out the least loyal (and armed) followers from W. So he is incentivized to reduce W while keeping S large and break his promise or set up a rigged election system so he gets the most rewards. And he is incentivized to kill the previous L, since reward for getting power back would be large for the previous L (various South American nations for example).

When W is defeated by a more splintered S-W and essentially has multiple equally powerful C’s, or a weak figurehead C, this is not possible. And each group in the new winning coalition does not want to give the other parties too much power. So generally that is more likely to result in democracy (India for example). Although I found the argumentation here a bit shallow. It seems like strength of institutions should also play a role?

Shortcomings of the selectorate theory

The first thing that jumped out is that India is an actual democracy while China is an autocracy. China currently is a classic Large S small W system. Yet China has become far richer than India, with far more public goods:

So this is in contradiction with the selectorate theory right? It should be the other way around. And we are talking about more than a third of the global population here. 

What is also confusing to me is that the book argues that low W/N systems have high personal tax rates. When I look at tax rates in countries with authoritarian regimes I see mostly low tax rates. Plus it is really easy to evade taxes in poorer countries. Also if tax rates are higher, why would you work less? Poor economies are less efficient, they have to work harder to get the same result. Maybe he used a lot of historical data and is taking feudalistic states into account. Which means that technology probably plays a role in this. Also I saw very little arguments on differences between natural resource rich and poor countries. 

Then there is the issue of affinity building. And the definition of affinity. Which was glanced over. How is affinity built? Why do some leaders not survive the initial affinity building stage and some do? What do reward structures look like within W? What determines who becomes the leader in W? How exactly can only 10% of the population dominate the other 90%? What are the mechanics there? What role does information play? How does size of a group affect the power structure? What determines exactly how much L spends in public resources, since this number is usually not zero? What role does geography and population density play? And how do we define who is in W? Since this is not exactly a black and white matter. Generally power fades the further you go down.

But maybe expecting to have all those questions answered in one book would be too much to ask.

Overall though, the book is a nice base to start from. It is a nice macro view of how incentives influence politics, with a lot of historical data and examples.

In my next post I will talk about what I have learned from other books and what I have come up with that is not in those books, and how I would answer some of the above questions.