The Dictator's Handbook

The Dictator’s Handbook

by Bruce Bueno de Mesquita and Alastair Smith, 2011

I was impressed with this book. Its tone is very cynical, which I enjoyed, but in the end it is strongly pro-democracy. This book clearly explains why authoritarian regimes tend to be unpleasant places for the common people to live, and democracies are far preferable.

Bueno de Mesquita and Smith’s main argument boils down to the importance of coalition size.

In any regime, there are three groups of people. There are the “interchangables”, basically anyone with a modicum of political power, whether they are voters in a democracy or potential officials in a dictatorship. Then there are the “influentials”, the people with actual influence. In a democracy, these are the people who actually vote. (Or, if there’s bloc voting going on, these are the people who control the votes.)

Finally, there are the “essentials”, also called the winning coalition. This is the smallest possible number of people whose support the leader actually needs, in order to stay in power. In a democracy, the coalition is a sizable fraction of the population. In an authoritarian regime, it is a very small fraction of the population. This is the primary difference between democratic and authoritarian governments and for this reason the book generally refers to large-coalition and small-coalition regimes to describe the two types.

According to The Dictator’s Handbook, every difference between the behavior of democracies and authoritarian governments can be explained by the differing incentives in place for large-coalition and small-coalition leaders. In the authors’ words, “when the coalition of essential backers is small and private goods are an efficient way to stay in power, then the well-being of the broader population falls by the wayside”. (p 13) What does this mean?

In a small-coalition regime, the leader can buy the loyalty of the essentials through extravagant salaries, gifts, or encouraging them to use corruption to extract money where they can. In a large-coalition regime, the leader cannot possibly afford to buy the support of enough individual people this way. Instead, the government must spend money on public goods to gain enough support. If the coalition is large enough, these goods benefit all the people, whether they are actually in the winning coalition (in other words, they voted for the leader) or not.

In a small-coalition regime, leaders are incentivized to provide primary-school education and enough health care to keep workers in shape, but not to fund quality higher education (You don’t want to give the unwashed masses any dangerous ideas! The elites send their children abroad to be educated) or good health care for babies and elderly people (they don’t work). In a large-coalition regime, these things must be funded to keep the large number of essentials happy. This is why clean drinking water is far more widely available in democratic countries than in autocracies of similar economic status.

In a small-coalition regime, corruption takes the bite out of regulations designed to minimize the impact of natural disasters, and leaders are primarily incentivized to make sure the influentials are not harmed -- for instance, neighborhoods where favored constituencies live tend to be better protected against flooding. In a large-coalition regime, these regulations are actually followed -- this is why earthquakes tend to produce far higher death tolls in authoritarian regimes than democracies. Also, if there’s a perception that the government’s response to a disaster was feeble or incompetent, this could severely damage a democratic leader’s standing (see Bush and Hurricane Katrina), but not an autocrat’s (see Myanmar’s military leaders and Cyclone Nargis, which killed far more people than Katrina). Democrats have incentives to minimize damage and relieve the common people’s suffering. Authoritarian leaders, generally speaking, do not. Heck, if a leader is really lucky, a huge death toll might mean loads of foreign aid payments!

In a small-coalition regime, the leader can tax the people extortionately (through a variety of formal and informal means) and then turn around and provide subsidies to favored groups. In practice, this usually means a transfer of wealth from the poor to the rich. In a large-coalition regime, people might feel heavily taxed, but that’s because every cent they pay in tax is through official channels. What’s more, they benefit from far superior public services.

The authors make a spirited argument that democracies provide far better public services and are just plain nicer to live in than authoritarian regimes.

Several months ago, I read Democracy for Realists by Achen and Bartels (towards the bottom of the linked post), which convincingly argues that elections in democracies do not necessarily result in good governance. But the book’s authors still prefer democracy to the alternative, and in the last chapter they list some of the benefits of democracy as they see it (which I summarize in that post’s last bullet point).

Frankly speaking, I think that what Achen and Bartels come up with with pale in comparison with the defense of democracy in The Dictator’s Handbook, which gives one well-supported reason after another why a government that is forced to be accountable to the people is far nicer to live under, even when factors such as differing levels of economic development are accounted for.

The authors hold that it is advantageous for the coalition size to be very large -- in other words, they think as many people ought to be involved in selecting leaders as possible. They are not fans of gerrymandering or the Electoral College in the USA. They also think it’s a bad idea for a democracy to have large numbers of long-term residents who can’t vote, and so they strongly support all democracies liberalizing immigration laws to give immigrants a plausible path to citizenship.

None of this is meant to imply politicians in democracies are morally better! Democracies are far nicer places for the common people to live, but democratic leaders are just as cynical as despots; they just have different incentives in place.

Look at the use of foreign aid. We may have this idealistic image of aid going to help poor individuals in developing countries, but democracies know very well that most of the aid given to authoritarian regimes gets skimmed off. That’s fine with them, because the real purpose of aid is to persuade the recipient to adopt policies favored by the donor. During the Cold War, Western governments spent substantial sums of money to persuade small-coalition regimes to tilt toward them and away from Communist countries.

It’s far more cost-effective to influence small-coalition regimes than large-coalition regimes in this way. In the 2003 Iraq war, the USA would have liked to use Turkey as a base to invade northern Iraq, but Turkey was just democratic enough that the government’s price was simply too high. Letting the Americans invade Iraq from Turkey would have been highly unpopular among the Turkish people, and the government would have had to spend far too much to keep its winning coalition together. Authoritarian Kuwait and Saudi Arabia, with fewer people who needed appeasing, were much cheaper.

The authors supply plenty of case studies taken not only from governments (the rise and fall of Liberia’s Samuel Doe) but also corporate politics (the rise and fall of HP’s Carly Fiorina). There is a rich variety of other examples.

Unfortunately, I don’t know just how trustworthy these examples are. When they wrote a paragraph on Taiwan’s democratization, something I actually know something about, I found their description to be insultingly dumbed-down and inaccurate. This, in turn, caused me to wonder what other problems were lurking in other examples which I don’t know so much about. As the book relies heavily on these historical examples to illustrate its points, this is a significant flaw. That said, the main themes of the book still stand.

My description above doesn’t come close to covering the whole book. Some other bits of note:

• When we hear about authoritarian leaders borrowing huge sums of money and spending it lavishly on their families or their supporters, we might condescendingly think they are foolish and ignorant when it comes to money, but in fact they are acting entirely rationally. Debt can be left for future regimes, but their small coalition needs to be kept happy in the short term or they may find themselves thrown out of power. 

• And speaking of money, debt relief for developing countries sounds like the morally right move in theory, but in practice it tends to merely extend a financial lifeline for authoritarian governments. In the authors’ opinion, it is better for the affected countries in the long term to force profligate autocracies into financial crises that are likely to bring about their ultimate downfall.

• Autocrats do not necessarily have an incentive to provide good infrastructure for the people. It’s enough to maintain the absolute minimum to keep local economies grinding along. If the country has natural resource wealth, that’s a different story; the government of course has to keep those extraction industries humming along, but they’re there to enrich the leaders, not help the people. (Large infrastructure projects that could legitimately help the people do have their use: they are prime opportunities for large-scale corruption.)

• Encouraging government officials and business leaders to be corrupt not only makes them fat and happy, but it’s also a way to enforce compliance and loyalty. If everyone is guilty of corruption to some extent, anti-corruption drives (which are, of course, selectively enforced) become a very useful tool to eliminate whatever potential troublemakers the regime sees as a threat.

• We often hear that democracies do not go to war with other democracies. But apparently what’s actually going on is that democracies are extremely reluctant to go to war unless they are certain of winning, and will try to exhaust all diplomatic solutions first. Authoritarian governments are far more willing to go to war, and are much more likely to cut their losses and quit when things go badly. This is because a loss in a war (or just suffering significant casualties) can bring down a democratic government, but is less likely to end an authoritarian regime. Also, democracies tend to go to war to achieve policy goals; autocrats tend to fight for land or treasure. This does not necessarily give democracies the moral high ground; there are many examples of democracies picking on weak opponents, some of which have large-coalition regimes of their own. “War for democrats is just another way of achieving the goals for which foreign aid would otherwise be used.” (p 237)

• It’s awfully risky for the boss to have competent people close to him, because they could be potential rivals. In pre-modern times, employing competent eunuchs was often the solution, as they would never be acceptable leaders in their own right. That doesn’t work so well nowadays, but there are other ways around the problem; note that Saddam Hussein’s right-hand-man Tariq Aziz was a Christian, which meant he was never a plausible replacement for the boss.

• There’s an inverse correlation between the straightness of the road connecting the capital city and its major airport, and how democratic the government is. In a dictatorship, you don’t need to worry so much about the people you displace when you build a major highway. Also, a nice straight highway to the airport is useful if your regime is collapsing and you need to leave the country in a hurry.

• The process of democratization is an inherently good thing -- it’s a shame established democracies so often pay lip service to it but don’t actually mean it. Democratization doesn’t necessarily mean having a benevolent leader -- for instance (if I can trust the book’s example), Jerry Rawlings of Ghana was an authoritarian ruler who led his country down the path to democracy because it was the best option open for him to stay in power. Also, simply holding elections is a much less meaningful step on the road to democratization than is often assumed, and free elections should not be seen as the main objective. (Joshua Kurlantzick also makes this point in Democracy in Retreat.) The authors write, “Ultimately, elections need to follow expanded freedom and not be thought of as presaging it!” (p 278)