I – II – III – IV – V – VI – VII – VIII – IX – X – XI – XII – XIII – XIV – XV – XVI – XVII – XVIII – XIX – XX – XXI – XXII – XXIII – XXIV – XXV – XXVI – XXVII
[Single-page view]

Of course, even that argument – that liberal democracy is preferable to authoritarianism – has its occasional detractors. When it comes to solving difficult social problems, some people will argue that it’s better to have an autocratic government that can straightforwardly “get things done” in a unilateral way, without any institutional checks on its power that might hinder its aims, than to have a whole big bureaucracy that has to go to all the trouble of getting popular approval for its actions any time it has to do anything, and is therefore constantly gridlocked and unable to act effectively. Is there anything to this argument? Leaving aside the specific example of Singapore (which has characteristics of both democracy and authoritarianism) and just considering the matter in general terms, is it better to have a single ruling authority, like a monarch or a small core of party leaders, that makes decisions for everybody – the proverbial “benevolent dictator” – or to have decision-making power dispersed across the entire voting population, as in a democracy? We’ve spent a lot of time up to this point talking about all the things government can do and all the reasons it can be a valuable complement to the private sector – and hopefully we’ve adequately established that this is the case – but with that groundwork having been laid, it’s now worth turning to this question of how best to implement those functions and what form an ideal government should actually take. How can a government most effectively accomplish what’s truly in everyone’s best interest?

Well, probably the first thing we’ll have to grapple with here is that phrase “everyone’s best interest.” It seems safe to say that whatever form of government we choose, we won’t be able to make everyone happy all the time, just as a matter of sheer logistics. As Wheelan explains:

Making Public Policy Is Hard

Assume that some twist of fate has made you president of your homeowners’ association. Your primary responsibility is planning the annual block party, which involves arranging for a catered meal and a film that will be projected on a screen outdoors in a common area. The whole event will be paid for out of the annual budget, which comes from mandatory dues. The majority of your neighbors are genial, reasonable people, but there are a handful of folks apt to complain vociferously when they do not get exactly what they want, even when most other homeowners want something entirely different.

You can watch only one film at the block party—because there is just one outdoor screen and the movie is meant to be a communal experience. This is the first test of your leadership. The families with children are pushing for a Disney movie to keep their kids occupied during the party; the families without kids do not want to be subjected to a film about princesses or talking donkeys. The weird guy at the end of the street has generously offered to share his “world class adult movie collection.”

You sensibly opt against a porn film, as well as an anodyne children’s movie, and select a PG-13 adventure flick with decent appeal across the age spectrum (but arguably inappropriate for the youngest of the kids).

On the food front, your dinner committee has chosen grilled chicken, burgers, and hot dogs, and a pasta dish as a vegetarian option, though the two vegetarian families are still peeved that their annual dues are being used to support a banquet built primarily around grilling animal flesh.

Meanwhile, a sizeable minority of residents do not think you should have had the party at all. Yes, 70 percent of the homeowners voted to allocate the funds for the party at your last meeting, so you are delivering what most people want, but that still means 30 percent prefer not to spend money this way. And within the majority who supported the party, a minority was pushing for a lobster bake with a live swing band and ice sculptures.

In the end you will have one outdoor film, one meal, and one budget. Even if you offer superb leadership, adhere to the will of the majority, and make defensible decisions at every step in the process, the only possible outcome is that many people will be watching a film they don’t want to see, eating food they would not have chosen, and paying for a party they did not want in the first place. And if you had opted not to have the party a different and larger group of “constituents” would have been inflamed. That’s public policy.

I use this example when I am speaking in public, and the “That’s public policy” line always gets a laugh. But the homeowners’ block party, albeit somewhat silly, is public policy in the sense that it represents a set of shared decisions that are binding on the entire group, even those who vehemently disagree with the decisions. When there is only one possible shared course of action, we somehow have to come to agreement on what that course of action is going to be.

Most communal policy decisions are a heck of a lot more difficult to reach than choosing a dinner menu and a movie. Think about military engagements, such as the ones in Iraq and Afghanistan. We have one shared military, and we must agree as a country on how to use it.

Or consider social issues like abortion. Abortion is either legal or illegal. (Those who are in favor of legalized abortion would argue that “choice” represents a policy of agreeing to disagree; their opponents would point out that the aborted fetuses do not have much choice in the matter.)

Tax policy has the same basic challenge. It is not possible to have an income tax in which everyone pays what they believe to be fair.

We have hard collective decisions to make on these kinds of issues. The notion that politicians should stop bickering and do the “right thing” makes for fine cocktail-party banter, but it is, in fact, idiotic in a public policy context. Even in the block party example, there is nothing particularly helpful about the homeowner who stands up at the planning meeting and demands, “Let’s just do the right thing here!” Is that the steak or the vegetarian buffet? Is it the Disney movie or the porn film?

Different factions have profoundly different ideas about that “right thing,” and there is no scientific or empirical process for choosing one over another.

All of this stands in stark contrast to the private sector, which is driven entirely by mutually beneficial voluntary market transactions. No one has to agree on anything. I like coffee; I don’t need to get forty-nine other people to agree on the merits of coffee for me to walk into a Starbucks and order a cup. Similarly, Starbucks can make plenty of money by ignoring those who don’t like coffee, can’t afford coffee, live in countries where dictators won’t let them drink coffee, have disabling diseases that prevent them from drinking coffee, and so on.

I buy coffee on any given occasion because I believe it is the very best use of my money; Starbucks makes money for its shareholders by anticipating and meeting my needs. That is all that needs to happen.

The entire private sector operates around these self-organizing transactions, and that is a great thing. But the private sector—no matter how efficient—is not going to defeat the Taliban, keep a “dirty bomb” out of the country, feed or clothe the indigent, reduce carbon emissions significantly, or do anything else to address most of our serious communal challenges.

The notion that we can or should run government like a business is another goofy platitude that often passes for wisdom. The whole point of government is to do things that the private sector cannot or will not do. Businesses have the benefit of a bottom line. Success or failure can be measured with a single metric: profits. If the aforementioned block party were run like a business, all of the complexity of the decision-making would be solved with one simple analytical exercise: Which option makes us the most money? If that is a steak dinner followed by an adult-film marathon, then that is what the block party will look like. Too bad for the vegetarian families with kids.

The profit-maximizing approach is fine for Starbucks; people who do not like coffee can spend their money elsewhere. It does not work for any shared endeavor in which those who disapprove of some course of action cannot simply go elsewhere.

Public policy involves two inexorable realities: 1) We have no objective measure of the “best” course of action in many situations. 2) In such situations, many stakeholders have significantly different opinions on what the best course of action ought to be. If there is a “bottom line,” it’s that we can’t always get what we want, particularly when most other people want something else.

It is ironic that the Tea Party has idolized the self-indulgent (and relatively easy) act of hurling tea into Boston Harbor. The truly extraordinary and unique contribution of that generation of patriots was the Constitutional Convention, which was a long, grinding series of compromises, most of which were profoundly objectionable to some faction or another.

It’s common for certain anarchist types to underscore the necessity of unanimity in decision-making – to say that if a government makes any decision without the full unanimous consent of the people, it’s illegitimate. But the idea that it would be possible to have literally everyone in a community agree on every collective action decision is, to put it lightly, not very realistic. It might be one thing if we were just talking about a tiny group of friends who only had to decide on one or two issues together, like picking a meal or a movie for the evening. But as soon as we start talking about larger groups that have to continually make collective decisions about vital issues (even if the group is just the size of a single neighborhood, as in Wheelan’s example), it quickly becomes clear that maintaining total unanimity at all times simply isn’t a workable long-term strategy. Public policy involves thousands of collective action decisions, not just one or two – and they aren’t the kind of decisions that can just be dropped if no unanimous consensus can be reached (at least not without making the population worse off). So unless we want to constantly be running into situations like, say, a community agreeing on the need for a fire department, but failing to come to a unanimous consensus on where the fire station should be built, so that ultimately no fire station ever gets built at all, it’s simply inevitable that at least some of the time, some people will have to get their way while others don’t. Some anarchists might argue that unanimity could be achieved by splitting up society into millions of tiny micro-communities and having everyone move to whichever micro-community perfectly matched their political beliefs, so they’d never have to compromise with anyone who disagreed with them; but even if we imagined that it would somehow be logistically feasible to relocate everyone like this (which it definitely wouldn’t be), and even if we ignored the fact that people’s political beliefs are constantly evolving and changing over time (so they’d constantly be having to pack up and move any time they changed their opinion on anything, and communities would constantly be fracturing as new political issues emerged and divided public opinion), we’d still run up against the fact that this unanimity-only solution is itself something that not everyone would want to agree to in the first place. People have all kinds of reasons for wanting to live in particular communities aside from just agreeing with their neighbors politically; they typically prefer to live near their extended family members, for instance (whose political ideologies almost never match up perfectly with their own), as well as their friends, their employers, and so on (who likewise invariably have a diversity of beliefs themselves). So even if it were somehow possible to have as many political jurisdictions as there were combinations of political beliefs, that still wouldn’t be enough to get everyone to sort themselves into completely like-minded groups that always agreed internally on everything, because there would always be other considerations at play. For better or worse, there will always be reasons for people to live alongside others who disagree with them on at least some issues; and that’s something that the vast majority of people (at least the non-anarchist ones) willingly accept. In a sense, it’s the biggest thing they do collectively agree on – that they won’t always agree on everything, and they won’t always get their way, and that’s okay.

So all right then, if unanimous consensus isn’t a viable strategy for running a government, then what’s the best alternative for at least keeping the dissatisfaction to a minimum – having everyone vote democratically on how they want their government to be run, or just putting one person (or group of people) in charge and having them make all the decisions? For most of us, our instant answer will be that it’s obviously the former; if we all had to submit to the will of one person without ever having a chance to express our own preferences, that would strike most of us as a major injustice. Having said that though, we can’t necessarily justify this choice by just saying that democracy is always the best choice in every circumstance, and that it’s therefore the right answer by default; after all, there are plenty of situations in which giving a single person all the decision-making power really is the more reasonable choice. When you decide what kind of hairstyle you want to have, for instance, or what color socks you want to wear, it’s not a matter of putting the question up for a public vote and then doing whatever the majority says; you’re the only one whose vote counts, and rightly so. So then what’s the difference between this kind of situation and the broader context of public policy? As Michael Albert explains, it’s actually pretty simple:

Suppose that you have a desk in a workplace. You are deciding whether to place a picture of your child on that desk. How much say should you have? Or suppose that instead of a picture of your child, you want to place a stereo there and play it loudly in the vicinity of your workmates. How much say should you have about that?

There is probably no one who wouldn’t answer that as to the picture you should have full and complete say, but as to the stereo you ought to have limited say, depending on who else would hear the music and therefore be affected by your choice. And suppose we then ask how much say other folks should have? The answer, obviously, depends on the extent to which the decision would affect them.

The norm we favor is thus that to the extent that we can arrange it, each actor in the economy should influence economic outcomes in proportion to how those outcomes affect him or her. Our say in decisions should reflect how much they affect us.

In other words, when it comes to things like personal choices and private market transactions that don’t involve any outside parties, it makes perfect sense that individuals should have the power to make their own decisions unilaterally, without any kind of democratic process. They’re the only ones affected, so they’re the only ones whose preferences are really relevant. But by the same logic, when it comes to public policy – the kind of stuff that affects all of society – all of society should be involved in making the decisions, because everyone’s preferences are relevant. Their preferences might not always be what we’d consider ideal – and the policies they vote for might not always even turn out to be what’s in their own best interest – but as George Mitchell puts it:

Democracy doesn’t guarantee a good result, [but] it guarantees a fair process.

Of course, we can’t just gloss over the fact that voting does sometimes produce bad results, because that’s one of the biggest arguments that critics level against democracy as a concept. According to these critics, the fact that people don’t always make the most accurate judgments about what will best satisfy their preferences – the fact that they sometimes vote for candidates and policies that ultimately make them worse off – means that they could be made much better off if decision-making power were taken out of their hands entirely, and all the decisions were left to someone who actually knew what they were doing – i.e. a benevolent dictator. And in theory, this isn’t necessarily a false argument; in theory, everyone probably would be better off if all the decisions were made by someone who always knew exactly what was best for them and always acted in accordance with their best interests. But to say that this is a very big “if” is an understatement. When we look at all the real-world examples of autocracy in action, it very quickly becomes apparent why its apologists always have to add the qualifier “benevolent” when they use the word “dictator” – because benevolence is most definitely not an innate characteristic of dictatorship that can just naturally be assumed by default (nor is superior judgment). On the contrary, the word “dictatorship” has become practically synonymous with “oppressive regime,” not “enlightened leadership” – and that’s not a coincidence. As Bueno de Mesquita and Smith explain, it’s not just a random accident of history that autocratic regimes almost always end up serving those at the top, at the expense of the broader population, rather than the other way around; it’s a product of the fundamental structure of autocracy itself:

Money, it is said, is the root of all evil. That can be true, but in some cases, money can serve as the root of all that is good about governance. It depends on what leaders do with the money they generate. They may use it to benefit everyone, as is largely true for expenditures directed toward protecting the personal well-being of all citizens and their property. Much public policy can be thought of as an effort to invest in the welfare of the people. But government revenue can also be spent on buying the loyalty of a few key cronies at the expense of general welfare. It can also be used to promote corruption, black marketeering, and a host of even less pleasant policies.

The first step in understanding how politics really works is to ask what kinds of policies leaders spend money on. Do they spend it on public goods that benefit everyone? Or do they spend mostly on private goods that benefit only a few? The answer, for any savvy politician, depends on how many people the leader needs to keep loyal—that is, the number of essentials in the coalition.

In a democracy, or any other system where a leader’s critical coalition is excessively large, it becomes too costly to buy loyalty through private rewards. The money has to be spread too thinly. So more democratic types of governments, dependent as they are on large coalitions, tend to emphasize spending to create effective public policies that improve general welfare pretty much as suggested by James Madison.

By contrast, dictators, monarchs, military junta leaders, and most CEOs all rely on a smaller set of essentials. As intimated by Machiavelli, it is more efficient for them to govern by spending a chunk of revenue to buy the loyalty of their coalition through private benefits, even though these benefits come at the expense of the larger taxpaying public or millions of small shareholders. Thus small coalitions encourage stable, corrupt, private-goods-oriented regimes. The choice between enhancing social welfare or enriching a privileged few is not a question of how benevolent a leader is. Honorable motives might seem important, but they are overwhelmed by the need to keep supporters happy, and the means of keeping them happy depends on how many need rewarding.

Commenter fox-mcleod breaks it down this way:

Democracy is not the greatest system of government because it “takes into consideration what the population wants”. That’s like a nice bonus. The core mechanism that makes democracy valuable is that democracy diffuses power effectively.

Power corrupts. And democracy works by diffusing the corrupting influence across many millions in order to retard the inherent corrosion of a society’s institutions. Democratization of a system isn’t the aspect of putting things to a vote; rather it is the diffusion of power. Voting is just a means to an end.

[…]

Think about alternatives to a “democracy”. In any alternative system, to varying degrees power is concentrated to either a smaller group within the population or to a limited group or individual. But what is power and why can’t we have a “benevolent dictator”?

There’s a reason you don’t actually see the “benevolent dictator” system in the real world. Political power is essentially the quality of having other powerful people aligned to your interest. And those other powerful people get their power in turn from people further down the chain being aligned to them.

In order to keep those chains of alignment of interest, you have to benefit the people who make you powerful. But you have no need to benefit anyone else. In fact, benefitting anyone else comes at the cost of benefitting those who make you powerful. It’s a weak spot that can be exploited by a usurper. Right?

If you’re going to be a “benevolent dictator”, whose selfish interest do you need to prioritize in what order?

• tax collectors?
• military generals?
• educators?
• farmers?
• engineers?
• doctors?

Well without the military, you’re not really in charge and you can’t defend your borders or your crown from other potential rulers. And without the tax collectors you can’t pay the military or anyone else for that matter. But you can probably get away without educators for decades. So your priorities are forced to look something like this:

1. Military
2. Tax collection
3. Farming
4. Infrastructure projects
5. Medicine?
6. Education??

And in fact, any programs that benefit the common person above the socially powerful will always come last in your priorities or your powerful supporters will overthrow you and replace you with someone who puts them first. So it turns out as dictator, you don’t have much choice.

In short, even in the best-case scenario where an utterly selfless autocrat with the noblest of intentions somehow manages to come to power, the very structure of that power will make it nearly impossible for them to act in a way that puts the general population first, as opposed to serving the most powerful interests (or else being replaced by someone who will). By contrast, in a democracy, the general population is itself the powerful interest that can replace any leader who fails to meet its needs, so leaders must serve those needs first and foremost. As William Easterly writes:

Government institutions such as courts, judges, and police […] help make the market work in rich countries by enforcing contracts, protecting property rights, providing security against predators, and punishing lawbreakers.

The Achilles’ heel is that any government that is powerful enough to protect citizens against predators is also powerful enough to be a predator itself. There is an old Latin saying that goes, “Quis custodiet ipsos custodes?”—which translates freely as “Why would you trust a government official any more than you would a shoplifting serial killer?”

Democracy’s answer to “Who will watch the watchers?” (the more conventional translation of the above) is everyone. The other great invention of human society besides free markets is political freedom. According to the simplest view of democracy, an open society with a free press, free speech, freedom of assembly, and political rights for dissidents is a way to ensure good government. Free individuals will expose any predatory behavior by bad governments, and vote them out of office. Voters will reward with longer terms of office those politicians who find ways to deliver more honest courts, judges, and police. Political parties will compete to please the voters, just as firms compete to please their customers. The next generation of politicians will do better at delivering these services. Of course, no democracy works anywhere close to this ideal, but there are some that come close enough to make [the improvement of their society] possible.

Mind you, none of this is to say that it’s literally impossible for any political leader to ever overcome all the adverse structural incentives and become something like a benevolent dictator; such aberrations have happened on rare occasions throughout history. Once again, people will often cite Singapore here as a modern example of this, with its founding leader Lee Kuan Yew having used his unusual degree of unilateral power to significantly improve his country when it was first getting off the ground. But as with our earlier discussion, the people who bring up Singapore in this context will typically have a hard time coming up with a whole lot of other modern examples aside from just that one – and there’s a good reason for that. Government power that’s not accountable to the people tends to result, unsurprisingly, in government actions that don’t primarily serve the people.

(People will also sometimes cite China’s ruling party as a modern example of an undemocratic government using its unilateral power to make the country significantly better off – but this is more of a stretch, considering that the best thing the party has done for China over the last half-century has been to ease up on its authoritarianism and let its people engage in more free market activities. In reality, it’d be more accurate to describe China as a country that has improved despite the authoritarianism of its government than as one that has improved because of it.)

Leaving aside this whole problem of adverse incentive structures, though, there’s an even more basic issue when it comes to autocracy, which is that even if a benevolent dictator has both noble intentions and the ability to act on them, that still does nothing to guarantee that their good intentions will actually translate to good policy judgments, much less ones that are superior to all other alternatives. Putting a crown on someone’s head, after all, does nothing to change the fact that they’re still flawed humans capable of getting things wrong. And while it’s of course true that the typical voter also gets things wrong all the time, the key difference is that in a majoritarian democracy, the only way for such a mistake to become a national policy failure is if over half the voting population is simultaneously wrong about it – while in an autocracy, all it takes is for the one person at the top to be wrong. There’s just a single point of potential failure around which the entire system hinges – so if and when that one person does make a mistake, it can throw the whole country into turmoil. And because there are no democratic checks on their power, there’s nothing that the people who ultimately suffer the damage can do about it.

But how likely is it really that the kind of benevolent dictator we’re talking about will be more prone to such mistakes than the median voter? If we’re really talking about a leader who’s well-educated and experienced in statecraft, whose intentions are good, and who knows how to get things done, then won’t they be much more likely to get things right overall? Well, sure, if we’re going to rig the question by stipulating that we can only ever have this kind of perfectly ideal autocrat, then of course it’s easy to imagine how they might produce better outcomes. Again though, here in the real world there’s simply no such mechanism for ensuring that autocrats will be anything close to that kind of ideal – and in fact, the general rule has historically been the exact opposite. As it turns out, the kinds of rulers who want to have the power to control entire populations without having to account for their preferences tend not to be the kinds of leaders who have the most extraordinarily superior judgment; more often, they’re just the kind of leaders who think their judgment is always perfect, and who are therefore more likely to stubbornly persist in their misguided beliefs even when they produce abysmal real-world outcomes. Alexander provides a list of examples here to illustrate the point, from Ivan the Terrible – whose paranoia and megalomania led him to destroy his own economy, burn and pillage his own cities, and torture and massacre thousands of his own people – to Henry VIII – whose decision to convert his whole country to a newly-invented religious denomination just so he could marry Anne Boleyn (whom he would later have beheaded) led to thousands of deaths and a series of wars that would last for generations. Alexander concludes:

This is exactly the sort of problem non-monarchies don’t have to worry about. If [an American president] said the entire country had to convert to Mormonism at gunpoint as part of a complicated plot for him to bone Natalie Portman, we’d just tell him no.

There’s another important aspect here too. Reactionaries – ending up more culpable of a stereotype about economists than economists themselves, who are usually pretty good at avoiding it – talk as if a self-interested monarch would be a rational money-maximizer [suggesting that such a monarch would presumably want to maximize their country’s wealth as well so they’d have a bigger tax base]. But a monarch may have desires much more complicated than cash. They might, like Henry, want to marry a particular woman. They might have religious preferences. They might have moral preferences. They might be sadists. They might really like the color blue. In an ordinary citizen, those preferences are barely even interesting enough for small talk. In a monarch, they might mean everyone’s forced to wear blue clothing all the time.

You think that’s a joke, but in 1987 the dictator of Burma made all existing bank notes illegitimate so he could print new ones that were multiples of nine. Because, you see, he liked that number. As Wikipedia helpfully points out, “The many Burmese who had saved money in the old large denominations lost their life savings.” For every perfectly rational economic agent out there, there’s another guy who’s really into nines.

Continued on next page →