A New and Arguably Better Way of Doing Things

Specifically, making political decisions

"Ninety percent of everything is crap."

—Sturgeon’s Law

Among the modern flavor of Very Serious People, particularly those in academia, there seems to be widespread agreement that the way we do things currently is very bad. You’ll find (some) fascists, communists, anarchists, monarchists, and every other flavor of strong political view. Heck, communists in academia are perhaps as common as Republicans, particularly in some fields.

And yet there is one ideology never seen in academia. In fact, it’s not clear it can even be called an ideology. After all, would we call the view that we should accept any moral view that we hear first, and reject any that come after it a moral view? Can such a set of views that never dusts the halls of the cathedral truly be called an ideology?

It’s unclear whether that which is not worth naming can be usefully referred to as a concept. After all, the segment of air in front of me lacking a name isn’t conceptualized as a thing. Nor is the random star shaped segment of my desk that hasn’t been considered distinct from the rest of the desk up until this point.

But I digress—as interesting as such points are about meaning and semantics, they’re not directly relevant to the topic at hand. No, the project here is much grander. We are, after all, figuring out how to make decisions go better.

So what is this ideology that no one defends? “Get to it man,” I can hear the hypothetical reader saying.

It’s something I shall refer to as status quoism. This idea is, as the rather banal title suggests, the view that we should generally keep the status quo. Note for the one confused reader, this is not the ideology I’ll defend in this article. After all, the status quo cannot be a better way of doing things. How can the status quo be better, after all, than the status quo?

For all their talk of conserving things, Chesterton’s Fence, and the inherent harms of change, conservatives are not interested in conserving the status quo. They don’t, for example, want to conserve Obamacare. Some aren’t interested in conserving anything, others are interested in conserving the way that things were. Yet no one seems to like the way things are now. It seems everyone agrees that the status quo is indefensible.

Yet why is status quoism not an ideology defended by anyone? There are, after all conservatives. There are monarchists too, who want to bring back monarchy. Surely the status quo is more defensible than monarchy? And yet despite that, no one defends the status quo. Why is that?

Perhaps it’s because the status quo is indefensible. And yet that doesn’t seem to be a very satisfactory account. The status quo was indefensible in communist countries, yet there were defenders of the status quo. Additionally, the status quo seems pretty good. Quality of life is improving after all.

Perhaps, the reason is that the status quo wouldn’t make a coherent ideology. After all, ideologies have to have a common theme running through their political views. Bidenism, for example, isn’t much of an ideology. Yet this account is also unsuccessful (though it’s similar to the correct account, as we shall see). One could very easily imagine a type of status quoism.

After all, status quoism gives across-the-board prescriptions about policy—keep things as they are! One could very easily imagine status quoist manifestos relying on their views in political theory—employing Chesterton’s fence and defending status quo decision-making as a generally good mechanism that we oughtn’t deviate from. We don’t criticize letting Doctors make medical decisions, even though Doctors sometimes make bad medical decisions because they make better decisions than others.

Take a moment to really consider the question before reading the next paragraph, when I give it away. I would, of course like my readers to think for themselves.

No, the reason status quoism is not an ideology is a lot like an argument I’ve employed against god, namely the problem of maximization. The problem of maximization objects to the existence of god on the basis of the universe not being optimized to maximize for anything. After all, if the universe was designed by a paper-clip maximizer, it would be purely paperclips. If it were made by a utilitarian, it would be pure utility. Thus, the fact that the universe is a confusing mishmash of things, not clearly optimizing for anything serves as strong evidence against god.

If there were a perfectly good god, the universe would be optimized for goodness. There would be nothing unnecessary, no excess stuff. No unnecessary stars, planets, and certainly no unnecessary evil. The universe would purely be good things. It would be filled to the brim with good things. Heck, it probably wouldn’t even have a brim—just infinite goodness.

Why does this mean that no one is a status quoist? Well, status quoism would be like this universe in that it’s not optimizing for anything in particular. The status quo neither maximally furthers the goals of the left or the right. It’s a confusing mix of lots of different things. Virtually no one would think that the optimal healthcare policy happens to be that which exists in each of the 50 states.

Thus, there is a coherent political view that involves supporting, communism. There is not for the status quo, which is not optimizing for anything in particular. That’s why Stalin has more passionate modern defenders than the status quo. No one holds up signs advocating keeping things as they are across the board.

And yet this also goes to show one significant problem with modern political ideologies. They tend to be fairly simple, applying a simple formula to all political issues. Yet as we know from Tetlock’s research, it’s better to be a fox than a hedgehog. It’s better to not see the world through a single frame. The correct assemblage of political views will not fall along simple partisan lines. It will instead be a mix of lots of different things. We have no reason to expect there to be a correlation between left wingers being correct about abortion and them being correct about trade policy.

Another problem plagues nearly every modern political ideology. This poses a general problem for radicalism in ideology. Well, it’s not clear if we can really call it a problem, it operates more as a higher-order consideration than a specific problem.

This problem is…

drum-roll please…

The general problem of political knowledge. Political knowledge is exceedingly difficult and most people are deeply wrong. Why think this? I have to credit Huemer for one line of evidence for this conclusion. What would we expect if political knowledge was exceedingly difficult and we weren’t rationally discovering political knowledge? If political knowledge was more like a sports team, a political tribe, even a religious affiliation, than a genuine intellectual project?

A) We’d expect people to hold very strong views. After all, people might have weak leanings when it comes to views on philosophy of language. But when issues are complicated, most informed people don’t have excessively strong opinions. Think about how ludicrous it would be to, for example, see someone holding up signs talking about the many worlds interpretation of quantum mechanics being obvious. Okay, perhaps Eliezer Yudkowsky would, but no one else!

And yet many political issues are arguably more difficult than many worlds. Despite this, average people who know very little lacking specific knowledge about it or significant mental ability have very, very strong views about politics. Views strong enough that they put it on shirts, signs, and even yell at you on college campuses about them.

So on the hypothesis that no one knows what they’re talking about politically, we’d expect people to be extremely adamant about and confident in their political views. This is exactly what we observe. Thus, this is evidence for the theory. It should factor into the consideration of any good Bayesian! And if we here at the bentham’s bulldog blog are anything (other than utilitarian), we’re good Bayesians.

B) We’d expect political disputes to be either impossible or very difficult to resolve. After all, people who apportion their beliefs to the evidence are much more likely to change their minds when evidence comes along. People often change their minds, about philosophical topics in which they are not deeply emotionally invested. They are much less likely to change their minds on topics that they tie to their identity.

Thus, the intractability of real world political disagreement serves as more evidence for the theory. The fact that reasoning seems inadequate to settle such disputes seems to confirm the theory rather strongly.

C) We’d expect political belief to correlate much more with non cognitive traits. After all, it politics is not about genuine information but is rather about signaling to ones tribe, then we’d expect different demographics to have different political views. This is exactly what we see. Racial minorities, members of the entertainment industry, and people in academia have very different political views from the rest of people.

So we have even more evidence for the theory. And yet the evidence can keep pilling up. It’s even more decisive than it currently appears. We are just getting started.

D) We’d additionally expect political beliefs to form loosely related, and in fact opposite clusters. After all, in philosophy while we see a correlation between some related views (like non-naturalism and platonism), we don’t see the same type of correlation between wholly unrelated views (like philosophy of language and philosophy of mathematics). And yet in politics, we do see that.

Even if it were the case that the liberal view, for example, was correct across the board, that wouldn’t be able to explain why some people are conservative across the board. What are the odds that very smart conservatives would reason about every single distinct political issue—and get every single one of them wrong?

In fact, we’d seem to expect a priori to have several views be correlated in the opposite direction from which they actually are correlated. We’d expect those who oppose immigration to be more supportive of big government, and less supportive of free markets. Yet what we see is exactly the opposite.

I’ve already expressed similar such sentiments here.

Additionally, despite people’s very strong views, they seem wholly unaware of what they’re talking about. As Huemer once again points out (his paper was sufficiently good that it’s worth quoting at length).

First, a related theory. The theory of Rational Ignorance holds that people often choose—rationally—to remain ignorant because the costs of collecting information are greater than the expected value of the information.1 This is very often true of political information. To illustrate, on several occasions, I have given talks on the subject of this paper, and I always ask the audience if they know who their Congressman is. Most do not. Among senior citizens, perhaps half raise their hands; among college students, perhaps a fifth. Then I ask if anyone knows what the last vote taken in Congress was. So far, of hundreds of people I have asked, not one has answered affirmatively. Why? It simply isn’t worth their while to collect this information. If you tried to keep track of every politician and bureaucrat who is supposed to be representing (or serving) you, you’d probably spend your whole life on that. Even then, it wouldn’t do you any good—perhaps you’d know which politician to vote for in the next election, but the other 400,000 voters in your district (or the 200,000 who are going to turn up to vote) are still going to vote for whomever they were going to vote for before you collected the information. Contrast what happens when you buy a product on the market. If you take the time to read the Consumer Reports to determine which kind of car to buy, you then get that car. But if you take the time to research politicians’ records to find out which politician to vote for, you do not thereby get that politician. You still get the politician that the majority of the other people voted for (unless the other voters are exactly tied, a negligible possibility).2 From the standpoint of selfinterest, it is normally irrational to collect political information.

The data does bare out this theory—one echoed by Churchill’s famous quote about the best argument against democracy. Unfortunately, it turns out that quote wasn’t said by Churchill, but it’s a good quote nonetheless.

A 2010 survey found that most voters didn’t know who controlled the house.

As Crain points out

Roughly a third of American voters think that the Marxist slogan “From each according to his ability to each according to his need” appears in the Constitution. About as many are incapable of naming even one of the three branches of the United States government. Fewer than a quarter know who their senators are, and only half are aware that their state has two of them.

The excellently named Campos points out

This sounds like a surreal joke, but then stories about contemporary politics often do. There is no polite way to phrase this: When it comes to politics, the average person is an idiot. Depressing evidence for this claim can be found in a recent New Yorker essay by Louis Menand, which surveys the political-science literature regarding why people vote the way they do.

The conclusions from this literature include:

— No more than 10 percent of the population can be said to have a coherent political belief system, using even a loose definition of that term. Most people's political beliefs, to the extent they have any at all, suffer from a lack of what political scientists call "constraint" — i.e., little or no logical connection exists between the positions they hold. For example, a large proportion of voters see no contradiction between being in favor of both lower taxes and increased government services.

— Perhaps a quarter of all voters vote on the basis of factors that have no "issue content" whatever. They vote for candidates who seem likable, or optimistic, or for those whose campaign posters are particularly eye-catching. According to Princeton political scientists Christopher Achen and Larry Bartels, millions of voters in the 2000 presidential election based their votes on what the weather had been like lately.

— Voters are remarkably bad at calculating their own self-interest, even when their self-interest and their political beliefs coincide. Bartels gives the following example. Only the richest 2 percent of Americans pay estate taxes. Yet among people who believe that the rich ought to pay more taxes, and who also believe that growing income inequality is a bad thing, two-thirds also favor repeal of the estate tax. Menand observes that this sort of data helps explain the otherwise puzzling fact "that the world's greatest democracy has an electorate that continually 'chooses' to transfer more and more wealth to a smaller and smaller fraction of itself."

Even if we ignore how many people have no coherent political beliefs, or base their voting on irrational factors, the sheer ignorance of the average American should take us aback. Seventy percent of Americans can't identify their senators or their congressman. Around 30 million can't find the United States on a map.

The Annenberg public policy center points out

• More than half of Americans (53 percent) incorrectly think it is accurate to say that immigrants who are here illegally do not have any rights under the U.S. Constitution;

• More than a third of those surveyed (37 percent) can’t name any of the rights guaranteed under the First Amendment;

• Only a quarter of Americans (26 percent) can name all three branches of government.

So this seems to be a rather consistent theme. The average voter just doesn’t know much. They seem deeply confused and badly misinformed across the board.

And on top of that, they don’t care much about the more important issues. The bombing of Sudan’s only medicine factory, greenlighting of the Rwandan genocide, specific sanctions regimes which (while disputable how many killed) was deemed worth it by Albright even under the assumptions that it killed half a million small children, broad sanctions regimes which arguably kill millions, reduction of the numerous existential risks, factory farms, all are largely ignored by the American public. It’s thus no coincidence that lots of very terrible policies pass. 81% of economists support market based solutions to environmental policies like pigouvian taxes, yet they’re rarely implemented. In the rare cases when they do, well—people tend to get rather upset about them. They don’t like the taxes, even though carbon taxes are arguably less regressive than other actions to combat pollution.

The information people are bombarded with for news consumption is largely irrelevant—culture war issues of little concern. People entirely ignore the problem of inadequate innovation, despite it being widely regarded by economists as one of the biggest problems. New, innovative ideas are ignored by governments, even ones proposed by Nobel prize winners. FDA delays which kill lots of people, yet are ignored politically.

Is the average voter really fit to make complex political judgements about Ukraine? About NATO? About tariffs or FDA policy or environmental policy, or really anything else? Probably not.

This is why it’s agreed upon that everything sucks to some degree—that many political decisions made are very bad. If you put the blind mad god behind the whims of unreflective humans in charge of making complex political decisions about issues they don’t care about, regarding countries they’ve never heard about and certainly couldn’t locate on a map, in terms of issues where brilliant scholars are in fierce disagreement—well, don’t be surprised when you end up with Al-Sheifa. Or the Iraq war. Or any of the whole host of other disastrous policies done by the U.S..

When the attention span of the average voter is less than that of small children with ADHD—our politics is unlikely to be sculpted by prolonged reflection. Rather, it would be expected to be an eclectic mix of policies—many good, many bad, and many just nuts. And that’s exactly what we see! When most people don’t know what Sudan is, it’s no wonder they don’t care about the U.S. casually committing a 9/11 esque terror attack on their medical facilities.

This explains why our modern political state is something of a disaster. The average voter is no more equipped to make political decisions than they are to make medical decisions. And we don’t let democracy determine our medical decisions.

The reason why nuclear energy is regulated into oblivion is because people have a vivid mental picture of Chernobyl and it sounds scary. That’s it. That’s what holds back a form of clean energy that would save lots of lives.

This also opens up a problem with general political knowledge. Those who are advocating for major shifts in the political system don’t adequately consider the possibility that they’re wrong. Just because it seems to them that particular policies are good does not mean they would in fact be good. The problem is less one of the particular decisions made and more of how decisions are made.

When considered from an outside view, the probability that the major political shift you advocate is a bad idea is reasonably high. We have lots of examples of that being the case. The Russian revolution turned out to be rather imperfect, to say the least. As did the Arab Spring.

Perhaps there’s a single, simple political organization that would make things go best—one that we just need to discover. However, I suspect, in line with Tetlock’s research, that the true optimal slate of policies won’t be the type of thing easily summarized—rather, it will be a confusing assemblage of different policies. We have no reason to, for example, expect there to be much correlation between the optimal U.S. policy towards Iraq and that towards Brunei.

Thus, those who want to shift radically in a particular political direction have their sights set on the wrong goal. They look for a solution rather than looking for a way of finding solutions. This is a rather big problem. The solution to politics is something too complicated for any single person. The political terrain is just too big and varied. No person has the time to sit through and read unqualified reservations articles, for example, to see if there’s a grain of truth in the things said.

Would any single person really be fit to comment on the views of Kotz, Hanson, Friedman, and every other person who has proposed a radical political prescription—and read their critic’s replies, and their replies to critics and so on to a great enough competency to form detailed evaluations? Could we really be sufficiently confident in their verdicts to put their drastic proposals into detail—when history books are littered with the blood and bones of those whose lives were cut short by radical new proposals?

This leads to an important caveat—before trying the idea on a large scale, it should start out small. While I think the proposal does have considerable advantages over other failed radical proposals for change—things always have the potential to backfire. (If only Marx had included that small caveat).

This idea is not original to me—its progenitor is the economist Robin Hanson. The idea is futarchy. The slogan for futarchy is vote values, bet beliefs. It works in the following way. We start with a metric for determining how well things are going. An example of a metric would be GDP+ which includes both GDP, people’s subjective assessment of their quality of life, and environmental quality.

Once a metric is set up, betting markets would be set up that bet on what the GDP+ would be if policies pass vs if they don’t pass. Thus, one could purchase essentially a stock that pays out some percentage of the GDP+ conditional on policies passing. They could also purchase a stock that pays out some percent of the GDP+ conditional on the policies not passing. If one purchases a stock conditional on the policy passing but the policy doesn’t pass, they’d get refunded. If the estimated stock value was higher conditional on the policy passing than on it not passing, then the policy would automatically pass.

Thus, bettors have a financial incentive to do rigorous research and decide which policies are worth passing. This creates a financial incentive to decide upon good policy and bet on the good policy. If we want people to have an incentive to do what is in the best interest of society, giving bettors an incentive to make good policy is a fantastic idea.

Thus, we have theoretical reason to expect betting markets to work well. But there’s also a significant body of empirical data supporting this. Hanson notes

While government policy may often suffer info failures, speculative markets show striking info successes. Most markets for stocks, bonds, currency, and commodities futures are called speculative markets because they let people speculate on future prices by buying or selling today in the hope of reversing their trades later for a profit. Such opportunities to “buy low, sell high” occur when identical durable items are traded frequently in a market with low transaction costs. Given such opportunities, everyone is essentially invited to be paid to correct the current market price, by pushing that current price closer to the future price. Such invitations are accepted by those sure enough of their beliefs to put their money where their mouth is, and wise enough not to have lost their money in previous bets. Betting markets are speculative markets trading assets designed let people speculate on particular matters of fact, such as which horse will win a race. Final bet asset values are defined in terms of later official judgments about the facts in question. By construction, such assets are durable, identical, and can be created in unlimited supply. Betting and other speculative markets have been around for centuries, and for decades academics have studied their info properties. The main robust and consistent finding is that it is usually quite hard, though not impossible, to find info not yet incorporated in speculative market prices.xvi In laboratory experiments, speculative markets usually aggregate info well, even with four ignorant traders trading $4 over four minutes.xvii

He later notes

The key issue is not absolute accuracy, but accuracy relative to other institutions, on the same topics, given similar resources. We have some data on this. In addition to lab studies, a few studies directly compare real speculative markets with other real info institutions. For example, racetrack market odds improve on the prediction of racetrack expertsxxii; orange juice commodity futures improve on government weather forecastsxxiii; stocks fingered the guilty firm in the Challenger crash long before the official NASA panelxxiv; Oscar markets beat columnist forecastsxxv; gas demand markets beat gas demand expertsxxvi; betting markets beat Hewlett Packard official printer sale forecastsxxvii; and betting markets beat Eli Lily official drug trial forecasts.xxviii The largest known tests compare bet prices to major opinion polls on US presidential elections. For example, in 709 out of 964 comparisons, bet prices were closer to the final result.xxix This gain comes in part from prices being disproportionately influenced by active traders, who suffer less from cognitive biases.xxx

Thus, we have significant evidence that current institutions make very bad decisions and betting markets do much better. This establishes a strong prima facie case for futarchy. However, there may be some problems with futarchy that make it unworkable. Let’s explore some potential problems.

One might worry about market manipulation. Such fears are, however, misplaced. Rhode notes

In every manipulation that we study there were measurable initial changes in prices. However, these were quickly undone and prices returned close to their previous levels. We find little evidence that political stock markets can be systematically manipulated

Around the 2012 election, someone tried to manipulate betting markets by spending 60,000 dollars in one night. However, within a few hours, prices reverted back to normal.

The reason for this is simple—if betting markets are manipulated, then making a rational bet pays out more. Thus, as Hanson notes, market manipulation makes markets more accurate in lab experiments—betting accurately becomes more profitable.

One might worry that it’s impossible to adequately decide on values. This concern is misplaced. For one—policy makers are already optimizing for a metric; votes of uninformed people. Thus, even if imperfect, futarchy would have a better utility function for policy makers.

Hanson notes

It seems reasonably clear, for example, that on average people in Ethiopia today are less satisfied with their lives than people in Sweden today. This can be true even if we would find it difficult to advise any random Ethiopian on what he should eat for lunch. The development of national accounts, i.e., the collection of statistics such as GDP, has been justifiably called one of the greatest economic innovations of the twentieth century. While such measurements are made with error, and leave out a great deal, they seem to be a sufficient basis for many policy recommendations. For example, most researchers in empirical growth economics seem willing to presume that policies are good if they causally induce sustainably higher GDP. And frequent travelers find it hard to escape the impression that, comparing nations with large differences in measured GDP, most people who live in the high GDP nations are richer and better off than most people who live in the low GDP nations.2 Furthermore, economists spending modest budgets have already explored many promising extensions to current GDP measures, such as lifespan, home production, leisure, environmental assets (Boskin, 2000; Nordhaus, 2000). Economists also have well developed frameworks for discounting future welfare, and for dealing with risk aversion and inequality. Direct measures of happiness can even be included (Oswald, 1997). It therefore seems plausible that with greatly increased funding and the full attention of our elected representatives, we could devise even more accurate and robust measures of national welfare. Consider a person who is willing to recommend a certain policy in a given situation because statistical analysis suggests that those policies tend to cause GDP to rise in such situations. Such a person should also be willing to recommend this policy if speculative markets were to estimate that this policy was expected to increase GDP in this situation. Speculative market estimates are probably no less accurate than a typical statistical analysis. After all, if a statistical analysis is persuasive then speculators should be persuaded by it, and if speculators disagree with a statistical analysis they probably have good reasons. Therefore, if one is willing to recommend policies that statistical studies suggest will increase (a time average of future) GDP, one should be willing to recommend policies that speculative markets estimate will increase GDP, and so one should be willing to consider a form of policy making which relies more on such market estimates in choosing policies. While GDP leaves out many things, nations that consistently adopt the best possible policies for increasing (a time average of future) GDP would in a few decades probably be much richer (and happier) than they would otherwise have been. And one should be even more willing to consider basing policy on speculative market estimates of their effects on GDP+, i.e., an improved measure of national welfare.

Hanson gives 5 possible metrics.

options include:

1) As I said above, the most conservative option is to have an elected legislature vote on edits to an existing explicit function. Because that’s the sort of thing we do now.

2) A simple, if radical, option is to use a “market value” of the nation. Make all citizenships tradeable, and add up the market value of all such citizenships. Add that to the value of all the nation’s real estate, and any other national assets with market prices. With this measure, the nation would act like an apartment complex, maxing the net rents that it can charge, minus its expenses. (A related option is to use a simple 0 or 1 measure of whether the nation survives in some sufficient form over some very long timescale.)

3) A somewhat more complex option would be to define a simple measure over possible packages of measured quantities, then repeatedly pick two random packages (via that measure) and ask a random citizen which package they prefer. Then fit a function that tries predict current choices. (Like they do in machine learning.) Maybe slant the random picks toward the subspaces where citizen choice tests will add the most info to improve the current best fit function.

4) An option that requires a lot of complexity from each ciziten is to require each citizen to submit a welfare function over the measured quantities. Use some standard aggregation method to combine these into a single function. (For example, require each function to map to the [0,1] interval and just add them all together.) Of course many organizations would offer citizens help constructing their functions, so they wouldn’t have to do much work if they didn’t want to. Citizens who submit expensive-to-compute functions should pay for the extra computational that they induce.

5) Ralph Merkle (of Merkle-tree fame) proposed  that “each citizen each year report a number in [0,1] saying how well their life has gone that year”, with the sum of those numbers being the welfare measure.

To conclude, futarchy might go terribly wrong. Radical shifts to governance often backfire. But putting aside these meta considerations, we have lots of object level reasons to support futarchy. Democracy may not be the best system of government.