Monthly Archives: December 2019

Governance By Jury

Among the many proposed forms of governance, some are “direct democracy” wherein all citizens vote on key choices, and some are variations on “demarchy”, i.e., assigning key roles to, or filling legislatures with, random citizens. The following proposal is similar in some ways, but seems different enough to be worth treating separately. I’m not sure if “jurarchy” is a good idea, but it seems to me simple and elegant enough to be worth considering.

Here is an especially simple version, though variations (some discussed below) may be better:

There is always a status quo set of government policies, including who sits in each key role. At any time, anyone can propose a change to these policies, if they pay fee $A. A court case then ensues, overseen by a random judge and decided by a random jury of N citizens. A key government agency is charged with defending the status quo in these cases. The judge can declare the proposal unconstitutional, or say that recent changes have invalidated it. But if not, and if M jurors support the proposal, then it becomes official policy, and challenger is awarded bounty $B.

And that’s it; everything is decided this way (aside perhaps from constitution changes). If the cost of pursing a case is $C, then we expect such challenges to be made from purely financial motives when the chance P of winning the case exceeds (A+C)/B.

Of course we might want some jury rules, such as no bribes to buy juror votes. Jurors might or might not be allowed to consult outside advisors, and might or might not be told of jury decisions on recent similar cases. Jurors might be chosen new for each case, or they might learn via sitting on juries that work together on many cases over many months.

One potential problem with the above system is that parties who stand to gain a great deal from a policy change may keep re-trying the same proposal until they happen to get a favorable jury. If they gain $G from the change itself (not via bounty), and if juries make bad decisions at error rate E, then this approach is profitable on ave when E*(B+G) exceeds A+C. Observers who believe a change was made in error would expect to profit by proposing a reversal. But is this solution enough?

A futarchy-based variation might help here. After a jury has ruled in favor of a proposal, we could immediately open up a betting market on the chance that another random jury would also favor that proposal, in a new court case. This new case might use the same values of A,B,N,M, or it might scale these up in the hope of getting a more considered judgment. This new case might be created with chance F. The original jury decision might be said to be confirmed, and implemented, only if this betting market estimated at least a conditional chance Q of confirmation. Yes, markets can also make mistakes, so this in essence just lowers error rate E.

Another potential problem is that this jury process might be too slow to make key changes. To deal with this, we might create a similar betting market as soon as a proposal is officially made, about that first jury process. A proposal might be immediately adopted if that market estimated at least a chance Q’ of the proposal winning.

I’m sure we could think of more problems, and more potential fixes. But there’s a real risk of fixes making things worse, especially as the system gets more complex, and as the citizen audience who must oversee it gets bored with complex details. So I’m attracted to very simple proposals, and tempted to just accept modest problems, instead of adding many complex fixes.

See also: Explainable Governance.

GD Star Rating
Tagged as: ,

Injustice For All

In their new book Injustice for All: How Financial Incentives Corrupted and Can Fix the US Criminal Justice System, Chris Surprenant and Jason Brennan suggest many ways to change the US crime system.

They spend the most space arguing against jail; they want to cut long jail terms, and to offer most criminals a choice of jail or non-jail punishments such as caning. (I also dislike jail.)

This and most of their other suggestions can be seen as fitting a theme of favoring defendants more, relative to government. For example, they want a lot fewer acts to be punished at all, more bad acts to be punished as torts instead of as crimes, loser pays lawyer/court costs, crime law to be clear and simple, a requirement to show the accused could easily know act was criminal, no cash bail, no private prisons, no asset forfeiture, fewer no-knock raids, the same lawyers and resources given to public defense as to prosecution, juries to choose between punishment plans offered by protection & defense, notifying juries of their jury nullification ability, and more grand juries before and during trials who can cancel trials.

While this theme is quite popular today, I’m wary of this focus on changing policy to favor defendants over government. Yes the pendulum may now favor government too much, but someday it will swing the other way, and I’d like to do more than just help push this one pendulum back and forth.

Many other suggestions in the book fall under a theme of spreading out incentives, to make incentives weaker for any one party. These authors attribute many current problems to overly strong incentives, such as that induce small towns to make speed traps. They want government-managed victim restitution funds, no elected judges or prosecutors, local governments to pay more for jail costs, state governments to pay more non-jail costs, and no revenue given to police agencies based on particular cases. And they suggest that the state pay for investigate torts:

For most tort claims, the state would need to bear the responsibility and financial cost of collecting and processing evidence, as well as finding and interviewing witnesses. This information would then be available to both the would-be plaintiff and defendant.

Instead of having the state manage tort investigations, I’d rather we did more to ensure tort damages can be paid, perhaps by adding bounties. Then we could rely more on private incentives to investigate well, instead of trusting the state to do that. More generally, I want to introduce stronger elements of paying for results into criminal law, instead of just weakening incentives all around to avoid bad incentive problems.

Below the fold are many quotes from the book: Continue reading "Injustice For All" »

GD Star Rating
Tagged as: ,

Capitalism Uses Hate; That’s Good

“Good! Your hate has made you powerful. Now, fulfill your destiny.” (More)

The most natural human social structure is based on prestige. People compete to look impressive, and then everyone defers to those who seem most impressive. We let them run the things they want the way they want, if only they will let us gain some prestige via association with them. Which is often a big problem, as in the modern world the way to look most impressive is often not the best way to run things.

When the way to seem an impressive doctor is not the best way to heal patients. When the way to seem an impressive lawyer or judge is not the best way to win or rule on cases. When the way to seem an impressive warrior is not the way to win wars. When the way to seem an impressive cook is not to make cheap tasty nutritious food. In such cases, letting the most prestigious folks do things their way can lead to wasteful inefficient outcomes.

In “capitalism”, big firms are run by rich greedy bossy managers in the service of even richer and greedier owners. For many, a natural ancient human reaction to such a situation is “hatred.” Or at least strong distrust, wariness, and suspicion. Many of us are primed to think the worst about these people and this situation.

Which is great, because this enables us to hold such people and firms accountable. We are willing to switch from firms who supply us with products and services when other options look better. We are willing to quit jobs we don’t like, and go home when we feel done for the day. And when firms fail to satisfy customers and employees, we are willing to let those firms die, and let their investors lose their shirts. Because we hate them.

Unfortunately, our hate also makes us more willing to regulate such firms, and to take from such people. Some regulation and taking may be useful, but too much can kill or at least emaciate the goose that lays the golden eggs of capitalism. Our related suspicions of big powerful politicians and their supporting organizations helps to mitigate this problem somewhat, but alas it seems we don’t hate such people and orgs remotely as much as we should.

Beware of love; sometimes hate is what we need.

GD Star Rating
Tagged as: ,

Might Disagreement Fade Like Violence?

Violence was quite common during much of the ancient farming era. While farmers retained even-more-ancient norms against being the first to start a fight, it was often not easy for observers to tell who started a fight. And it was even harder to get those who did know to honestly report that to neutral outsiders. Fighters were typically celebrated for showing strength and bravery, And also loyalty when they claimed to fight “them” in service of defending “us”. Fighting was said to be good for societies, such as to help prepare for war. The net effect was that the norm against starting fights was not very effective at discouraging fights during the farming era, especially when many “us” and “them” were in close proximity.

Today, norms against starting fights are enforced far more strongly. Fights are much rarer, and when they do happen we try much harder to figure out who started them, and to more reliably punish starters. We have created much larger groups of “us” (e.g., nations), and use law to increase the resources we devote to enforcing norms against fighting, and the neutrality of many who spend those resources. Furthermore, we have and enforce stronger norms against retaliating overly strongly to apparent provocations that may have been accidental. We are less impressed by fighters, and prefer for people to use other ways to show off their strength and bravery. We see fighting as socially destructive, to be discouraged. And as fighting is rare, we infer undesired features about the few rare exceptions, such impulsiveness and a lack of empathy.

Now consider disagreement. I have done a lot of research on this topic and am pretty confident of the following claim (which I won’t defend here): People who are mainly trying to present accurate beliefs that are informative to observers, without giving much weight to other considerations (aside from minimizing thinking effort), do not foresee disagreements. That is, while A and B may often present differing opinions, A cannot publicly predict how a future opinion that B will present on X will differ on average from A’s current opinion on X. (Formally, A’s expectation of B’s future expectation nearly equals A’s current expectation.)

Of course today such foreseeing to disagree is quite commonplace. Which implies that in any such disagreement, one or both parties is not mainly trying to present accurate estimates. Which is a violation of our usual conversational norms for honesty. But it often isn’t easy to tell which party is not being fully honest. Especially as observers aren’t trying very hard very to tell, nor to report what they see honestly when they feel inclined to support “our” side in a disagreement with “them”. Furthermore, we are often quite impressed by disagreers who are smart, knowledgeable, passionate, and unyielding. And many say that disagreements are good for innovation, or for defending our ideologies against their rivals. All of which helps explain why disagreement is so common today.

But the analogy with the history of violent physical fights suggests that other equilibria may be possible. Imagine that disagreement were much less common, and that we could spend far more resources to investigate each one, using relatively neutral people. Imagine a norm of finding disagreement surprising and expecting the participants to act surprised and dig into it. Imagine that we saw ourselves much less as closely mixed groups of “us” and “them” regarding these topics, and that we preferred other ways for people to show off loyalty, smarts, knowledge, passion, and determination.

Imagine that we saw disagreement as socially destructive, to be discouraged. And imagine that the few people who still disagreed thereby revealed undesirable features such as impulsiveness and ignorance. If it is possible to imagine all these things, then it is possible to imagine a world which has far less foreseeable disagreement than our world, comparable to how we now have much less violence than did the ancient farming world.

When confronted with such an imaged future scenario, many people today claim to see it as stifling and repressive. They very much enjoy their freedom today to freely disagree with anyone at any time. But many ancients probably also greatly enjoyed the freedom to hit anyone they liked at anytime. Back then, it was probably the stronger better fighters, with the most fighting allies, who enjoyed this freedom most. Just like today it is probably the people who are best at arguing to make their opponents look stupid who enjoy our freedom to disagree today. Doesn’t mean this alternate world wouldn’t be better.

GD Star Rating
Tagged as: ,

Automation As Colonization Wave

Our automation data analysis found a few surprising results. We found that labor demand is inversely correlated with education. As if, when facing a labor shortage for a particular kind of worker, employers respond in part by lowering education requirements. And even though more automation directly lowers demand for a job, it seems that labor demand changes, relative to labor supply changes, becomes a smaller factor for jobs where automaton rises more.

But the most interesting surprise, I think, is that while, over the last twenty years, we’ve seen no noticeable change in the factors that predict which jobs get more automated, we have seen job features change to become more suitable to automation. On average jobs have moved by about a third of a standard deviation, relative to the distribution of job automation across jobs. This is actually quite a lot. Why do jobs change this way?

Consider the example of a wave of human colonization moving over a big land area. Instead of all the land becoming colonized more densely at same rate everywhere, what you instead see is new colonization happening much more near old colonization. In the U.S., dense concentrations started in the east and slowly spread to the west. There was little point in clearing land to grow stuff if there weren’t enough other folks nearby to which to sell your crops, and from which to buy supplies.

If you looked at any particular plot of land and asked what factors predict if it will be colonized soon, you might see those factors stay pretty constant over time. But many of those factors would depend on what other land nearby had been colonized recently. In a spatial colonization wave, there can be growth without much change in the underlying tech. Instead, the key dynamic can be that there are big time delays to allow an initial tech potential to become realized via spreading across a large landscape. A colonization wave can be growth without much tech change.

Now think about the space of job tasks as a similar sort of landscape. Two tasks are adjacent to other tasks when the same person tends to do both, when info or objects are passed from one to the other, when they take place close in place and time, and when their details gain from being coordinated. The ease of automating each task depends on how regular and standardized are its inputs, how easy it is to formalize the info on which key choices depend, how easy it is to evaluate and judge outputs, and how simple, stable, and mild are the physical environments in which this task is done.

When the tasks near a particular task get more automated, those tasks tend more to happen in a more controlled stable environment, the relevant info tends to be more formalized, and related info and objects get simpler, more standardized, and more reliably available. And this all tends to make it easier to automate such tasks. Much like how land is easier to colonize when nearby land is more colonized.

Among the job features that predict automation in our analysis, the strongest is: Pace Determined By Speed Of Equipment. This feature clearly fits my story here; it says you coordinate your task closely with a task done by a machine. Many others fit as well; here is more from our paper:

Pace Determined By Speed Of Equipment picks out jobs that coordinate closely with machinery, while Importance of Repeating Same Tasks picks out jobs with many similar and independent small tasks. Variety picks out an opposite case of dissimilar tasks. The job features Wear Common Safety Equipment and Indoors Environmentally Controlled pick out tasks done in calm stable environments, where machines function better, while Hearing Sensitivity picks out less suitable complex subtle environments. In jobs with frequent Letters and Memos, such memos tend to be short and standardized. Jobs with more Advancement are “results oriented”, with more clearly measurable results. Simple machines tend to be bad at Thinking Creatively, Innovation and Mathematics. Physical Proximity picks out jobs done close to humans, usually because of needed human interactions, which tend to be complex, and where active machines could risk hurting them.

We have long been experiencing a wave of automation passing across the space of job tasks. Some of this increase in automation has been due to falling computer tech costs, improving algorithms and tools, etc. But much of it may simply be the general potential of this tech being realized via a slow steady process with a long delay: the automation of tasks near other recently automated tasks, slowly spreading across the landscape of tasks.

GD Star Rating
Tagged as: , ,

Why Not RFID Tag Humans?

Today, across a wide range of contexts, we consistently have rules that say that if you have a thing out there in the world that can move around and do stuff, you need to give it a visible identifier so that folks near that thing can see that identifier, look it up in a registry, and find out who owns it. That identifier might be a visible tag or ID number, it might be an RFID that responds to radio signals with its ID, or it might be capable of more complex talk protocols. We have such rules for pets, cars, trucks, boats, planes, and most recently have added such rules for drones. Most phones and tablets and other devices that communicate electronically also have such identifiers. And few seem to object to more systematic collection of ID info, such as via tag readers.

The reasoning is simple and robust. When a thing gets lost, identifiers help us get it back to its owner. If a thing might bother or hurt someone around it, we want the owner to know that we can hold them responsible for such effects. Yes, there are costs to creating and maintaining IDs and registries (RFID tags today cost ~$0.15). Also, such IDs can empower those who are hostile to you and your things (including governments) to find them and you, and to hurt you both. But we have consistently seen these costs as worth the benefits, especially as device costs have fallen dramatically over the decades.

But when it comes to your personal body, public opinion seems to quite strongly opposed:

My 14 law&econ undergrads all agreed when I assigned this topic on their final exam today. People oppose requiring identifiers, and as face readers are now on the verge of making a new ID system, many want to legally ensure a right to wear masks to thwart it.

Yet the tradeoffs seem quite similar to me; it is just the scale of the stakes that rise. When we are talking about your body, as opposed to your car, pet, or drone, you can both do more to hurt others, and folks hostile to you might try to do more to you via knowing where you are. But if the ratio of these costs and benefits favor IDs in the other cases, I find it hard to see why that ratio would switch when we get to bodies.

Added 5Mar2020: The number you get from an RFID tag need not directly tell you the public name or location of the person behind it. You might instead need a subpoena to get that from the number.

GD Star Rating
Tagged as: ,

Automation: So Far, Business As Usual

Since at least 2013, many have claimed that we are entering a big automation revolution, and so should soon expect to see large trend-deviating increases in job automation levels, in related job losses, and in patterns of which jobs are more automated.

For example, in the October 15 Democratic debate between 12 U.S. presidential candidates, 6 of them addressed automation concerns introduced via this moderator’s statement:

According to a recent study, about a quarter of American jobs could be lost to automation in just the next ten years.

Most revolutions do not appear suddenly or fully-formed, but instead grow from precursor trends. Thus we might hope to test this claim of an automation revolution via a broad study of recent automation.

My coauthor Keller Scholl and I have just released such a study. We use data on 1505 expert reports regarding the degree of automation of 832 U.S. job types over the period 1999-2019, and similar reports on 153 other job features, to try to address these questions:

  1. Is automation predicted by two features suggested by basic theory: pay and employment?
  2. Do expert judgements on which particular jobs are vulnerable to future automation predict which jobs were how automated in the recent past?
  3. How well can we predict each job’s recent degree of automation from all available features?
  4. Have the predictors of job automation changed noticeably over the last two decades?
  5. On average, how much have levels of job automation changed in the last two decades?
  6. Do changes in job automation over the last two decades predict changes in pay or employment for those jobs?
  7. Do other features, when interacted with automation, predict changes in pay or employment?

Bottom line: we see no signs of an automation revolution. From our paper‘s conclusion:

We find that both wages and employment predict automation in the direction predicted by simple theory. We also find that expert judgements on which jobs are more vulnerable to future automation predict which jobs have been how automated recently. Controlling for such factors, education does not seem to predict automation.

However, aside perhaps from education, these factors no longer help predict automation when we add (interpolated extensions of) the top 25 O*NET variables, which together predict over half the variance in reported automation. The strongest O*NET predictor is Pace Determined By Speed Of Equipment and most predictors seem understandable in terms of traditional mechanical styles of job automation.

We see no significant change over our time period in the average reported automation levels, or in which factors best predict those levels. However, we can’t exclude the possibility of drifting standards in expert reports; if so, automation may have increased greatly during this period. The main change that we can see is that job factors have become significantly more suitable for automation, by enough to raise automation by roughly one third of a standard deviation.

Changes in pay and employment tend to predict each other, suggesting that labor market changes tend more to be demand instead of supply changes. These changes seem weaker when automation increases. Changes in job automation do not predict changes in pay or employment; the only significant term out of six suggests that employment increases with more automation. Falling labor demand correlates with rising job education levels.

None of these results seem to offer much support for claims that we are in the midst of a trend-deviating revolution in levels of job automation, related job losses, or in the factors that predict job automation. If such a revolution has begun, it has not yet noticeably influenced this sort of data, though continued tracking of such data may later reveal such a revolution. Our results also offer little support for claims that a trend-deviating increase in automation would be accompanied by large net declines in pay or employment. Instead, we estimate that more automation mainly predicts weaker demand, relative to supply, fluctuations in labor markets.

GD Star Rating
Tagged as: , , ,

Unending Winter Is Coming

Toward the end of the TV series Game of Thrones, a big long (multi-year) winter was coming, and while everyone should have been saving up for it, they were instead spending lots to fight wars. Because when others spend on war, that forces you to spend on war, and then suffer a terrible winter. The long term future of the universe may be much like this, except that future winter will never end! Let me explain.

The key universal resource is negentropy (and time), from which all others can be gained. For a very long time almost all life has run on the negentropy in sunshine landing on Earth, but almost all of that has been spent in the fierce competition to live. The things that do accumulate, such as innovations embodied in genomes, can’t really be spent to survive. However, as sunlight varies by day and season, life does sometimes save up resources during one part of a cycle, to spend in the other part of a cycle.

Humans have been growing much more rapidly than nature, but we also have had strong competition, and have also mostly only accumulated the resources that can’t directly be spent to win our competitions. We do tend to accumulate capital in peacetime, but every so often we have a big war that burns most of that up. It is mainly our remaining people and innovations that let us rebuild.

Over the long future, our descendants will gradually get better at gaining faster and cheaper access to more resources. Instead of drawing on just the sunlight coming to Earth, we’ll take all light from the Sun, and then we’ll take apart the Sun to make engines that we better control. And so on. Some of us may even gain long term views, that prioritize the very long run.

However, it seems likely that our descendants will be unable to coordinate on universal scales to prevent war and theft. If so, then every so often we will have a huge war, at which point we may burn up most of the resources that can be easily accessed on the timescale of that war. Between such wars, we’d work to increase the rate at which we could access resources during a war. And our need to watch out for possible war will force us to continually spend a non-trivial fraction of our accessible resources watching and staying prepared for war.

The big problem is: the accessible universe is finite, and so we will only ever be able to access a finite amount of negentropy. No matter how much we innovate. While so far we’ve mainly been drawing on a small steady flow of negentropy, eventually we will get better and faster access to the entire stock. The period when we use most of that stock is our universe’s one and only “summer”, after which we face an unending winter. This implies that when a total war shows up, we are at risk of burning up large fractions of all the resources that we can quickly access. So the larger a fraction of the universe’s negentropy that we can quickly access, the larger a fraction of all resources that we will ever have that we will burn up in each total war.

And even between the wars, we will need to watch out and stay prepared for war. If one uses negentropy to do stuff slowly and carefully, then the work that one can do with a given amount of negentropy is typically proportional to the inverse of the rate at which one does that work. This is true for computers, factories, pipes, drag, and much else. So ideally, the way to do the most with a fixed pot of negentropy is to do it all very slowly. And if the universe will last forever, that seems to put no bound on how much we can eventually do.

Alas, given random errors due to cosmic rays and other fluctuations, there is probably a minimum speed for doing the most with some negentropy. So the amount we can eventually do may be big, but it remains finite. However, that optimal pace is probably many orders of magnitude slower than our current speeds, letting our descendants do a lot.

The problem is, descendants who go maximally slow will make themselves very vulnerable to invasion and theft. For an analogy, imagine how severe our site security problems would be today if any one person could temporarily “grow” and become as powerful as a thousand people, but only after a one hour delay. Any one intruder to some site who grew while onsite this could wreck havoc and then be gone within an hour, before local security forces could grow to respond. Similarly when most future descendants run very slow, one who suddenly chose to run very fast might have a huge outside influence before the others could effectively respond.

So the bottom line is that if war and theft remain possible for our descendants, the rate at which they do things will be much faster than the much slower most efficient speed. In order to adequately watch out for and respond to attacks, they will have to run fast, and thus more quickly use up their available stocks of resources, such as stars. And when their stocks run out, the future will have run out for them. Like in a Game of Thrones scenario after a long winter war, they would then starve.

Now it is possible that there will be future resources that simply cannot be exploited quickly. Such as perhaps big black holes. In this case some of our descendants could last for a very long time slowly sipping on such supplies. But their activity levels at that point would be much lower than their rates before they used up all the other faster-access resources.

Okay, let’s put this all together into a picture of the long term future. Today we are growing fast, and getting better at accessing more kinds of resources faster. Eventually our growth in resource use will reach a peak. At that point we will use resources much faster than today, and also much faster than what would be the most efficient rate if we could all coordinate to prevent war and theft. Maybe a billion times faster or more. Fearing war, we will keep spending to watch and prepare for war, and then every once in a while we would burn up most accessible resources in a big war. After using up faster access resources, we then switch to lower activity levels using resources that we just can’t extract as fast, no matter how clever we are. Then we use up each one of those much faster than optimal, with activity levels falling after each source is used up.

That is, unless we can prevent war and theft, our long term future is an unending winter, wherein we use up most of our resources in early winter wars, and then slowly die and shrink and slow and war as the winter continues, on to infinity. And as a result do much less than we could have otherwise; perhaps a billion times less or more. (Thought still vastly more than we have done so far.) And this is all if we are lucky enough to avoid existential risk, which might destroy it all prematurely, leading instead to a fully-dead empty eternity.

Happy holidays.

GD Star Rating
Tagged as: , ,