Tag Archives: Prediction Markets

Coming Commitment Conflicts

If competition, variation, and selection long continues, our worlds will become dominated by artificial creatures who take a long view of their future, and who see themselves as directly and abstractly valuing having more distant descendants. Is there anything more we robustly predict about them?

Our evolving descendants will form packages wherein each part of the package promotes reproduction of other package parts. So a big question is: how will they choose their packages? While some package choices will become very entrenched, like the different organs in our bodies, other choices may be freer to change at the last minute, like political coalitions in democracies. How will our descendants choose such coalition partners?

One obvious strategy is to make deals with coalition partners to promote each other’s long term reproduction. Some degree of commitment is probably optimal, and many technologies of commitment will likely be available. But note: it is probably possible to over-commit, by committing too wide a range of choices over too long a time period with too many partners, and to under-commit, committing too few choices over too short a time period with too few partners. Changed situations call for changed coalitions. Thus our descendants will have to think carefully about how strongly and long to commit on what with whom.

But is it even possible to enforce deals to promote the reproduction of a package? Sure, the amount of long-term reproduction of a set of features or a package subset seems a clearly measurable outcome, but how could such a team neutrally decide which actions best promote that overall package? Wouldn’t the detailed analyses that each package part offers on such a topic tend to be biased to favor those parts? If so, how could they find a neutral analyses to rely on?

My work on futarchy lets me say: this is a solvable problem. Because we know that futarchy would solve this. A coalition could neutrally but expertly decide what actions would promote their overall reproduction by choosing a specific ex-post-numeric-measure of their overall reproduction, and then creating decision markets to advise on each particular decision where concrete identifiable options can be found.

There may be other ways to do this, and some ways may even be better than decision markets. But it clearly is possible for future coalitions to neutrally and expertly decide what shared actions would promote their overall reproduction. So as long as they can make such actions visible to something like decisions markets, coalitions can reliably promote their joint reproduction.

Thus we can foresee an important future activity: forming and reforming reproduction coalitions.

GD Star Rating
Tagged as: ,

Protecting Hypocritical Idealism

I’m told that soldiers act a lot more confident and brave when they are far from battle, relative to when it looms immediate in front of them.

When presented with descriptions of how most citizens of Nazi Germany didn’t resist or oppose the regime much, most people claim they would have done different. Which of course is pretty unlikely for most of them. But there’s an obvious explanation of this “social desirability bias”. Their subconscious expects a larger positive payoff from presenting an admirable view of themselves to associates, relative to the smaller negative payoff from making themselves more likely to actually do what they said, should they actually find themselves in a Nazi regime.

When the covid pandemic first appeared, elites and experts voiced their long-standing position that masks and travel restrictions were not effective in a pandemic. Which let them express their pro-inclusive global-citizen liberal attitudes. Their subconscious foresaw only a small chance that they’d actually face a real and big pandemic. And if that ever happened, they could and did lower the cost of this previous attitude by just suddenly and without explanation changing their minds.

For many decades it has been an article of faith among a large fraction of these same sort of experts and elites that advanced aliens must be peaceful egalitarian eco-friendly non-expansionist powers, who would if they saw us scold and lecture us about our wars, nukes, capitalism, expansion, and eco-damage. Like our descendants are presented to be in Star Trek or the Culture novels.

Because in this scenario aliens would be the highest status creatures around, and it is important to these humans that the highest in status agree with their politics. I confidently predict that their attitudes would quickly change if they were actually confronted with unknown but very real alien powers nearby.

This predictable hypocrisy could be exposed if people would back these beliefs with bets. But of course they don’t. They aren’t exactly sure why, but most just feel “uncomfortable” with that. Visible and open betting market odds that disagreed with them would also expose this hypocrisy, but most such also oppose allowing those, mostly also for vague “uncomfortable” reasons. Their unconscious knows better what are those reasons, but knows also not to tell.

GD Star Rating
Tagged as: , , ,

Who Wants Curated Democracy?

Most democratic systems are pretty simple. To a first approximation, anyone can run for office, any adult citizen can vote, and voters can use all their usual ways to associate and talk to evaluate and coordinate on who to vote for in upcoming elections.

Imagine that some academics instead develop and advocate for a “curated” system of democracy. They research how democratic outcomes vary with the candidates, who votes on which candidates, and who talks to who about which candidates and topics. These academics say that if you put someone who knows this literature well in charge of “curating” democracy, you can get better outcomes.

Assume that these researchers have the usual level of academic competence at doing their research. They study a real phenomena and make real progress, but have the usual academic biases, such as playing usual games of hindering rivals via insider clubs and method fashions. Their results tend to be complex, though news media can sometimes offer deceptively simple summaries of them.

How eager are you to replace your simple democratic system with a voting system curated by an expert credentialed by these academics? That is, to put these curators in charge of who can run for office, who can vote on what, and who can talk to who how about what political topics? They wouldn’t suggest simple rules that we could then debate and choose whether to adopt. No, they’d make many detailed context-dependent choices that they couldn’t well explain to us; we’d just have to trust them.

Most of us wouldn’t trust them, and thus would be wary of such curated democracy. Because democracy is less about having a well-oiled machine and more about having a simple neutral system that we can trust when we don’t together trust any particular people that much to run our system.

This is how I feel about the forecasting systems and contests that are now popular among academics, relative to simple prediction markets. In a simple prediction market, you set up a topic on which to bet cash, and then let any individual or group bet cash there at any time, in any amount, and the current price is your best estimate. Biases are to be fixed by traders profiting from finding and correcting them. Yes, each market has some mechanical details, but those matter less when there is lots of trading, and it usually works okay to let people compete to pick details of the markets they pay to create.

In contrast, in curated prediction contests, the curators pick who participates on which questions, assign them to teams in which they work together, assign them each a weight in a final consensus function that they choose, say how and in what units each is rewarded as a function of their predictions and outcomes, adjust their consensus for various “biases” they see. Curators say that in their studies that this approach gives more accurate predictions.

Which may well be true. Except they don’t do the crucial test where a lot is at stake in the decisions that the markets influence, so much that interested parties try to corrupt the curators themselves. By bribing curators, threatening to get them fired, or just taking over the whole process by which they are trained and selected. The more details that curators control, and the harder to understand their reasons for making adjustments, the more room there’d be for curator corruption.

Institution/system/mechanism design is a very different problem between when (a) you can trust someone to run it, and make discretionary adjustments as needed, and (b) there is no one we can agree to trust, so we need to agree on something simple and clear that will run with few such adjustments. I’m most interested in that second kind of design problem.

GD Star Rating
Tagged as: ,

“If We Win, You Win”

One of the big wins of capitalism is that it creates strong private incentives for some kinds of social change. If someone has an idea for change, they can get investors and employees to work with them, in the hope of rewards if the change earns profits.

Alas, the most fundamental problem with social change in our world is that capitalism doesn’t encourage many other kinds of changes. Yes, under democracy elected politicians can get some weaker rewards for proposing changes. But for anything but small local changes, it isn’t worth a politican’s time to work out change details, explain it to voters, and organize support. That sort of thing is left to social movements and organized interest groups.

While many will deny it, the main promise that movements make to potential recruits is this: “If we win, you win”. Thus we mainly see movements around changes that can credibly make such promises. For example, crypto promises to reward investors with more money, and workers with valued job skills. Academic and technical movements promoting particular tools promise rewards to those who invest in these tools, relative to those who invest in competing tools.

Sometimes a movement has a vague label, and the real message is “As we ‘own’ this label, if our movement grows then we can send rewards to the high status loyalists among us.” Sometimes the movement’s implicit message is simply “We need to replace old folks with young folks like us in positions of influence.”

In entertainment and fashion movements, the reward can just be looking and sounding more knowledgable and “with-it”. For example, if I watch a lot of Game of Thrones, and it is popular, then in conversations I can relate to and say more about what others discuss. If locally sourced foods get popular, then I can seem more with-it when I cook such foods or recommend their restaurants. And if I grow or sell local food, I can gain even more. If I do or don’t wear masks, and then my mask side wins, I can brag that I supported the winning side.

The key point is that there are a lot of good ideas for change, including ideas that most people will admit are good ideas upon examination, where it is hard to organize supporting movements this way. For example, you can make a movement around a new way to teach kids, as you might start a school or be a teacher that uses it, or you might have your kid taught with it. But it is much harder to make a make a movement around the idea that there should just be a lot less school, unless you push a particular alternative to school.

Colleges rate professor teaching via student evaluations, which seems to have zero correlation with how much students learn, even though learning is the main reason given to attend college. But it seems hard to start a moment to fix this. We probably could construct ways to evaluate teacher effectiveness at student learning, but that would take resources away from other things, and would interfere with letting teachers teach any way they like. And a movement to just stop using current evaluations would admit to the public that we don’t care much about teaching quality.

More generally, when the public will mainly listen to people who specialize in X regarding changes in X, it is hard to make a movement to cut back on X. You can have movements to increase investments in X, or change how X is done, but the people who gain from cutting X are not the people listened to much on X.

Note that early on, movements can just promise gains via personal association with prestigious founders. It is later on when movements need to offer other rewards.

Futarchy would solve this, as it could give much stronger rewards for initiating changes. (At least for problems that government can solve.) But what would be gained by those who joined a movement to promote futarchy? The mechanism is simple, so there’s little to gain from investing in learning how to use it. It doesn’t promise to promise the young over the old, or to promote any particular policies for which we could identify the winners. Just making the world, or your nation better, inspires little passion.

GD Star Rating
Tagged as: , ,

Shoulda-Listened Futures

Over the decades I have written many times on how prediction markets might help the intellectual world. But usually my pitch has been to those who want to get a better actionable info out of intellectuals, or to help the world to make better intellectual progress in the long run. Problem is, such customers seem pretty scarce. So in this post I want to outline an idea that is a bit closer to a business proposal, in that I can better identify concrete customers who might pay for it.

For every successful intellectual there are (at least) hundreds of failures. People who started out along a path, but then were not sufficiently rewarded or encouraged, and so then either quit or persisted in relative obscurity. And a great many of these (maybe even a majority) think that the world done them wrong, that their intellectual contributions were underrated. And no doubt many of them are right. Such malcontents are my intended customers.

These “world shoulda listened to me” customers might pay to have some of their works evaluated by posterity. For example, for every $1 saved now that gains a 3% real rate of return, $19 in real assets are available in a century to pay historians for evaluations. At a 6% rate of return (or 3% for 2 centuries), that’s $339. Furthermore, if future historians needed only to randomly evaluate 1% of the works assigned them, then if malcontents paid $10 per work to be maybe evaluated, historians could spend $20K (or $339K) per work they evaluate. Considering all the added knowledge and tools to which future historians may have access, that seems enough to do a substantial evaluation, especially if they evaluate several related works at the same time.

Given a substantial chance (1% will do) that a work might be evaluated by historians in a century or two, we could then create (conditional) prediction markets now estimating those future evaluations. So a customer might pay their $20 now, and get an immediate prediction market estimate of that future evaluation for their work. That $20 might pay $10 for the (chance of a) future evaluation and another $10 to establish and subsidize a prediction market over the coming centuries until resolution.

Finally, if customers thought market estimate regarding their works looked too low, then they could of course try to bet to raise those estimates. Skeptics would no doubt lie waiting to bet against them, and on average this tendency of authors to bet to support their works would probably subsidize these markets, and so lower the fees that the system needs to charge.

Of course even with big budgets for evaluations, if we want future historians to make reliable enough formal estimates that we can bet on in advance, then we will need to give them a well-defined-enough task to accomplish. And we need to define this task in a way that discourages future historians from expressing their gratitude to all these people who funded their work by giving them all an A+.

I suggest we have future historians estimate each work’s ideal attention: how much attention each particular work should have been given during some time period. So we should pick some measure of attention, a measure that we can calculate for works when they are submitted, and track over time. This measure should weigh if the dissertation was approved, the paper was published and where, how many cites did it get, etc. If we add up all the initial attention for submitted works, then we can assign historians the task of (counterfactually) reallocating this total attention across all the submitted works. So to give more attention to some, they’d have to take away attention from others.

Okay, so now they can’t give every work an A+. (And we ensure that bet assets have bounded values.) But our job isn’t done. We also need to give them a principle to follow when allocating attention among all these prior works. What objective would they be trying to accomplish via this reallocation of attention?

I suggest that the objective just be intellectual progress, toward the world having access to more accurate and useful beliefs. A set of works should have gotten more attention if in that case the world would have been more likely to have more quickly come to appreciate valuable truths. And this task is probably easier if we ask future historians to use their future values in this task, instead of asking them to try to judge according to our values today.

These evaluation tasks probably get easier if historians randomly pick related sets of works to evaluate together, instead of independently picking each work to evaluate. And this system can probably offer scaled fees, wherein the chance that your work gets evaluated rises linearly with the price you paid for that chance. There are probably a lot more details to work out, but I expect I’ve already said enough for most people to decide roughly how much they like this idea.

Once there were many works in this system, and many prediction markets estimating their shoulda-been attention, then we could look to see if market speculators see any overall biases in today’s intellectual worlds. That is, topics, methods, disciplines, genders, etc. to which speculators estimate that the world today is giving too little attention. That could be pretty dramatic and damning evidence of bias, by someone, evidence to which we’d all be wise to attend.

One obvious test of this approach would be to assign historians today the task of reallocating attention among papers published a century or two ago. Perhaps assign multiple independent groups, and see how correlated are their evaluations, and how that correlation varies across topic areas. Perhaps repeating in a decade or two, to see how much evaluations drift over time.

Showing these correlations to potential customers might convince them that there’s a good enough chance that such a system will later correctly vindicate their neglected contributions. And these tests may show good scopes to use, for related works and time periods to evaluate together, and how narrow or broad should be the expertise of the evaluators.

This whole shoulda-listened-futures approach could or course also be applied to many other kinds of works, not just intellectual works. You’d just have to establish your standards for how future historians are to allocate shoulda attention, and trust them to actually follow those standards. Doing tests on works from centuries ago here could also help to show if this is a viable approach for these kinds of works.

Added 7am 28Apr: On average more assets will be available to pay for future evaluations if the fees paid are invested in risky assets. So instead of promising a particular percentage chance of evaluation, it may make more sense to specify how fees will be invested, set the (real) amount to be spent on each evaluation, and then promise that the chance of evaluation for each work will be set by the investment return relative to the initial fee paid. Yes that induces more evaluations in state of the world where investments do better, but customers are already accepting a big chance that their work will never be directly evaluated.

GD Star Rating
Tagged as: , ,

Try-Two Contest Board

Imagine that a restaurant wants to ask its associates (cooks, servers, etc.) what are the best two menu items to put on its menu as specials on a particular night. They have a large set of possible menu items to consider, the measure of success is menu item sales revenue, and they want a mechanism that is both fun and easy. (Which rules out conditional prediction markets, at least for now.)

Here’s an idea. Start with a contest board like this, on a wall near associates:

Continue reading "Try-Two Contest Board" »

GD Star Rating
Tagged as: ,

Response to Suri Re Futarchy

If by chance one of your writings strikes a chord, and is cited by folks decades later, your main reward may be to repeatedly hear the same misunderstandings and off-target counter-arguments that you’ve repeatedly tried to head off in your writings, but which critics apparently can’t be bothered to read. Sometimes, though not usually, I bother to respond. Case in point: Sunil Suri’s complaints about futarchy in Politics With Skin In The Game.

His summary of futarchy mechanics seems fine to me, though it might mislead readers into thinking that one needs to pick a new outcome for each new policy choice. I instead suggest picking just one standard outcome measure to use for most all big choices. I’d only pick specialized measures for decisions too small to sufficiently impact the standard measure.

Suri admits to some positives:  

futarchy creates financial incentives to be a better-informed citizen. This could transform our politics by:

  • Reducing our consumption of low-quality information and our susceptibility to cognitive biases – both of which distract us from what matters.
  • Making real expertise matter again – while democratising it. …

Suri then lists ten objections. But five of those objections merely point to general features of the problem that futarchy is trying to address, which are thus issues that bedevil any solution to its problem.

To review, the problem is how to make key government policy choices, the sorts of choices now made when bills are passed by a legislature, or when executives issue orders. These choices are typically made in a complex world under great uncertainty regarding relevant outcomes, outcomes which are often spread out over many decades. A great many values and preferences are relevant for these choices. These values, and the relevant info needed to make these decisions well, are all housed within opaque, distracted, and often irrational humans, who must somehow be induced to sufficiently reveal them.

Here are Suri’s five applies-to-all-solutions objections:  Continue reading "Response to Suri Re Futarchy" »

GD Star Rating
Tagged as: ,

Lost Advanced Civilizations

Did life on Earth start on Earth, or did it start on Mars and move to Earth? If you frame such panspermia as an “extraordinary claim” for which you demand “extraordinary evidence”, you will of course conclude that this should be treated “skeptically” as unlikely and sloppy unscientific “speculation”. To be disdained and not treated as serious by respectable academics and science journalists. But that’s not really fair.

You see the early Mars environment is, a priori, about as likely a place for life to start as the Earth environment. So if the rate at which life is transferred between the planets were high enough, then equal chances of life starting first in both places would result in equal chances for Earth life to have started in either place. We should take the expected time difference between life starting in the two places, and ask how high is the chance that life would move from one planet to the next during that period. The more often rocks are thrown from one place to the other, and the more easily life could survive for the travel period within those rocks, then the more likely it is that Earth life started on Mars.

In addition, Mars, being further from the Sun, would have cooled first, and had a head start in its window for life. Making it more likely that life would start there and spread to Earth than vice versa. Of course life starting first on Mars would have implications for what we might see when we look at Mars. If we had expected Mars life to continue strong until today, then the fact that we see no life on Mars now would be a big strike against this hypothesis. But if we expected Mars life to have died out or at least gone dormant by now, then the issue is what we will see when we dig on Mars. With enough data on such digs, we may come to reject to Mars first hypothesis even given its initial plausibility.

A similar analysis applies to panspermia from other stars. You might think it obvious that the rate at which life-filled rocks from a star make it to seed other stars is very low, but most stars are born in large groups close together in stellar nurseries. So if life arose early enough within our star’s nursery, there might have been high rates of moving that life between stars in that nursery. In which case the chance that Earth life came from another star could also be high, and the best place to look for life outside our star would be the other stars from our stellar nursery.

Now consider the possibility of lost advanced civilizations. Not just civilizations at a similar level of development to those around them in space and time; that’s quite likely given that we keep finding new previously-unknown settlements and developed places. No, the more interesting claims are about substantial (but not crazy extreme) decreases in the peak or median level of civilizations across wide areas. Such as what happened late in the late Mediterranean Bronze Age, or at the fall of the Roman Empire. Could there have been “higher” civilizations before the “first” ones that we now know about in each region, such as the Sumerians, Egyptians, and Chinese Shang dynasty? (I’m talking human civs, not others.) Continue reading "Lost Advanced Civilizations" »

GD Star Rating
Tagged as: , , , ,

Seeking Robust Credible Expertise Buyers

On Jan 19, 2000, I posted an email to the Extropians mailing list, giving the first public mention of the futarchy idea. (I also have a detailed PPT on the idea dated June 22, 2000, and the first pdf paper I posted is dated “July 2000.”) So the general idea is just over two decades old now.

Coincidentally, some new prediction platforms have been announced recently, and some have asked me why I do not act more excited about them. So this seems a good time to review my agenda.

I seek to jumpstart stable decision-advising info markets, wherein bias-robust widely-credible expertise is bought and sold. Let’s walk through these terms one at a time.

By jumpstart stable I mean that I’m seeking to start a new regular practice, not just proof-of-concept demonstrations of related technologies. I’m okay with some party subsidizing them at first, to help move to a new equilibria. But that sponsoring party either needs to stay indefinitely, or the market must soon find a way to pay its way without that subsidy. To become a regular practice, relevant parties need to see a long enough track record of how such info markets have worked and performed in their particular topic areas.

By decision-advising info I mean that my goal isn’t to add to or change general talk, gossip, and chatter, much of which is too vague to see what exactly it means, and most of which influences little outside the world of chatter. My goal is instead to influence real and important decisions, via better info. So I want to see info markets that sell clear, precise consensus estimates that can be understood in probability terms, so they can be fed into traditional decision analysis.

To better influence decisions, these estimates should also be as actionable as possible. That is, estimates should sit clearly close to actual decisions, so that decision-makers can see their relevance, and see how different estimates naturally lead to different decisions.

By bought and sold, I mean that we need two kinds of participants, buyers and sellers. While there will sometimes be an overlap, in general the people who know things, the info sellers, just aren’t the same as the people who want to know things, the info buyers. And we can’t presume that the sellers will sell info for free. Instead, buyers must offer sufficient rewards to distract sellers from alternate activities.

By markets I mean to integrate these new systems with our many other markets in our mostly market economy. This isn’t a world apart. Most individuals and organizations in our society should be free to participate, if they so choose, as either buyers or sellers of info. And we should expect money to be the usual currency used to make deals.

By expertise, I mean that estimates should be accurate, due to embodying more information. While we must accept that there will be error, i.e., differences between estimates and truth, but on average errors should be minimized. More precisely, for each topic on which the markets offer an estimate, I want that estimate to be as accurate as possible given the costs paid for it. And it should usually be possible to pay more to get more accuracy.

By credible, I mean that estimates need to not just be accurate, but also to seem accurate to key audiences. And by widely I mean credible not just to a few audiences, but to many audiences. There should be a widely held common belief in their accuracy. For the set of topics to which they are said to apply, and holding constant the cost spend, these estimates need not usually seem more accurate than other key sources, but they should rarely seem to be much less accurate.

So I’m not just trying to create a tool that some people will see as useful, if they have certain compatible abilities and attitudes, and after they’ve practiced with it and developed a personal style of usage. Not just a private advisor who might happen to be trusted by a particular decision maker. I’m instead looking for an institution that many people with different goals and agendas can share, and trust together. That is, I seek the most accurate institution that many can share, even if some Individuals think they know of better sources.

For example, the accuracy of estimates shouldn’t depend greatly on the quality of management by key central administrators. Unless most everyone can agree on a reliable way to achieve high management quality, it just isn’t enough to have some people believe in a high quality of current management, if many others are skeptical. If any parts of these markets require central management, we need ways to pick managers that which don’t require unusual and unshared confidence in particular administrators.

The key attraction of widely credible info markets is that they can be used by decision makers who seek not just to make good decisions, but also to convince key audiences that they have made good decisions. And this can help us all to more easily trust agents who make decisions on our behalf. By checking that decisions made match the estimates from related info markets, we can check on decision makers. Or if market estimates can be make directly relevant and actionable enough, we might must put them directly in charge of key decisions.

By robust, I indicate that I want estimate accuracy to be high not just sometimes, but across a wide range of topics and information contexts. And by bias-robust I mean that I want estimates that are robust to situations where many parties would like to bias and distort the estimates, consciously or unconsciously, to influence decision makers. It is no good having something that works well in the lab, or on small unimportant topics, but falls apart when the stakes get high. To be a shared institution on important topics for parties with differing goals and agendas, we need a wide perception that accuracy persists even when many parties seek to distort and manipulate the estimates.

Okay, now that I’ve explained what I want, I can better explain when I get excited.

In the last few decades, dozens of groups have written new software to support info markets of varying forms. Such software is almost always tied to a particular project, and when that project fails the software almost never becomes available for other projects. And most of these groups see software and management as the only project parts worth paying for, in cash, stock, etc. Other parts are left as an exercise for to-be-determined “users”. So I find it hard to get excited about software unless it is tied to an exciting further project. Even software that comes with new features.

Sometimes sponsors are found to help pay to collect a set of regular users (i.e., info sellers) who talk on a set of regular topics. Sometimes it is the users themselves who are the sponsors, willing to pay in time and money to express their opinions on topics of interest to them. But rarely do such projects put much effort into soliciting participation and support from particular info buyers, choosing topics close to their key decisions. And, alas, the rare projects that at least pitch to potential info buyers tend to pick system designs sensitive to management quality, and less clearly robust to manipulation efforts.

Yet to my mind it is the info buyers who should come first in info market project planning. Info sellers are second, and software last. First find a set of estimates that would be useful in advising some set of important decisions. Especially where there’s a plausible trust advantage from widely-credible estimates, so that key audiences can better trust decision makers. Find parties to whom more credible accuracy would be valuable, and ask them how much they’d be willing to pay for it. They don’t need to be convinced of such accuracy in the start, but they do need to be willing to pay once sufficient accuracy is demonstrated. If you can’t find info buyers, you can’t make info markets.

Yes, when many potential info buyers want similar info, they can each be tempted to free ride on the efforts of others. So it makes sense to look more to cases where info gains are concentrated in a few parties. Alas, an even larger obstacle to finding info buyers is that we often justify our activities in terms of info collection and processing, when those activities are better described as local politics. We pretend to want accurate info far more often than we actually do.

I’m quite willing to work with most any group that seems to have at least a chance of putting together all the needed parts. But my best guess for the most promising project is still the one I first posted on over 24 years ago: fire the CEO markets re the Fortune 500. I doubt I have another 24 years, so I do hope someone tries this before then. For this project the plausible info buyers are firm investors, represented by the board of directors, who subsidize these markets. Likely info sellers are stock analysts and stock traders, who would profit from trading in these markets.

Simple money-based conditionally-called-off stock markets should produce bias-robust widely-credible estimates, at least if trading liquidity is high enough. That has been a widely shared belief on speculative financial markets for many decades. To get high liquidity, use large market-maker based subsidies on only a few firms to start with, firms chosen via a prize system as most likely to see fire-CEO recommendations. Once these prices get enough attention, especially from CEOs trying to manipulate them to make themselves look good, their liquidity can be self-reinforcing, and subsidies can be transferred to the next set of firms.

Yes, this fire-the-CEO project faces substantial legal obstacles if anyone is allowed to participate; may have to do this one offshore. Legal issues are much less of a problem for most projects that ask firm employees and contractors to advise firm decisions, as the firm can pay for their initial stake. For those projects the main obstacle is political disruption; existing players in the firm tend to be bothered to see their advice contradicted by a system with higher proven accuracy.

Of course I can get excited by a great many other project concepts; I’ve posted on many here over the years. But to get excited about an info market concept, I need to at least hear about the intended info buyers willing to pay to get bias-robust widely-credible expertise. A mere project to develop software, or even to collect a regular set of users, not so much.

GD Star Rating
Tagged as:

Pandemic Futarchy Design

Researchers … use … feeds from a global network of students, staff and alumni to construct a “stringency index” that boils down to a single number how strictly governments in 160 countries are locking down their economies and societies to contain the spread of the virus. … plans to include state-by-state measures of stringency in the U.S. … latest version … draws on 17 indicators to determine the stringency of the government response. (More)

Not that I hear anyone eagerly clamoring to try, but let me sketch out how one would use decision markets to set pandemic policy. Just to plant another flag on how widely they could be usefully applied, if only enough folks cared about effective policy.

As you may recall, a decision market is a speculative (i.e., betting) market on a key outcome of a decision, conditional on which discrete decision is made. To apply these to the current pandemic, we need to pick

  1. key ex-post-measurable outcome(s) of interest,
  2. likely-enough key decisions which could substantially influence those outcomes,
  3. participants capable of learning a bit about how decisions related to outcomes,
  4. sponsors who care enough about informing these decisions, and
  5. legal jurisdictions that may allow such markets.

Regarding participants, sponsors, and permission, it makes sense to be opportunistic. Seek any sponsors interested in relevant questions, any participants you can get to trade on them, and any jurisdiction that let you want to do. Alas I have no sponsor leads.

For key decisions, we could consider using bills before legislatures, administrative rulings, or election results. But there are a great many of these, we don’t get much warnings about many, and most have little overall impact. So I’d prefer to aggregate decisions, and summarize policy via three key choice metrics per region:
Lockdown Strictness. As described in the quote above, some have created metrics on lockdown strictness across jurisdictions. Such metrics could be supplemented by cell-phone based data on trips outside the home.
Testing Volume. The number of tests per unit time, perhaps separated into the main test types, and perhaps also into accuracy classes.
Tracing Volume. The number of full-time equivalent tracers working to trace who infected whom. Perhaps supplemented by the % of local folks use apps that report their travels to tracing authorities.

Yes, worse pandemic outcomes will likely cause more lockdown, tests, and tracing. But one could look at outcomes that happen after decisions. Such as how average future outcomes depend on the decisions made this month or quarter.

For key outcomes, the obvious options are deaths and economic growth.

For deaths, we can avoid testing problems by looking at total deaths, or equivalently “excess” deaths relative to prior years. It helps to note the ages of deaths, which can be combined with local mortality tables to estimate life-years lost. Even better, if possible, note the co-morbidities of those who died, to better estimate life-years lost. And even more better, have estimates of the relative quality of those life-years.

For economic growth, just take standard measures of regional income or GDP, and integrate them many years into the future, using an appropriate discount factor. Assuming that the temporary disruption from a pandemic is over within say 10 years, one could end the bets after say ten years, projecting the last few years of regional income out into the indefinite future.

As usual, there will be a tradeoff here re how far to go in accounting for these many complexities. I’d be happy to just see measures of life years lost related to lockdown strictness, perhaps broken into three discrete categories of strictness. But I’d of course be even happier to include economic growth as an outcome, and tests and tracing as decisions. Either aggregate all outcomes into one overall measure (using values of life years), or have different markets estimate different outcomes. For decisions, either separate markets for each type of decision. Or, ideally, combinatorial markets looking at all possible combinations of outcomes, decisions, and regions.

GD Star Rating
Tagged as: ,