Futarchy: Robin Hanson on prediction markets

https://richardhanania.substack.com/p/futarchy-robin-hanson-on-how-prediction

by jcarterwil • 5 years ago

150 193 5 years ago

richardhanania.substack.com

Prediction markets would influence the processes they're supposed to be impartially observing.

When there's a lot of money riding on something happening, it tends to happen. Sports and traditional financial markets have been dealing with this problem since forever.

iakh• 5 years ago

I hadn't heard that before. Can you give an example? Or do you mean in the "fixed game" way?

spaetzleesser• 5 years ago

Markets are not based on fixed principles but are basically a game of psychology mixed with a few principles. So expectations play a big role in market outcomes.

azta6521• 5 years ago

If I had 500M to manage, I may just buy a lot of real estate in a specific place to drive up the prices in that area. I want prices to increase and with money I can create circumstances that would favor my preferred outcome.

robertk• 5 years ago

I feel like this is left as an exercise to the reader and no proof is needed. Bettors in a prediction market are not divine speculators causally divorced from the real world. They are embedded, and when there are a lot of them and their financial incentives are towards a particular outcome, they might act in ways that aggregate to a greater likelihood of the event transpiring than in the counterfactual setting where they are pure observers of the simulations waging in a vacuum. It seems to me there are multiple ways to formalize and prove this and this contributes to my perception the original comment seems self-evident. If that is not the case then something is wrong in my intuition.

pjc50• 5 years ago

Years ago there was a libertarian proposing this as "assassination politics": simply create a system that allows people to bet that "politician X will be assassinated before time T". People who want X assassinated take the other side of that bet. Eventually there may be enough money in the pot that someone considers it worth making the hit and collecting the bet, and the people on one side have paid for it without directly paying for it.

PragmaticPulp3• 5 years ago

Imagine a prediction market had an entry for whether or not someone would streak across the field during a Super Bowl. Someone might see this, buy into the market, and then go streak across the field to force the outcome to favor their position.

It’s not even theoretical. This actually happened (or rather was attempted with traditional betting markets, not prediction markets): https://www.insider.com/super-bowl-streaker-bet-on-himself-p...

Prediction markets give financial incentive to force specific outcomes. They aren’t just observations: They become incentives to influence the outcome. The bigger the market, the bigger the incentive.

chaboud• 5 years ago

You can find many simple but observed real world examples of unintended consequences from indexing on observation by searching for examples of Goodhart's Law (which, as commonly generalized, should actually be called Strathern's Law):

"When a measure becomes a target, it ceases to be a good measure."

See Wikipedia to get started, then google search for examples: https://en.m.wikipedia.org/wiki/Goodhart%27s_law

inter_netuser• 5 years ago

Numerous scandals in european soccer. an endemic problem in any big popular sport, really.

https://en.wikipedia.org/wiki/French_football_bribery_scanda...

https://www.dw.com/en/police-expose-european-soccer-bribery-...

etc etc etc

colinmhayes• 5 years ago

There was a prediction market on the number of times celebrities tweet a week. Someone found one of the celebrities live streaming and kept paying them to delete their tweets.

naasking2• 5 years ago

The solution seems simple: predictions should be blind until the deadline. This also exploits the "wisdom of the crowd", where crowds are smarter when each individual's decision is independent of the others.

What am I missing?

miketery• 5 years ago

The price of a prediction depends on what people are currently predicting. So you can't keep it blind. Well I guess one way would be to put in limit orders on odds you're willing to take, and not revealing the market price. However I don't know if such a market condition / feature would have people using the prediction market.

dragonwriter• 5 years ago

What am I missing?

That by so preserving the independence of predictions, you’ve also removed their utility, which is predicated on their availability.

karpierz• 5 years ago

What do you mean by predictions are blind? The issue the parent is describing is (as an extreme case): I go on a prediction market for when someone will die, put all of my money on tomorrow, and then kill the person tomorrow.

You could imagine a lighter version where I ruh for some public office, bet a billion against me winning, and then drop out of the race.

PragmaticPulp3• 5 years ago

Prediction markets are literally markets.

You can’t have a market if the price is secret.

The goal isn’t to have people place bets and see who is right later. The goal is to expose the predictions to market forces and make people put their money on the line, thereby (theoretically) improving the quality of predictions.

nxpnsv• 5 years ago

Sounds clever, until you realize you want to predict stuff before it happens

__MatrixMan__3• 5 years ago

If you reduce transparency you also lose out on a lot of utility that centers around the wagered amount being an indicator of certainty.

Suppose you're using prediction markets to encourage thorough code reviews, specifically with an eye towards catching malicious commits. Run of the mill non-malicious PR's get lots of little yes wagers, and are merged without exceptional scrutiny from the package maintainers. Then a malicious commit comes along and a reviewer wagers $100 on "not merged". This captures the attention of the maintainers and they give the PR extra scrutiny. Turns out it has a malicious commit, so it doesn't get merged and the reviewer who found the flaw is rewarded with the money wagered by those who didn't find the flaw (plus some from the stakeholders, who seed the market with some "no" money to encourage participation in the game even in boring non-adversarial times).

If you hide the predictions and the amounts, you can't use the unsettled bets as inputs for decision making.

croon• 5 years ago

If you know the price and payout of a bet, it's no longer blind. And if you don't, you wouldn't buy a ticket.

aazaa• 5 years ago

This is why, for example, decentralized life insurance policies have been compared to hit contracts.

bchjam• 5 years ago

"An engine, not a camera" is a great book about this effect in financial markets.

https://mitpress.mit.edu/books/engine-not-camera

robocat• 5 years ago

Review: http://quixoticfinance.com/the-perennial-criticism-a-review-...

hogFeast3• 5 years ago

He has written a new book about HFT which covers some of the same ground.

hugh-avherald3• 5 years ago

That's where Hanson's idea of futarchy (as opposed to prediction markets) comes into play. You don't bet on an event A, you bet on A given B or on A given not B. So the prediction market wouldn't be for an asset paying $1 if there's a streaker, but for two assets: one which pays $1 if there's a streaker given there's 100 or more security guards and $1 if there's a streaker given there's fewer.

lvass• 5 years ago

So the potential streaker has one extra step, discovering the amount of guards, before placing the bet? How does this solve the problem instead of creating even more of a plutocracy? Gaming the system will always be possible, except for a smaller amount of people.

pdonis2• 5 years ago

I don't see how this helps any with the underlying problem. All it does is introduce an additional incentive for another party to manipulate the outcome in order to win money by betting: whoever runs the security guards.

hamburga5• 5 years ago

Yet we still use financial markets.

Isn't the idea that we put up regulations to minimize this, just like we have restrictions on insider trading?

hamburga5• 5 years ago

From the transcript:

  Richard: They say Bin Laden is just going to put all his money in the market and then attack?

  Robin: Well, that was crazy because these were relativity thin markets, and they have a lot of money at stake. Basically a fact that people don’t know about the markets is that many people criticize by saying, “Well, somebody will try to manipulate the markets by betting on one side not because they know better, but because they’re willing to lose money in order to distort the market price.”

  That is true. There are people willing to manipulate markets, but that actually makes the prices more accurate. For example in the fire the CEO market you say, “Well, the CEO wants to keep his job, so he will bet in these markets in order to make himself look like the price will be higher if he stays, and lower if he leaves.”

  Yes he would have an incentive to do that, but when other traders know that somebody will be trying to manipulate in the market they know to increase their trading and their efforts and that compensates, and actually on net makes the prices more accurate. That’s something we see in theory and we’ve seen in the lab, and we’ve seen in the field. These markets are robust to attempts to manipulate. In fact people who want to manipulate them make the prices more accurate.

jstanley• 5 years ago

That's actually a slightly different thing.

The quote from the transcript is talking about attempts to manipulate the market.

optimalsolver is talking about manipulating reality in order to win money in the market.

RandomLensman4• 5 years ago

Oddly enough, in financial markets people do not rely on self-correction for manipulation, but rather there is a tendency for it to be illegal. Market manipulation is in general not believed to make markets more accurate.

Any kind of the suggested self correction here relies on the other traders having sufficient funds/capital/risk taking capacity to overcome the manipulator and to coordinate their views to some extent. In the CEO example, how would other traders even now that the CEO was making the bet? What if some other informed party made that bet? In small, open, and rather serene settings that might work out, but in a real market with fast movements, market makers, etc.?

DINKDINK4• 5 years ago

Prediction markets would influence the processes they're supposed to be impartially observing.

Who said anything about impartiality (/cordially a strawman)?

Changing behaviors is a feature not a bug. Your health insurance writer (who's bought "No optimalsolver will not get sick") loses money if you do in fact get sick. Your fire insurance writer has an incentive to provide you free fire inspections because it reduces their payouts. A farmer plants a lucrative but fragile crop because a meteorologist can better price weather risks than they can.

Swapping exposure across space and time is a productive act.

dwohnitmok4• 5 years ago

Swapping exposure is only productive if you have many guardrails outside of the system that absolutely constrain what actions players inside the system are allowed to do. Otherwise, and this has been borne out it in reality, you end up with very distorted incentives.

This is covered today by what we call fraud (whether that be insurance fraud or market manipulation fraud), which tries to set bounds on what acceptable behavior is so that you can try to eliminate pathological edge cases. I don't see how this would be handled if everything at a top-level is handled through prediction markets.

bobbyd2323• 5 years ago

A lot of dynamic Econ models have some condition where my choice today depends on an expectation of the future. Like how much I choose to eat today depends in part on my best prediction of how much I’ll eat tomorrow. For aggregate decisions, getting better predictions should improve welfare.

__MatrixMan__3• 5 years ago

The alternative to putting your money into the prediction market to influence the outcome is to try to influence the outcome in some other way (e.g. a disinformation campaign).

At when if bad actors put their money into the prediction markets, they draw further attention to their false claim--something that can't be said for the alternatives.

ddp266• 5 years ago

This is true, but in practice a very small problem.

People who can control or majorly influence these big outcomes - elections, sports matches, acquisitions, etc. - generally have a much, much larger stake in the outcome itself than any side bets.

IMAYousaf• 5 years ago

It also creates adverse incentives for bad behavior. If I have a stake in a prediction, I am incentivized to make that happen by hook or by crook.

pphysch7• 5 years ago

If there are no limits on betting, then this is just plutarchy with extra steps. Guy comes along with $100B and directly changes the policy landscape.

If there are limits on betting, then it's direct democracy with extra steps (everyone gets one "vote" worth up to $limit).

What if improving society was a reward-in-itself and we trusted competent managers to do the complicated bits?

jacquesm3• 5 years ago

We are already in that situation, but we have several intermediaries (media ownership, lobbying) to pretend that we do not.

pphysch7• 5 years ago

There is no law that says you have to be a corrupt politician and enable plutarchy, although it is common.

At any rate, "futarchy" is clearly a step backwards.

RandomLensman4• 5 years ago

None of these things are are really new, but I wish folks like this would better understand what makes markets work:

Markets need participants, capital, liquidity, effort, etc. to become efficient. A lot prediction markets suffer from the fact that volatility in them is limited and that the outcome is not of interest to many people, so the markets cannot really do what markets are supposed to do.

If markets are aggregating crummy data and information, no useful market will form. And that is observable even in much more developed financial markets, where some things really have trouble on price formation. And then you might use auction methods, for example.

Similarly, if stuff happens rarely and you only get very limited shots at being right, then averaging and aggregation or not so useful because you cannot get an average (e.g., can do one thing for the next five years and better be right, for example).

pgustafs• 5 years ago

Adding to your point, most successful markets are positive-sum -- hedgers gain value from mitigating their structural risks, and speculators get paid to assume price risk. For example, the wheat futures market has two natural participants -- farmers and bakers. Farmers can sell future produce to buy seeds right now. Bakers can hedge their wheat price exposure to reduce their chance of getting ruined by a bad harvest. Speculators get paid to hold onto wheat futures contracts if a farmer wants to sell a future when the bakers aren't around to buy (presumably baking), selling to a baker later for a higher price reflecting the price risk assumed. All of these participants derive value from the market.

It's not clear to me that prediction markets usually have natural hedging participants (maybe political operatives, but the tx costs are probably too high relative to the value at stake).

wpietri3• 5 years ago

Excellent point.

Prediction markets have been a thing for at least 25 years. I get the intellectual appeal, and they may be useful in certain niches. But I think their lack of significant uptake or impact is telling.

nickff• 5 years ago

Prediction markets have huge positive externalities, as they help non-PM participants predict the future! One problem PMs have is that non-PM participants often feel that people betting short (or in some cases long) are 'hoping for failure'.

zerobits2• 5 years ago

It’s a bit presumptuous that these folks don’t understand what makes markets work.

They went into quite a bit of detail but can’t cover everything. Most of the examples were quite top of funnel, fire the CEO markets, economic impacts of new laws being passed. I wouldn’t expect for major bills or companies there would be any shortage of liquidity there.

It’s typical of comments in any forum to mostly be critical, but what takes more guts and cleverness is to connect the dots to improve upon the idea. You seem to have a good mind so I’d encourage you to try applying it in that way.

RandomLensman4• 5 years ago

Bit presumptuous to assume I have not been involved with the creation of markets, trading, etc. I have seen stuff work as well as fail up close in large arenas. [Edit: sorry, I should not snark. You are right that criticizing is easy creation is not.]

Going from academic ideas of markets and experiments to actual deep and useful markets is surprisingly difficult.

EDIT: the failure to create proper working markets for GDP-linked derivatives is good example of something that should but actually ain't. Not enough market making risk takers, limited hedges, unbalanced demand between long and short demand, index issues, ...

hogFeast3• 5 years ago

Even when you have a market with immense amounts of liquidity, the market won't be efficient if all the participants are wrong.

I remember back in 2012 InTrade had a market for the US Presidential election, the odds were wrong, stayed wrong for months, and actually got more wrong close to the event. You could get hourly liquidity in the tens of thousands almost throughout (I put on $20k in this market, I wasn't touching the sides, it was incredible).

Most of the people who talk a lot about prediction markets haven't worked in markets. Markets aren't magic. They work better with binary outcomes but they cannot be smarter than the people making bets in that market...and they aren't (I have most experience with financial markets, which just don't work well at all, but have quite a bit of experience with binary markets too...they have only become more accurate as our knowledge about the underlying events increased...if you look at binary markets where knowledge is limited in some way, markets are not efficient).

RandomLensman4• 5 years ago

Seconded, predictive quality of markets is not at all a given. For example, financial markets can have real arbitrage opportunities in size open that no-one removes (if capital requirements are high/volatile) - so no unique predictive state then

amvalo• 5 years ago

Wrong how? Skewed R?

ddp266• 5 years ago

| A lot prediction markets suffer from the fact that volatility in them is limited and that the outcome is not of interest to many people

+1. This is why I'm more interested these days in prediction markets within (large) institutions, like tech companies, universities, or government agencies.

That way, the participants have (a) a shared domain of interest, and (b) can speculate about internal, nonpublic matters.

Perhaps most importantly, if the prediction market improves the wisdom of the institution, the participants benefit - even if they aren't profitable in the market itself.

elil172• 5 years ago

Last year a man took a large bet that there would be a streaker at the Super Bowl, then went to the Super Bowl and streaked. His winnings far exceeded the fine and the cost of the ticket.

My fear about a betting market for policy is that you’d have a bunch of rich/powerful people betting GDP will go down and then doing everything in their power to stunt economic growth.

paulgb• 5 years ago

Ha, first I'd heard of that. Apparently it didn't work out for him, which I guess is good for market integrity.

https://www.insider.com/super-bowl-streaker-bet-on-himself-p...

flerchin• 5 years ago

Is that different than shorting a broad range of stocks?

elil172• 5 years ago

That’s a good question - perhaps it’s not different.

chollida13• 5 years ago

Yes in the sense that for a bet like streaking the chance that you can affect the outcome is 100%.

The chance that you can short a "broad range of stocks" and affect their share price is alot less certain. Especially in liquid markets your chance of affecting a blue chips stock price is very small.

There are always a few cases, like Archageos, where you can get so big that you move the stock but those cases are so well known exactly because they are so rare.

gunshai2• 5 years ago

This is the exact perversion I was thinking of. Pun very much intended.

tshaddox7• 5 years ago

That doesn’t strike me as a terrible outcome of a prediction market, unless you think the act of streaking is itself so terrible.

dwaltrip• 5 years ago

It isn't a problem as long as no one is betting on whether or not "bad" things will happen. However, the things that people care about (or find interesting) – and want to bet on – are often heavily debated as being good or bad. This means that naive prediction markets will probably incentive some people to work towards outcomes that many view as bad.

ajkjk• 5 years ago

To me it just sounds like a dystopia. Like: "market forces are so fun, let's have more of that!" It is wishing for a giant system that has a mind of its own and can't be reasoned about or controlled but affects everyone's life.

I suppose I already think that most of wall street is not, in fact, making anything much more efficient at all, and is instead a massive state-run casino. So maybe I am not the target audience.

DoubleDerper• 5 years ago

Prediction markets will exacerbate the "war on truth," and are illegal in the US, thankfully. "Shorting" is frequently accompanied by disinformation campaigns against a target. It's scary to think about what this will lead to.

tlholaday5• 5 years ago

"Shorting" is frequently accompanied by disinformation campaigns against a target. It's scary to think about what this will lead to.

Do you agree that it would lead to colossal losses by the disinformation promulgators and those they influenced when their predicted outcomes failed?

blurbleblurble2• 5 years ago

There's an episode of Mike Tyson's podcast where he talks to a former Mafia boss about how easy it is to get athletes who are in debt to Mafia bookies to throw games as payment for their debts.

Prediction markets bring the magic of perverse incentives to all walks of life.

tylermauthe• 5 years ago

Agreed, this post completely misses how the power dynamics of wealth affect the opinions and 'rationality' of players. It also never mentions how a whole section of society would be unable to take part in this system due to their socioeconomic status.

ddp266• 5 years ago

Are you opposed to futures markets for commodities, currencies, stock prices, etc. on the same grounds?

Sometimes I think opposition to prediction markets is just opposition to markets in general, which is fine, but a very broad critique.

cjcole2• 5 years ago

Hanson says regarding national metrics: 'Well, now we’re going to authorize this same sort of agency to estimate number like GDP except we’re going to tell them to put more things in the number. Bills before Congress would say, “Count more trees, and count leisure, and count international reputation.” They would just make a bigger formula that included all the stuff they cared about in this measure of national welfare.'

Here's the tricky part. It seems to me that the metrics which you don't choose to bundle into the aggregate measurement will get annihilated (optimized out) at the expense of those which you do choose to bundle. And the decisions as to which things you do bundle and with what weight represent ethical, moral, spiritual, aesthetic, and otherwise intangible judgments which are (unsurprisingly) difficult to quantify or even come to basic agreement on. He weasels a bit by using relatively uncontroversial examples ("more trees", "leisure") and by handwaving ("authorize": how?, "sort of agency": what sort?), but it's trivial to imagine metrics which produce wild disagreement regarding the magnitude or even sign of the weight to be applied.

(Shadows of paperclip maximization loom.)

At that point, I'm not sure where we've significantly improved things since we're still left with the problem of how to choose the people who decide which metrics go into the aggregate measure and at what weight. Can we make a market on that, too? At that point we're in some crazy recursion and I get lost. I'm deeply skeptical.

blfr• 5 years ago

His usual answer is "vote on values, bet on beliefs." So we would just vote on what to include, and then bet (through prediction markets) on how to achieve these goals.

pphysch7• 5 years ago

Who determines what is on the ballot, or how unstructured ballot items get classified? Are "make college free" and "make university free" the same or different policy proposals?

cjcole2• 5 years ago

That "just" is doing a lot of work.

It seems to me that many of these "rationalize all the things" schemes tend to bottom out at and build onto an "incorruptible kernel of truth" which, if corrupted, causes the whole thing to collapse (or worse: have a veneer of impartial truth while being secretly corrupt). It's a sort of microkernel approach to government.

With Hanson's futarchy, it's the hypothetical mechanism for "voting on values" in order to choose what does and doesn't go in the almighty bundle of metrics (and who gets to measure the result: they have the real power). If that can be corrupted then the whole thing falls apart. Yarvin's "neocameralism" has a "cryptographic decision and command chain" which everything else rides on (https://www.unqualified-reservations.org/2008/05/ol6-lost-th...).

This "all the eggs, one basket" approach seems fragile and ripe for subversion, especially when having fallible humans administer it. It seems more resilient to avoid concentrations of this kind of power.

SlapperKoala• 5 years ago

Hanson says regarding national metrics[... ]

The big issue I feel like he underrates is the massive incentive to take control of the agency that's supposed to be doing objective assessments. These are now effectively the most powerful people in the country. Even if they start off as saints the political incentive is going to be to find ways to influence it. And once that agency is captured then you have a dictatorship in all but name.

I feel like this reflects a common problem in political theorising, of coming up with an ideal institutional structure without thinking about the incentives around it and how it need to be sustained.

Historical analogues would be how originally non partisan district drawing processes are politicised, the politicisation of science and medicine, or soviet or Chinese gdp figures. The simplest solution is always to just rig the game.

chollida13• 5 years ago

Options and equites markets have a similar risk called pin risk where an underlying stocks price will tend to stick close to a strike price with alot of open interest.

Pinning refers to the potential for institutional option buyers to manipulate price action in the underlying as options expiration approaches. If these option buyers face the potential for a total loss of the option, they may try to pin the stock to a price just in the money by strategically entering buy orders at the last minute before the close

https://www.investopedia.com/terms/p/pinrisk.asp

You end up with this tail wagging the dog model where options end up moving the underlying as the option seller/writer doesn't know if they will be called on the option so they have to acquire it to hedge out their risk just in case.

Similarly the option buyer doesn't know if they will be assigned so they tend to short the stock to, again, hedge out their risk so the stock experiences both buy and sell pressure around the strike price due to bets placed on it.

gruez• 5 years ago

Similarly the option buyer doesn't know if they will be assigned

Don't you mean option seller/writer? If you buy an option (and therefore are an owner), you have the right but not the obligation to exercise it.

You end up with this tail wagging the dog model where options end up moving the underlying as the option seller/writer doesn't know if they will be called on the option so they have to acquire it to hedge out their risk just in case.

Right, but that's not any different than any other sort of leverage? eg. buying stocks on margin to drive up the price.

chollida13• 5 years ago

Don't you mean option seller/writer? If you buy an option (and therefore are an owner), you have the right but not the obligation to exercise it.

No, meant the buyer, alot of professionals want their exposure to be hedged out so they don't want to end up owning any shares after expiry so they need to be short shares for any calls that will end up in the money. So as an option pins the underlier close to the strike price the option holder ends up buy/selling to hedge out their delta exposure to zero.

Right, but that's not any different than any other sort of leverage? eg. buying stocks on margin to drive up the price.

Not really,

The tail waiving the dog effect occurs when a derivative instrument that is priced off an underlier actually ends up moving the underlier itself.

Leverage is a different animal. If you lever up 2x you're just buying twice as much. When you unwind you you sell 2x as much. Its still the underlying moving itself and not being affected by any other instrument.

glutamate4• 5 years ago

Am really I the only one who would rather have a Future Anarchy, where worker cooperatives take over the world?

tlholaday5• 5 years ago

where worker cooperatives take over the world?

Suppose that worker cooperatives have taken over the world.

Suppose then that the worker cooperatives must choose between policy options, for example { "increase payments to retired workers", "increase investment in disaster preparedness", "increase investment in infectious disease biochemistry" }.

What is your recommendation for how the worker cooperatives should decide how to allocate resources? Presuming you do not recommend prediction markets.

glutamate4• 5 years ago

One member, one vote. Worker cooperatives are democracies.

sudosysgen4• 5 years ago

There are many possible ways for the individual members to decide to vote, in the end it ends with one person one vote.

Otherwise the tools we have available are extremely powerful especially when fed good data that is freely available, which is what would be possible if you achieved the (perhaps impossible task) of switching most of the economy to worker coops.

avsteele3• 5 years ago

Commenters saying: "but someone can influence the market" should at least listen to the podcast (and really should also) familiarize themselves with the existing literature.

Do you think this critique has never been considered? If you disagree with the reasons why this is not a problem, go ahead and critique those directly.

Please don't reflexively pooh-pooh an idea you don't really understand.

PragmaticPulp3• 5 years ago

The link is a transcript of the podcast. Are we really missing anything by not listening to it?

Also, could you at least summarize some of the counter-arguments or even link to any of the literature?

avsteele3• 5 years ago

read==listen here.

Do a search for "manipulate" in the text for a simple counter-argument.

tellingly. None of the critiques I see address these counterarguments even, so I suspect knee-jerk responses are mostly whats posted here.

LanceH• 5 years ago

The simple reason you can't manipulate the market for profit is that everyone else can make the same bet you are making, quickly reducing the profit since you have the added expense of manipulating the market.

If a person can manipulate the market without expense, then they can do it without relying on the profit and the market isn't the problem. It will, however, allow them to bet on the market, providing a good indicator.

DuskStar• 5 years ago

Imagine that there's a market for which day a politician will die on. Initially, all of these daily contracts are priced very low because there is a very low chance of a politician dying on any given day. So, you buy quite a few of one particular day, and then go kill the politician.

SilasX• 5 years ago

Or, since you understand that literature so well, you could contribute to the discussion by explaining its key insights on those points.

BurningFrog• 5 years ago

I haven't seen Hanson address the "destructive insider trading" scenario where I buy a prediction that California air quality will be bad, and then spend the summer starting wildfires in California forests.

Today, people have no incentive to do that other than "wanting to watch the world burn". But if you can get rich burning forests, more forests will burn.

I don't doubt Hanson has thought about this. I just haven't seen it addressed.

tshaddox7• 5 years ago

My guess is that if a particular prediction were really so easy (low cost and risk) to manipulate then you wouldn't expect much trade volume. As an easier example, you wouldn't expect many people to bet money that no one will tweet a cartoon of a cat wearing a pink hat today, because it would be obvious to everyone that it would be trivial for an insider to manipulate this outcome.

As for your specific example, starting wildfires is an extremely serious crime that I expect the authorities would spend significant resources investigating, so I don't know how bettors would estimate the ease of manipulation for that market. Even if the prediction market were implemented in a perfectly anonymous system where you're guaranteed to be able to collect (and spend) your money without authorities knowing you were the one who profited from that prediction market, it seems like you have a pretty high risk of getting caught simply for the act of arson itself (not to mention the "risk" of firefighters controlling the fires that you start).

I'm also not sure how many people there are out there who refrain from starting wildfires only because there is no direct financial incentive, and not for other reasons (like not wanting to destroy massive amounts of ecosystem and property and potentially kill many people).

avsteele3• 5 years ago

I mean, someone can buy stocks short and then burn down a company's HQ/factory today.

This isn't a reason to not allow equities markets.

Imnimo8• 5 years ago

This might be a dumb question, since my understanding is limited to the article, but here goes.

Let's say we have the situation in the diagram, we have a two policy options, "A" and "B", and we're betting on "if we adopt this policy, will the GDP reach the target?". Let's say that currently A is unpopular, so "yes if A" has a low price, and B is popular. Now imagine an oracle enters the market, who knows that A actually has the best chance of reaching the target, but B also has a good shot. In fact, they know that "yes on A" and "yes on B" are both underpriced - just "yes on A" moreso. Ideally, we would like this oracle to put their money on "yes on A". But if they don't have enough capital to change which market leads, they'll break even (by having their money returned when B wins) if they do that. Instead, they should bet on B, which they know to be a worse option, because it'll actually fire and they'll still make some profit. The market doesn't get to learn all the information that the oracle has.

Is there some way to structure the markets so that our oracle instead bets on A?

qsort4• 5 years ago

Is there some way to structure the markets

Most of the prediction markets I've seen use sets of binary options that are complete an mutually exclusive. The entity that handles the market would only sell you for, say, $1, a complete set of those binary options.

So you could not buy "A", you could buy a pair made of a copy of "A" and a copy of "not A". If the oracle knows that "A" is true with p ~ 1, they would know that "A" is underpriced (it should be ~$1) and "not A" is overpriced (it should be ~$0), so they would buy ~infinitely many pairs of "A" and "not A".

Imnimo8• 5 years ago

I don't think that matches what they're describing in the article. The problem is that we only get to try implementing one of policy A and policy B, so if someone bets "I think policy A will achieve the goal", but we implement policy B, you have to just void their bet.

If we had already decided on policy A, and were just trying to predict whether it'll work, what you describe would be fine. But in the article, we're trying to decide whether to implement policy A or policy B, by having two separate markets, one for "what will happen if we do A" and another for "what will happen if we do B", and one of those two will get voided.

piaste2• 5 years ago

One improvement might be to make the total capital that investors have access to into public information.

That way the market should judge bets based not on their absolute value, but on the degree of risk that the bettor is willing to take on. An oracle betting 100% of their capital on A should be treated as a maximally strong signal regardless of the size of said capital. Now, if even that signal isn't enough to move investors away from B... well, you can't really stop them without turning the system into an aristocracy.

Of course, ensuring that the total capital information is accurate is going to be complex - how do you prevent rich people from creating 'proxy investors' who bet 100% of their borrowed funds? - but it seems on a similar order of magnitude of difficulty as 'prevent people from insider trading'.

Imnimo8• 5 years ago

Unless the two choices are extremely close, we probably don't want it to be the case that a single investor, even with 100% of their funds, can single-handedly change the outcome. We just want each investor to be incentivized to put their money behind what they believe to be the best policy, and hope that collectively they choose right.

Thus, we should expect that even if our oracle knows with certainty that A is the best policy, and invests all their money into it, they're unlikely to be able to unilaterally change which policy will actually be implemented. They will therefore be incentivized to bet on B succeeding (which is also a winning bet), and now that we're revealing that they bet 100% of their capital on it, all we've done is magnified the signal of that bet. But this is bad - our oracle is betting in a way that moves us away from the policy they know is best.

setori88• 5 years ago

If it's of any interest to anyone I recently did a podcast with Prof Robin Hanson too: https://www.youtube.com/watch?v=EIbOUjplFus [2hr 48min video]

lkey3• 5 years ago

If I recall correctly, this is the same man who said:

'I’m not a medical professional, so I can’t speak much to medical solutions.'

and in spite of this, proceeded to suggest we intentionally infect people at the outset of the coronavirus.

Claiming, falsely:

'As of yesterday total known deaths were 1384, a number that’s had a six day doubling time lately. At that rate, in four months deaths go up by a factor of a million, which is basically the whole planet. So unless growth rates slow by over a factor of four, there’s probably not time for a vaccine to save us.'

You'll have to forgive me if I don't take any of his claims about the future organization of society seriously, as he seems to lack something critical ever time he tries to imagine how people other than him might behave.

Also this:

'To me as a youth, I think the theory that many men want to have sex with men looked like a conspiracy theory: implausible, with no direct evidence shown, & the sort of thing people would want to claim even if not true. I now believe, though I've never seen very direct evidence.'

Truly a bizarre way to conceptualize the world.

tlholaday5• 5 years ago

Truly a bizarre way to conceptualize the world.

If familiarity is what you seek, look for echo chambers.

lkey3• 5 years ago

Pithy, but I can't think of anything more pointless than seeking out a solipsist and trying to convince them I exist.

Seeking contrary opinions for their own sake is not virtuous.

Seeking out "heterodox opinions" from people with an established track record of being misinformed on a topic is downright foolish.

Matticus_Rex2• 5 years ago

The numbers on variolation (the intentional infection he was talking about) still look like a pretty clear net positive. Were variolation trials not so legally problematic, they would likely have saved many, many lives, because as Hanson pointed out very early on the amount of exposure seems to be one of the larger factors in disease severity with COVID-19 (which was very predictable). And for most of his futarchy claims, he's relying directly on fairly good evidence. That's not a good reason to overhaul society in its image, but neither is it a good reason to dismiss the ideas.

I'm the first to agree that Hanson has pretty big blind spots, and he's about as non-neurotypical as they come (and therefore seems to often fall into a trap of terrible assumptions about how others will behave), but this is an awful example to use.

lkey3• 5 years ago

'Were variolation trials not so legally problematic' It's not like they weren't going to study this, but studying it in a way that doesn't violate the core tenants of medical ethics took longer than vaccine development. Indeed, I see that it has continued to be studied into the present year, but results are not unilaterally good, especially regarding secondary cases.

"they would likely have saved many, many lives"

This statement is loaded with unfounded assumptions. Governments and epidemiologists, if even if they'd agreed with Hanson, might have had a tough time convincing the population to go to their doctors to be infected with Covid-lite.

There are dozens of interventions at every point along the way that might have saved many many lives.

What I find most inexcusable is Hanson, a self-styled numbers guy, justifying his urgency with a completely unfounded extrapolation.

'At that rate, in four months deaths go up by a factor of a million, which is basically the whole planet'.

This is a completely, laughably unserious analysis.

As to the 'futarchy', I think other comments here point out the flaws in any such system better than I could.

AlbertCory7• 5 years ago

Old news with a catchy new name "futarchy".

In 2005 when I joined Google, they actually HAD prediction markets (no $$ involved) on business questions like "Gmail will have 50 million 7-day actives on April 1." It seemed so hip and with-it. Top management can get the real truth about things, rather than just asking their underlings!

The markets didn't take off, they withered. AFAIK they don't have them anymore. A friend of mine actually created an open source project for setting up your own prediction market. We can ask him how many users he got.

We've had betting markets on politics for a long time, too. I lost a few dollars on Amy Klobuchar in the 2020 primaries, although I had a 3x profit for a while (didn't sell, damn it).

So the interesting question isn't "is this a good idea?" but "why hasn't it taken over the world already?"

===== "Richard: Do you put futarchy in a larger intellectual tradition? Because a lot of people when they're coming up with an idea... Did you come up with this term by the way?

Robin: Yeah, and I've been ridiculed for it."

ALittleLight• 5 years ago

If there was no money involved it sounds like Google actually didn't have prediction markets. This is like saying that because nobody will take care of my lawn in exchange for Monopoly money lawn care businesses can't exist. It might be different if people got paid!

ddp266• 5 years ago

Agree, tangible incentives are important. But non-monetary prediction markets still have other benefits over traditional forecasting:

(1) Traders only make predictions in markets where they think they will gain;

(2) Traders control their risk, so one very confident person can make a larger bet than 10 people who are not very confident;

(3) Markets are continuously updated as new information arises;

(4) Profitable traders trade more, while losing traders exit the market

asdfasgasdgasdg2• 5 years ago

Prediction markets are probably not a panacea, but Google's various internal implementations of them (there have been several 20% attempts) are far from proof of this. If the people making the predictions don't have any actual skin in the game, the economic incentive to be correct is removed. That is the secret sauce that should supposedly make prediction markets more accurate than other forecasting venues.

AlbertCory7• 5 years ago

People DO bet where "ego points" are the reward, and my own personal proof is the Hollywood Stock Exchange [1]. For several years, I won my movie Meetup group's Oscar pool, just by taking all the predictions from HSX. (They don't do all the minor awards, which is a problem.)

Actually using real money is a major legal hurdle, and that's why most prediction markets don't do it. The political markets have a special "research" dispensation, last I checked.

[1] https://www.hsx.com/

ddp266• 5 years ago

You'd be surprised how motivated many Googlers are to be on a leaderboard of top traders, even without $!

arpinum2• 5 years ago

Robin has been at this for over 15 years. Some reasons why he failed in the early years were that he used government money for uncomfortable bets about assassinations of US politicians.

didibus2• 5 years ago

I just can't trust a social scientist personally telling me that something offers better predictions, because social science has lost my trust in it's methodology. I'd need to see a data scientist, mathematician or statistician arguing that it does for me to start considering it more.

The good thing though is, he could simply setup a decision market to have people decide the true value of decision markets no?

AlbertCory7• 5 years ago

Yeah, that's one of the few markets where I'm with the alarmists. You definitely can move the market by assassinating someone.

tshaddox7• 5 years ago

Futarchy also isn’t new. I remember it being a popular topic of discussion online 10 years ago, and apparently it dates back at least to 2007.

lend000• 5 years ago

The problem is mostly around the legal gray area of using them, and whether it constitutes gambling, instead of general buying and selling of assets.

If there wasn't a legal problem, I think they would have much more of an impact.

whimsicalism• 5 years ago

no $$ involved
The markets didn't take off, they withered.

This is surprising?

ddp266• 5 years ago

I'm not sure this is so clear. Google's first prediction market had 1,463 traders (>10% of the company at the time) and produced >250k predictions (though >50% of those were from trading bots).

The market was a surprisingly accurate forecaster too: http://static.googleusercontent.com/media/services.google.co...

wpietri3• 5 years ago

Another good example here is the Long Bets project, which I wrote the initial code for: https://longbets.org/

It has been an interesting exercise, and our largest bet was over $1m: https://longbets.org/362/

It has been a success as far as I'm concerned, in that it's still up and running almost 20 years later, and it has generated a lot of discussion about long-term topics, which is our goal.

But at the time I hoped it would turn into something much bigger. Lots of bets! Lots of money on table! Secondary bets! Maybe even options! That was definitely not the case. It has stayed niche, and so have prediction markets generally. I think there are good reasons for that.

AlbertCory7• 5 years ago

Checked it out. I think this is a great idea! Why do think it's stayed niche? There IS real money at stake, albeit philanthropic donations.

My personal feeling: people do not want accountability, and to paraphrase Jack Nicholson in A Few Good Men, they can't handle the truth. I think that's the real reason the Google prediction markets died: managers prefer to do what they want to do, regardless of whether it's going to work or not. Perhaps someone can theorize as to why that's a good thing? I can't.

ctchocula2• 5 years ago

The markets didn't take off, they withered. AFAIK they don't have them anymore.

Actually there is a new prediction market at Google now called Gleangen. It seems to be fairly active and flourishing despite not using real $$. Although there is a leaderboard, so perhaps the desire to want to rank in the top 10 is what motivates people to place bets.

AlbertCory7• 5 years ago

I'm not looking for any corporate secrets, but generally speaking:

* what sort of things can bets be placed on? Real sensitive business stuff, or more mundane things?

didibus2• 5 years ago

Has there actually been any empirical proof that betting markets yield better predictions though? From what I read, it doesn't seem so, seems that people still aren't sure if they offer better predictions or not than alternatives, and since it isn't really obvious if they do or don't, it seems ideological if you believe they do or don't and therefore support them or not.

Does anyone know if things have changed on that front?

max_• 5 years ago

For a relevant blog post i wrote[0].

[0] https://as1ndu.xyz/2020/06/the-wisdom-of-rationals/

DINKDINK4• 5 years ago

A more accurate phrase is "Event Derivative"

A derivative (a contract which is representative of some good, right, entitlement) based on the outcome of an event occurring or not.

Event derivatives are all around us. You buying fire insurance grants you the right to sell, to an insurance writer, a smoldering pile of wood (worth $2) at a price of say 30% of the unburned value of your home (similar to an OTM put option). An accidental-death insurance contract being worth a negligible amount until the subjects death.

What people often get wrong about EDs is

1. Whether or not they accurately predict events: they don't; they reflect the price, or odds, at which a market is willing to swap exposure. e.g. One pays a insurance costs because it's more profitable to swap risk than deal with exposure.

2. Confusion around incentives: Yes, each party swapping exposure will change their behavior -- which is a feature, not a bug. A company buying fire insurance can now enter into commerce otherwise prohibitive unlocked by swapping fire risk with an insurance writing company who has incentives to prevent a fire risk -- and do so at scale, coordinating multiple parties.

3. Lack of information about unfettered demand for the products. People claim there's no demand for EDs but neglect to take into account regulations: prevent people from purchasing them freely, political manipulation of prices when buyers are unhappy with the market price for risk, regulatory capture creating anticompetative producer protections, observe that because because of bans the inferior counts some EDs have been forced into are insufficient.

The global policy failure of covid has been around risk pricing, risk-exposure swapping, and effective, at-scale incentive coordination.

An exercise: An assassination market opens for the price of your head. In which case(s) should you be most concerned for your life:

[A] Exposure available at .99 on the 1

[B] Exposure available at .01 on the 1

[C] Exposure available at .5 on the 1

@HarryDCrane on Twitter is an applied researcher in this area, read him if you're interested in more -- I know I have.

lifeisstillgood2• 5 years ago

What does "Exposure available at .99 on the 1" mean please?

(I hate guessing in stats and odds - the jargon used to express things has so many subtleties)

DINKDINK4• 5 years ago

The available price is 99 cents

AlbertCory7• 5 years ago

And now for something completely different: prediction markets and patents!

I had the idea, back in 2004, of a prediction market for "patent NNN will be invalidated." The idea is that this creates a financial incentive to build the case, usually with prior art, to invalidate garbage software patents. Because most software patents ARE invalid.

I actually pursued this at Google for some period. The implementation details are kinda prohibitive, though:

must ALL the claims be invalidated, or just some?

can the patentee amend his claims to avoid the invalidity ruling?

by what date must the invalidation take place?

who's going to bear the cost of invalidating it?

Nice idea, though.

benlivengood• 5 years ago

An interesting case study is the Nenana Ice Classic[0]. There is existing research[1] on the accuracy of the betting pool as a prediction market, but to me it looks like outcomes farther from the median result are not predicted well.

[0] https://en.m.wikipedia.org/wiki/Nenana_Ice_Classic [1] https://climatecommunication.yale.edu/wp-content/uploads/201...

a9h74j• 5 years ago

Their step 6: reward those who won in the prediction.

And is there a predictable step 7: one dollar one vote.

What if "billionairs" already feel they they have won up to step 6, and feel legitimated in that step 7.

jonstewart• 5 years ago

The only confident bet I feel like I can make about the future is that most of my bets about the future would be wrong. I don’t think this is an uncommon feeling amongst humans, and perhaps indicates why prediction markets haven’t caught on beyond a core niche.

Futarchy: Robin Hanson on prediction markets

Related Stories