Truth Markets

3 Architecture FAQ

Truth Claims Market FAQ

How is the TCM different from a commodities market?
In a commodities market there is an underlying physical good such as wheat or gold that the contract is based on. In a simplistic view, you are buying and selling the underlying commodity when you enter into the contract. With Truth Claims, the commodity in question is a notional Claim, not a physical good. The distinction isn’t terribly important. We can use the following terms true-sider, buyer, and long investor interchangeably to refer to someone who believes a Claim will be judged TRUE. Those who believe in the falsity of the Claim are false-siders, sellers, or short investors.

How is the TCM different than the Foresight Exchange (aka IdeoSphere) and the Hollywood Stock Exchange?
Both of these sites are only for “pretend money”. When people have to bet their own hard-earned cash they are notoriously more accurate in their evaluation of truth, and they ignore whatever biases they may have. Long-term, those who are good “truth pickers” will have more money to invest and thus have more influence on the market. In a pretend-money economy it is very easy to open a new account and reload your bankroll. With real money, those who are deluded or buy into propaganda pay with their wallets and eventually go broke. Anyone who has played online poker knows this all too well.

Isn’t this just the “ideas futures” concept done with real money?
The TCM is interested mainly in current truth, while ideas futures deal explicitly with the future. Robin Hanson, inventor of the ideas futures concept, points out that any Current Truth Claim can be cast as a Future Truth Claim by wording it appropriately, e.g. “Will it be determined before date X that claim A is TRUE?” The issue is one of practicality. Let’s say you want to trade on a claim such as “The government is doing a good job”, call it GDGJ for short. You could craft a time-series of Futures with different judgment dates, for instance, at the end of each year. Then you could ask at any time before Jan 1, 2009: what’s the price of “2008 GDGJ” contracts? But once the judgment date has passed, you are forced to switch to the 2009’s or later.

What’s the right frequency for judgment? Yearly, monthly, weekly, or some other interval? The more frequent the judging, the more overhead required on everyone’s part. The less frequent the judging, the less relevant the Claim is to the present, which defeats the purpose. By creating an explicit form of Truth Claim that deal with current truth, we obviate the problems associated with trying to squeeze claims about the present into a mechanism designed for the future. Essentially, Currents are the time integral of an infinite series of Futures, much the way continuous interest integrates discrete compounding dates.

How is the TCM different than other real-money prediction markets like TradeSports and LongBets.org?
The main differences are that existing prediction markets and bet exchanges for real money rely on one of two mechanisms to assure that judging is beyond reproach: (1) a future event whose outcome is incontrovertible, or (2) a small panel of human judges (usually just one person). The “wisdom of crowds” judgment process outlined here allows the TCM to obviate the need for either condition by letting (a statistical sample of) all the people who are affected by the judgment decide the outcome for themselves. More importantly though, it creates a future Judgment event for Currents, which serves to ground the Claims in reality. More about this can be found here.

How do I know that PRICE corresponds to TRUTH?
Ultimately it’s a collective agreement, similar to currency. When you accept a twenty dollar bill as payment, you are taking a leap of faith that you will be able to use it for twenty dollars worth of value later. We all have to agree that the bill is worth $20. As long as the market is sufficiently liquid, we can accept that a cost of $80 for a Truth Claim means it’s very likely true, or mostly true. Conversely, a cost of $1 might indicate that either a mathematical proof has falsified it, or the Claimant has recanted.

How do I “sell” or liquidate my position in a Claim, for instance if I want to lock in a profit or limit my losses before judgment happens?
You do this by hedging. For instance, if you hold 30 true-side contracts of GDGJ, you can simply acquire 30 false-side contracts on the open market. The TCM will allow you reconcile positions that cancel each other out for bookkeeping purposes.

Does every Truth Claims have to be entered into the TCM as a written piece of text?
No. The TCM can easily be built to work on document repositories such as web pages cached by Google, or by unique identifiers (e.g. URLs) that specify where the Truth Claim can be found. Technology exists for assuring that the Truth Claim document is not altered after trading has begun.

How do new Truth Claims get onto the market?
In theory, anyone can submit a Truth Claim to the market as an IPO. All new issues can be set at $50. Market forces will drive the price to a “fair” point quickly after trading begins.

Isn’t it possible that a Current never gets judged based on luck of the draw?
Yes. However, it’s not important that the Claim is actually judged, but rather that the market participants believe that judgment will happen. The more certain it is that judgment will happen “soon”, the less likely that bubbles and other price anomalies will occur. It is easy to program in the most effective value for “soon” by tweaking the parameters in the judgment process and seeing how the market reacts.

What happens when a putative Claimant denies making a Truth Claim?
In such cases, the Authorship Derivative Claim (ADC) value for the underlying Truth Claim suffers greatly. The lower the ADC value, the less of an impact that Truth Claim will have on the NAV of the Claimant Bond of the Claimant. Thus, if someone lists a Truth Claim and the Claimant denies having made it, then assuming the Claimant is telling the truth, there will be little to no impact on their reputation in the Claimant Bond Market.

The judgment process seems bizarre, what’s the rationale behind it?
By choosing the judgment date randomly for Currents, we reduce the corruption potential by making prolonged propaganda campaigns expensive and impractical.

By choosing judges randomly from a subset of total potential stakeholders, we further reduce the profitability of corruption by diluting the efforts of those who would seek more direct means of market manipulation, i.e. you can’t bribe ’em if you can’t find ’em. But equally importantly we make the system practical by dramatically reducing individual judgment duty and by making the judgment process as simple as a quick IM chat exchange.

What portion of the total outstanding claims is chosen for Judgment on any given day?
A natural choice would be 1 in 365 so that the “mean time to judgment” is fixed and tangible. Varying this rate though may be useful for fine-tuning the system dynamics or keeping individual judging burden within an acceptable range.

How long is the Judgment Time Window?
Empirical testing is required. One thought is to keep it as short and unpredictable as possible while still allowing a critical mass of judges to comprehend and consider the claim thoroughly. The longer the window, the greater the chance of corruption.

What portion of the total registered market participants are chosen to judge a given claim?
There’s no reason it couldn’t be 100%, but the higher the percentage, the bigger the individual judging burden becomes. Assuming a minimum threshold for diversity of opinion, it would be better to keep the number as low as possible, and allow each claim to be more carefully considered. Also, the higher the percentage, the more likely a tragedy of the commons, wherein individuals compete to the point of dysfunction over the amount of total influence they hold. In other words, if you know you only have a 10% chance of being chosen, it might not be worth skipping dinner and other healthy activities on the off chance you could be called to duty.

Why not limit the potential judge pool to only those who have a stake in the particular claim?
This increases bias and corruption potential dramatically, even if you don’t know who the other stakeholders are. It’s important to have diversity for overall accuracy (see Wisdom of Crowds). Plus you might run into a case of pathologically low “judgment liquidity”, say in the instance where there is only one outstanding contract: the one between you and the person who took the other side.

Is voting done purely democratically, or is it weighted somehow, e.g. by each judge’s NAV on the CBM?
Interesting question. Weighting could be beneficial for overall “truthiness”, but also could lead to destabilizing feedback loops. Empirical testing is required.

Does a judgment have to yield a TRUE or FALSE outcome? Why can’t it be UNDECIDED?
For Futures it doesn’t make much sense. But for Currents, it may be beneficial to add an UNDECIDED outcome, for instance if the judgment falls in a range between 40% – 60%, indicating that the market is not too sure of the truth. The exact thresholds could either be globally determined, or they could be a part of the Claim itself. Currents which are judged UNDECIDED would be put back into the market for trading based on future judgment.

Are Claimants a necessary part of a Truth Claim?
For Currents they are an integral part of the model; when your favorite politician or news media makes a claim, you want to hold them accountable. For Futures, it’s not as important to attach a Claimant, however the more data we have on each Claimant, the better the CBM will function.

What happens with the annuity swaps from the original proposal?
While they didn’t ultimately solve the truth/price grounding issue, they could still be useful for canceling out time value premiums. Time premiums would be inherent in both Futures and Currents. And since the point of TCMs is to tie price and truth as closely as possible, there’s good reason to want them canceled.

What about market manipulation?
Who cares? The only people hurt are those who allow themselves to be misinformed. The smart investor does research on a stock before buying. The smart “claim staker” becomes well informed before investing in a Truth Claim. Eventually the market reaches the “true price” as the misinformation gets traded out.

Isn’t market manipulation illegal, immoral, or at least fattening?!
There are those who feel that insider trading and other forms of market manipulation should not be regulated as they are in most existing markets. Proponents of this view say that information that flows within company walls is really no different than the so-called public information. Insiders buy and sell stock all the time without being accused of insider trading. Outsiders create sophisticated computer algorithms using an army of math whizzes to glean small nuggets of insight on whether to buy or sell. The mantra of the markets is that eventually the truth will out in the form stock price.

The counter argument says that company insiders directly affect the information that is traded on by their actions as employees of the company. An otherwise bad business deal might look really attractive to a manager if they can trade on information that other people won’t have until the damage is done.

The good news about the TCM is that there is no distinction between insider and outsider. We are all outsiders trying to get better insight. If you are a better researcher, or you are better at letting your biases and ignorance be tempered by objective evidence, then you should be rewarded in the TCM. If a corporate or political interest invests a large amount in a Truth Claim so that they may profit on fluctuations, they will be thwarted by market forces. Yes, there will be those who are caught in the manipulation crossfire and who will lose perhaps worse than the manipulators themselves. But they knew what they were investing in when placed the trade. They just chose to follow the market rather than research the underlying value of the Truth Claim.

Are there market makers for the TCM like in other markets?
No. Market makers are middlemen with an inherent conflict of interest. Traditionally market makers have served an important role in providing liquidity for thinly-traded issues and complex derivatives. The TCM will match buyers and sellers at exact prices using stochastic algorithms that are guaranteed to be fair. If a TC has low liquidity it is because not enough people care about it or understand it well enough. Given the goal of the TCM to illuminate truth, it can be seen as a good thing that liquidity corresponds with relevance and clarity. Not all claims are equally important.

Claimant Bond Market FAQ

How is the CBM different than StarMine?
StarMine does objective analysis of market forecasters based on track record over time. StarMine and the CDM both serve a similar purpose, however the CDM is more general and is applicable in cases where the jury is still out on the underlying claims. Also, StarMine is not a market itself but rather an analysis of underlying market data and forecasts.

What exactly is the formula for NAV?
This should be determined empirically. However, it is clear that it should probably include the value of all Truth Claims that the Claimant is listed on, weighted by the corresponding ADC values.

Authorship Derivative Claim Market FAQ

Won’t there be market equivalents of slander and character assassination campaigns if the truth of claim authorship is allowed to be called into question?
Very likely. It may desirable to put mechanisms in place that automatically de-list Truth Claims whose ADC value drops below a threshold. At the very least, the ADC value will be displayed as a part of the ticker quote for the underlying Truth Claim. Thus anyone who is researching the value of a Truth Claim will always know how firm the authorship credit is.

If the ADC value is low, how does that affect the underlying Truth Claim market price?
It doesn’t necessarily. Truth Claims are ultimately about the world, not about who makes them. However, for the purposes of doing research on a Truth Claim and deciding to make a trade on it, the ADC value could be a relevant piece of information. A low ADC value could easily dry up the market for the underlying Truth Claim, depending on the nature of the claim.

General FAQ

What is the incentive for Truth Markets to come about? How does the market itself become profitable?
There are several potential revenue streams that make sense, including the following:

  1. Fixed transaction fees per trade, independent of trade volume
  2. Monthly subscription fee for all traders
  3. Selling advertising (e.g. pay-per-click web ads)
  4. Premium content offerings

Nobody is going to invest real money in “truth”. How is this really going to work?
Billions are already being invested in electronic marketplaces on information futures for politics, weather, entertainment, current events, etc. The twist suggested here is to invest in the current truth value of a proposition as opposed to its future truth value. If you have any doubts about the demand, see the following:

The Economist

Wired Magazine

“The price of orange juice futures has even been shown to accurately predict the weather”
“Traders on the Hollywood Stock Exchange last year correctly picked 35 of the 40 Oscar nominees in the eight biggest categories, according to The New Yorker magazine.”

Trade Exchange Network (TEN)

“The Global Marketplace where Communities of People Invest in their Opinions”
“3 platforms, 1 pool of liquidity, 65,000+ members & 1000’s of innovative products to trade”

Consensus Point

“Whether you’re forecasting sales revenue, project completion dates, forthcoming industry trends or deciding which features to include in the next product, Consensus Point offers an innovative set of software products — Foresight Server and Forsight On Demand — to meet your unique strategic planning and risk management needs”

Prediction Market Summit

“TEN has approval from the CFTC to operate as an exempt Board of Trade.”

Founder of TEN on Google blog

I’ve heard of the book, and the ‘predicitive power’ of money and opinion has long been recognised as being a VERY powerful tool….
We KNEW 100% certain that Arnie would win the California recall, before the announced result….as our contract had virtually closed out at 99 hours before the polls closed.
We also knew something was happening with George Tenet at the CIA 3-days before he resigned….the money moved significantly.
Saddam was captured, but hours before we saw massive volumes trading….
The question will always be about the ‘money following money’…but in reality seldom does a market shift unless there is supporting FACT behind the trading.
Its a great theory which in practise we have seen time and time again: “When people state their opinion and back it up with hard cash, its a hard opinion to ignore!”

Can the Truth Markets concept be extended to values other than “Truth”?
Yes! As long as the market participants agree on the value system that they are upholding during Judgment events, this same concept can be applied to other values, e.g.:

  • Legality / Constitutionality
  • Entertainment Value
  • Newsworthiness
  • Overall Human Wellbeing
  • Corporate Governance
Advertisements

1 Comment »

  1. […] rationality to the mix in the face of media, financial incentives, political agendas.  Perhaps the truth market concept can be tweaked to create “urgency […]

    Pingback by Notes from TED « Complex Adaptive Systems — December 16, 2008 @ 8:43 pm


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: