Betting Billions On Predicting the Future (Gnosis Part I)
The untold story of GnosisDAO's bold, if unsuccessful, experiment with futarchy, prediction markets and institutional innovation.
Hey Friends!
Today’s post is the first of two that I’m planning to write about Gnosis.
Gnosis is one of the most interesting DAOs in all of crypto today. Their products — Gnosis Chain, Omen, Gnosis Safe, COWSwap and Gnosis Zodiac — help power the Ethereum ecosystem. And the story of how they built them is deeply fascinating.
But we’re not talking about that today.
Instead we’re going to talk about Gnosis’s original ambition to remake the way markets and politics work.
We’re going to talk about how the crypto community became obsessed with a new form of government known as futarchy.
We’re going to talk about how that system, if successful, could help solve some of the biggest problems with democracy.
And we’re going to talk about how Gnosis launched a bold (if unsuccessful) experiment to become the first organization governed by futarchy.
Let’s dive in.
On Foundation and Futarchy
When I was in high school, my father introduced me to the novel Foundation by Isaac Asimov. In the book, the scientist Hari Seldon develops a mathematical approach to predict the rise and fall of civilization.
This idea: that we can use mathematics to model and predict human behavior — was intoxicating. What if we could use data to render the ambiguities of history and politics completely clear? It changed how I viewed math forever. I was hooked.
In College, I studied math, economics and game theory. I relished building and manipulating intricate models of human behavior. But then came graduation. In the real world, I learned, this type of prognostication is usually used for tracking assets markets. Somehow, that lacked the romance of predicting the outcome of civilization shaping changes.
Incentives, it turns out, really are destiny. There’s money to be made in market speculation so we learn to deploy all our best analytical tools in that realm. For our cherished alpha, we traffic in information. We build and refine models. But we still apply almost none of this rigor to questions of civilizational import.
Last year, America relied on two agencies to estimate the budget impact of President Biden's $2 Billion Build Back Better Plan. Our estimate for its impact on key policy objectives -- climate, education, poverty-- was even more threadbare. Even in matters of war, we fly blind. Thousands of finance employees estimate the impact of Russia's war on oil prices. Far less are estimating the likelihood of nuclear war.
But it doesn't have to be this way. Imagine if investors could capture the upside from predicting policy outcomes. Imagine if we could deploy the energy and resources of modern finance against our most pressing national challenges.
That, in short, is the pitch for futarchy.
Futarchy was first proposed by George Mason Professor Robin Hanson in 2001.
Hanson believed that democracy and free markets are both incredible systems. But we can better leverage them if we bring them closer together.
Democracy helps a community set its goals and priorities. But democracy is not always the most effective way to find successful policies. Emotions crowd judgment. Information overload can distract and confuse voters.
But financial markets excel at finding the informational needle in the haystack. Given enough time, a market will accurately reflect the consensus wisdom about a firm or strategy. To quote Warren Buffet, in the short run the market is a voting machine, but in the long run, it’s a weighing machine. The market excels at weighing options to find the right one.
So instead of constraining each to its proper sphere, Hanson proposes we marry them. Voters choose policy goals. The market allows speculators to bet on how effective a given policy will be in achieving those goals. When the speculator bets correctly, they get paid out. This creates strong incentives to predict outcomes correctly.
When Prediction Markets Meet the Blockchain
Now, if you've made it this far, you may be asking yourself: WTF does this have to do with DAOs?
I promise, we're getting there.
You see, betting markets are caught between a regulatory rock and a hard place. Depending on how you look at it, they’re either an unregulated security or they’re gambling. Either way, they are persona non grata in the eyes of the SEC and the CFTC. The original betting markets were all forced to shut down in the US.
Today, there are two popular legal prediction markets in the US. Both operate under significant restrictions. Neither can allow bets over $840. Hardly the requisite financial horsepower for an efficient information-aggregating capital market.
So here's where we were in 2015.
1. Betting markets provide a libertarian, market based solution to a massive problem.
2. They were being blocked by perceived regulatory overreach.
Yes, the table was set perfectly for a blockchain solution.
You see, one of the core virtues of a blockchain is censorship resistance. Once a betting market is deployed on Ethereum, it cannot be taken down as long as a single Ethereum node is running. Even if the government shuts down the sponsoring company.
For the first time, there could be a permissionless, unregulated, international prediction market.
A few major players sprung up to test this idea. Among them, Gnosis and Augur attracted the most attention.
Gnosis’s journey, in particular, is super fascinating. On the way to building a prediction market, they happened to launch a series of products that are essential to Web3. Among them — Gnosis Safe, Gnosis Zodiac, Gnosis Chain and COWSwap. But their innovation story is for another time.
For now, we’re going to fast forward five years to November 2020. That's the month that Gnosis announced the formation of a $1B organization shepherded by prediction markets. Ladies and gentlemen: GnosisDAO.
The GnosisDAO Experiment
Over their five years of operation, Gnosis had acquired 150,000 Ethereum (valued today at $440M). They also retained 8M of their own Gnosis tokens (with an implied valuation today of $2.5B).
In November 2020, Gnosis announced they were turning over management of that treasury to Gnosis token holders. The DAO’s mission would be to grow the Gnosis ecosystem by wisely investing its treasury.
To achieve that, the DAO would make decisions on investments based on a three step process. Any member could suggest an investment. If popular enough, a prediction market would evaluate the impact of that investment. Finally, guided by that market, the community would vote on whether to adopt the proposal.
This moment marked the realization of Gnosis's shared vision and the culmination of five years of effort. If anyone could pull this off, it was Gnosis.
And yet, just eight weeks later, they voted to abolish the prediction market entirely.
The Flaw in the Design
This section is going to get a little technical. If you want to skip ahead to the next section, I won't be offended (or, you know, notice...):
Here’s what happened:
Gnosis's prediction markets allowed users to bet on whether a policy would pass (Yes Or No). Users could choose to place their bet in either USD or GNO (Gnosis’ equity).
If the policy were not expected to have any impact on GNO's value, these two markets should be interchangeable.
If I am willing to bet 30 cents for a $1 payout, I should also be willing to bet .3 GNO for a 1 GNO payout. After all, my upside is 70% in both cases.
But, what if I believe that the policy is valuable to Gnosis and will increase the price of GNO? Then I should be willing to pay more to bet in the Gnosis market. Once the proposal passes, my GNO will appreciate and I will earn more than 70% on my bet.
Rational investors would price this in. Therefore, the difference in prices between the USD and GNO markets reveals the expected impact of the investment.
Unfortunately, this design has two big problems.
First, Gnosis had to play the role of the “house,” in this betting market. But this was expensive. Gnosis estimated it would cost them $100,000 to set up a given market to get reasonable data.
Second, the prediction markets and the community's policy votes were public. Votes were often overwhelmingly decisive in GnosisDAO (>95% YES votes were common). Thus, once a policy looked sure to pass, speculators could rush in and buy “YES” tokens for a guaranteed payout. This both distorted the market and ensured that Gnosis, as teh house, would took on a loss.
Neither of these problems are unsolvable.
But they do require some significant redesigns. So Gnosis pulled the plug.
Lessons from Democracy & Civic Society
Democracy (mostly) works in the United States because we have a strong(ish) civic society. We have a political tradition grounded in social contract theory. We learn about voting at a young age. We teach our democratic values early and often.
Our familiarity with democracy obfuscates that democracy has a steep learning curve. We’re two hundred years into the American experiment and we’re still working out some pretty gnarly kinks.
So it’s safe to say that an infant system like futarchy will have growing pains.
But institutional innovation is one of the most exciting areas in crypto. If we can find a way to make better decisions for society, the entire crypto enterprise will be well worth all the scammy NFTs.
So, no, I'm not going to criticize Gnosis for their experiment. We should celebrate their willingness to boldly push the limits on institutional design. But their experience does provide some cautionary notes.
The best forum for early experiments may not be in groups with billion dollar treasuries. We need to make testing our new forms of government much cheaper. I think we can do that with futarchy.
My group chats love to make bets — on sports, on stock prices, on political outcomes. If there were a way for us to experiment with using that type of betting to make collective bets or investments, we would use it. And using it would give the group some intuition for leveraging prediction markets. That’s the only way to develop a citizenry with the necessary skills to operate a futarchy.
GnosisDAO’s flirtation with prognostication was not an end. It’s like Athenian Democracy — an idea way ahead of its time with a lot of big kinks to work out. But if the rate of innovation in web3 is any guide, the time of futarchy is going to come way sooner than any of us think.
Get Involved
New this week, I’m going to start listing out opportunities to get involved with each organization I write about.
Or check out some open roles with the team:
Business and Ecosystem Development (m/w/d)
Remote, Berlin, Gibraltar