Put up or shut up performance-bond governance

[lirazsiri]

AKA “put your money where your mouth is"

Political systems suffer from the problem of demagogues because they reward form over substance.

The root of the problem is misaligned incentives in a system of governance that grants power to those who do not bear the burden of failure.

Instead political system obsess over the question over who should rule and not what constitutes winning and how to fairly reward those who take risks to win on our behalf.

Any coordination system that rewards self promotion over problem solving will get more self promotion and less problem solving. It will reward extroverted fakers who compete theatrically in the spotlight to win Twitter. It will penalize introverted achievers who are busy researching hard engineering problems.

When was the last time anyone saw an advertisement for an algo trading firm? Never.

An efficient coordination system naturally squeezes out the role of influence peddlers in all their forms (from marketers to sly politicians) because it has has no use for them. Resources are automatically allocated to the most productive problem solvers. There's no use in spending any resources on fancy packaging for good (or bad) ideas because it's the good ideas themselves that win, not the packaging.

To reward genuine achievement over fake achievement I propose a system of governance rooted in the idea of performance bonds.

Every election period all token holders vote not on who should govern but on what they consider a win and how much are they willing to pay for wins of various degrees. Instead of electing a person or organization to office they elect performance and compensation functions. For example this could be a decentralized Oracle feed that rewards governance in proportion to the average price of the token on secondary markets. Or by the number of transactions per second in all shards multiplied by the cost of gas, etc.

Signalling their preferences is where the role of token holders ends and the role of a private competitive policymaking industry takes over.

Policymakers compete to stake the highest bonds betting they can maximize the performance function and get rewarded the most for doing it. Essentially policymakers are competing to insure the community against their own incompetence. To the degree that they have proven themselves their cost of capital should go down which makes the reputation of the policymaker an out of band function that influences governance through the capital markets.

In the simplest version of this game the highest bid wins the contract. To the degree that they are successful the policy maker either makes a profit or loses their stake.

The use of performance bonds can be fractal so that the primary role of the policymaker is to define nested performance / compensation functions for different sub-roles (research, devops, core development, events coordination).

Ideally these would be mostly on chain as soon as the governance stack and supporting ecosystem is mature enough to support defining objective performance / compensation functions for a broad scope of required activities. However this is mostly a cost savings mechanism that could cheapen the cost of global coordination within a well aligned model of compensation. It could still work even if most of the coordination was happening off chain and still somewhat vulnerable to subjective measures of performance, as is the default today.

It is expected that performance bonds governance would almost quickly outperform existing systems of governance with misaligned incentives that encourage and reward political behavior, defined as a system that rewards form over substance, possibly without even realizing that there is a difference.

However it is expected that like all marketplaces we will only see the true potential of a private policymaking industry emerge after the market has matured, weaker policymakers have been weeded out and only the strong policymakers prosper because they got became wealthy actually coming up with good policies or at least recognizing and promoting good policies when they saw them.

Like with any economic game it will take multiple iterations for the best and brightest policymakers and policies to float to the top.

[kronosapiens]

I think the overall research direction by which we govern via functions and vote on parameters is a good one. My question here is whether the "objective" measures of success you suggest will be as effective as you hope. If I put up a bond to increase the price of a token, am I not incentivized to simply manipulate price, rather than create underlying value? If I put up a bond to increase network throughput, am I not incentivized to create fake traffic, at a cost of up to $1 less than the value I expect to gain?

[lirazsiri]

You could voice a similar objection to getting medical care for a brain tumor from a neuro-surgeon vs getting it from a witch doctor. If the neuro-surgeon cuts your head in the wrong place it could be worse the witch doctor dancing around you and doing nothing at all.

So yes, there is a problem of constraining performance / compensation functions for the performance bonds to those that will generate outcomes faithful to the community's intentions. And of course if we don't do that well then we won't get what we want and it may be worse than doing nothing. But not doing that at all opens governance to the misaligned incentives of form over substance. We see the result in today's decaying political systems most of which are eventually captured by demagogues who pander rather than doctor the body politic.

To clarify, I was not actually proposing that the performance / compensation function be the token price or the total spent on gas throughout all shards. Those were just quick and dirty examples, off the top of my head. Coming up with objective quantitative proxy measures that faithfully represent the community values, in all their qualitative nuance should be expected to be hard. But we'll never get there if search for or keys under the streetlight trying to answer the seemingly easier question of how to divide the loot.

In my mind this means we should be spending more time thinking long and hard about measurements, not less.

As a metaphor, pre-enlightnment thought was based on intuitive conviction as the measure of truth. Science, rooted in objectivity, was a harsher mistress that did not yield satisfying answers as readily. Being an alchemist was more romantic and intuitively appealing than being a chemist. It has taken us hundreds of years of objective scientific experimentation to ratchet up true knowledge of the laws of nature. If we ever want governance to develop being the stage of political alchemy we will have to think long and hard about what is it we want to achieve and how to measure it.

This idea that you can somehow magically increase quality by not measuring it reminds me of an old story in AI:

In the days when Sussman was a novice, Minsky once came to him as he sat hacking at the PDP-6. “What are you doing?”, asked Minsky. “I am training a randomly wired neural net to play Tic-Tac-Toe” Sussman replied. “Why is the net wired randomly?”, asked Minsky. “I do not want it to have any preconceptions of how to play”, Sussman said. Minsky then shut his eyes. “Why do you close your eyes?”, Sussman asked his teacher. “So that the room will be empty.”

At that moment, Sussman was enlightened.