Thursday, January 29, 2015

Random Democracy

The Problem
There are many versions of direct democracy. The version proposed in this blog should have its own name. 

The Solution
True to democracy, I am open to proposals but to avoid confusion I will refer to the version of direct democracy supported by this blog as Random Democracy.

What is Random Democracy?
Random Democracy is rule by a random sample of the people (and thus by the people).

How does Random Democracy work?

What are the advantages of Random Democracy?
Random Democracy copies the advantages of traditional direct democracy:
  • Direct Democracy ensures that decisions are made by the biggest possible group which reduces the possibility of elite capture of government.
  • Direct Democracy encourages everybody to take part in debate which not only ensures people are more aware about important issues, diversity will ensure that solutions are better and more creative as proven by crowdsourced solution.
  • Direct Democracy allows everybody to propose issues that are important to them so the debate is about things that matter to large parts of the population, not only to specific interests.
In addition, it solves the biggest problems of direct democracy:
  • Random Democracy is cheap and efficient for the state. Many laws can be voted on simultaneously and even if a new law is voted on every day, Random Democracy is still hundreds of time cheaper than 1 election every 5 years.
  • Random Democracy makes the state efficient by linking taxes to services. If every law and public service has its own budget, citizens can make an informed choice. Now the government budget is like a buffet: all you can eat for a fixed amount of tax.
  • Random Democracy is cheap and efficient for citizens.  If 1 new law would be voted on every day, an eligible voter would be called upon to do "voter duty" only once in a lifetime!
  • Random Democracy is really representative. Traditional democracy suffers from a voter bias because rarely more than 50% of the people show up to vote. 

The Transition
It will be hard to transition from the current system to Random Democracy. Not because the target is difficult to achieve but because the resistance of the status-quo is enormous. But that does not mean it is impossible. Versions of direct democracy that come very very close to Random Democracy already exist. More about that in the next blog.

Tuesday, May 7, 2013

Finding a New Name for PDIA

The Problem:
Various voices are calling for a new name for PDIA. PDIA stands for Problem Driven Iterative Adaptations. It is based on four principles:

  1. Solving locally nominated and defined problems in performance (as opposed to transplanting preconceived and packaged “best practice” solutions). 
  2. Create an authorizing environment for decision-making that encourages positive deviance and experimentation (as opposed to designing projects and programs and then requiring agents to implement them exactly as designed). 
  3. Embed this experimentation in tight feedback loops that facilitate rapid experiential learning (as opposed to enduring long lag times in learning from ex post “evaluation”). 
  4. Engage broad sets of agents to ensure that reforms are viable, legitimate, relevant, and supportable (as opposed to a narrow set of external experts promoting the top-down diffusion of innovation).

The Solution:
Call PDIA, agile reform since it is based on almost the same principles. 
Agile Reform is driven by a common goal and a plethora of trials and errors. It has 5 principles: 

  1. Have a Vision 
  2. Several solutions for the same problem 
  3. Just start doing 
  4. Change is managed at the lowest possible level (subsidiarity)
  5. Big change is the result of the iteration of small changes.

Principle 4 would cover PDIA 1. Principle 2 and 3 covers PDIA 2. Principle 5 covers PDIA 3. It would be nice if Principle 1 would have covered PDIA 4. While there is definitely some affinity, agile reform should learn iteratively and include more emphasis on constituency building. Principle 1 should thus be rewritten as:

  1. Have a vision to engage broad sets of agents for reform

The Solution:
The suggestion in this blog is of course a joke. But the message is clear whether it is called agile reform or PDIA, the paradigms are shifting.  
In a discussion with Duncan Green, Matt Andrews (one of the PDIA authors) said that "we have arrived at a ‘moment’ – a coming together of dissidents from numerous disciplines to reject the logframe/best practice culture and push for something more rooted in reality." 
Duncan provides a list of examples that are not really in line with the PDIA approach. This is probably a better list. 

Sunday, March 31, 2013

DD=OO: Direct Democracy is the Only Option

The Problem:
Karina asked this question on Is direct democracy good for state government?

The Solution:

Direct Democracy is the only way forward because:

  1. It ensures that decisions are made by the biggest possible group which reduces the possibility of elite capture of government.
  2. It encourages everybody to take part in debate which not only ensures people are more aware about important issues, diversity will ensure that solutions are better and more creative as proven by crowdsourcing.
  3. It allows everybody to propose issues that are important to them so the debate is about things that matter to large parts of the population, not only to specific interests.
But if it would be easy to do, some would have tried already. Luckily, information technology has allowed new ways to organizing. Efficiency remains the biggest challenge but it can be solved as follows:
  1. Everybody can propose issues for discussion on a government facebook page
  2. If 2% of the eligible voters click like on a proposal, it is put up for discussion.
  3. Everybody can have their say or make changes to the proposal, but to encourage compromise and limit the range of ideas, amendments must be voted up. The submitter of the original idea can revise a proposal taking into account various amendments to improve the quality of the proposal using the wisdom of the crowd and to increase the support for the proposal from a wider group.
  4. Once the proposal is final, everybody can say whether they are for or against. Again statements can be voted up. 
  5. From experience everybody knows that crowdsourcing relies on a very small group of people to contribute (the so called 1%), there are about 9% that are willing to vote up a proposal. The remaining 90% is silent but should not be voiceless! Therefore, voting cannot be done on a voluntary basis as we will reduce the voting public to the 10% active citizens. So here comes the crucial part of the solution: rather than asking everybody to vote on everything, we need to demand that a random sample of all eligible citizens votes on each proposal. If you demand 600 randomly selected citizens to vote, the result will be much more representative than when you ask all citizens to vote and only 70% actually vote (which is a very high figure for turn out). It is certainly more representative than asking 600 parliamentarians...

The Transition:
This is how direct democracy can work. But let's keep discussing, I am sure that the crowd has even better ideas.

Stealing from the Swiss!

The Problem:

The Swiss are often in the news because they have a system of direct democracy. While this is not really the case as Switzerland has a parliament (national council), it is true it uses referendums once in a while. 
In 2012 for example, the Swiss population was asked 12 questions in 4 referendums of which 4 questions were adopted. Just as a comparison, US congress introduced 3914 bills in 2012 of which 61 were adopted, and that was their worst performance since WOII. 

From experience the Swiss know that asking the People more often than 4 times a year is counterproductive because fewer people will show up. Participation in any election has already dropped below 50% in Switzerland, so any vote can be approved by 25% or less of the electorate. That is probably better than asking 200 parliamentarians but it falls short of direct democracy. Or at least, we can do better.

The Solution:

But there are little gems in the Swiss system: any proposal that gets 100,000 signatures (that is less than 2% of the registered voters) will have to be considered in a referendum. That is a beautiful idea that this blog fully supports.
But the most beautiful part of the Swiss system is not used to its fullest. Each time the Swiss have a referendum, a random sample of the population is selected to count the votes. That is of course a waste of time and resources. But the random sampling is perfect and the referendum should be replaced by asking this randomly selected group to vote for the Swiss population. The result would be closer to the will of the Swiss People then the referendum result. And a lot cheaper too. That is what I would call direct democracy.

The Transition

The key elements of Agile Direct Democracy already exist and have been tested. It is just a matter of putting them together into a functioning system.

Sunday, August 19, 2012

Liquid Feedback with Random Sampling

The Problem:

Liquid Feedback uses a system of direct democracy that allows voters to transfer their votes to other voters whom they trust. While I support the arguments in favour of vote delegation made in this blog (and comment), I would like to raise two argument why I have come to the conclusion why voting is the wrong approach in the first place.

This blog correctly identifies the problem with direct democracy: it violates the return on investment instinct of the rational voter. This means, some people do not have time or do not feel competent to engage in every discussion. Delegating the vote seems to be the most logical solution but has two important disadvantages:

One: Animal Farm Revisited:
The crisis in our current democratic system emerged from a well intended system of representation (i.e. parliament) that has lost its legitimacy because the representatives have been captured by narrow interests (i.e. most often the 1% top wealthy). This tiny group is controlling the 99% not illegally but through systems of influence they can afford: lobbying, election contributions, etc.
If we build a system that relies on direct democracy, the number one rule of crowdsourcing will kick into action: the 1:9:90 rule. Applied to political issues, this would mean, 1% of the people is willing to propose legislation and discuss it, 9% of the people is willing to vote and 90% of the people will remain inactive.
Can you imagine how easy it would be for the current powerful elite to take over such a system. It is not very different from the current system of representation. They only have to provide resources to the opinion they support and their view will win every vote.
Direct democracy will have to defy the 1:9:90 rule. This can only be done by moving away from general voting and opting for random sampling.

Two: Dumb down the Masses:
The second reason why vote delegation cannot be allowed in a direct democracy is to activate the public. The current system and elite have managed to dumb down the voter. They have done so consciously. Reversing the trend needs an active intervention because “dumbing” is a self-reinforcing process. Research shows that “rather than motivating an increased search for information, a lack of knowledge about a specific socio-political issue will (a) foster feelings of dependence on the government, which will (b) increase system justification and government trust, which will (c) increase desires to avoid learning about the relevant issue when information is negative or when information valence is unknown.” So once people are ignorant, they want to become even more ignorant. If you do not have time to read the academic paper, you can watch the short version 30 seconds.  Or the dumb down long version.
To reverse the trend, we will have to force those who have a strong interest in an issue to try to explain their position on the issue to people. That is very different from “please let me vote for you because I am attractive and you will probably not understand what this is about”.
Even the most complex issue can be explained in a way that the average grownup can understand. Climate change explained by Al Gore is a good example. It takes a bit of effort, but the alternative is just not acceptable.

The Solution:

But how do we deal with the problem that not every citizen can become a full time politician? We allow the 1% to express their opinion on a subject to ALL citizens. Then we select a random sample of citizens (or party members or whatever the voting group is). For this small group voting is mandatory and they can review all the arguments made by the 1%. The result of this vote is the same as if the whole population would have voted. (That is the beauty of statistics.) If this system would be applied at European level, a voter would probably have to vote on 1 out of every 1 million decisions. In Belgium the number would be slightly higher: every citizen would have to vote on 1 out of every 40’000 decisions. That sounds undemocratic, but it isn’t. 

The Transition:

Liquid Feedback is the ideal place to test the random sampling system. Initially, there will only be few members and thus the statistics force many people to vote. But as numbers of members grow, the percentage of those who have to vote will go down and we can focus on being part of the 1% on those issues that are very close to our heart. 

Friday, May 18, 2012

Harvard Business Review Confirms Power of Agile Reform

"Who needs an agenda if you have a vision" was the first post on this blog and explained that Agile Reform is an unfocused method to reform complex systems. Agile Reform is driven by a common goal and a plethora of trials and errors. It has 5 principles:
  1. Have a Vision
  2. Several solutions for the same problem
  3. Just start doing
  4. Change is managed at the lowest possible level (subsidiarity)
  5. Big change is the result of the iteration of small changes.

The Harvard Business Review (HBR) now confirms that about 70% of entrepreneurs who had a successful exit (that is, an IPO or sale to another firm), did NOT start with a business plan. "[The businesses] were conceived not with a document but with a feeling and doing for an authentic vision. Clarity of purpose and passion ruled the day with less time spent writing about an idea and more time spent just doing it."
"The best content for a business plan is real-world data based on testing aspects of the concept. ... You want simple, iterative tests."

Principle 2 is not really part of the HBR suggestions, but supposedly start-ups are not really complex enough to have competing solutions for the same problem.

Sunday, May 13, 2012

Crowdcracy Does Not Need Elections

This Presentation shows graphically why we are better off when we have abolished elections.