Imagine a country where up to 50% of the tax dollars are allocated by the people instead of congress/government. Once a year, the citizens file a vote with their tax return, which allocates percentages to the programs which they feel are most important. Each citizen controls an equal portion of the taxes (an equal fraction with no difference based on income).

The requirements for establishing a program are as follows:

  • 0.5% of the citizen population must provide their verifiable signature to support a proposed program.
  • Programs must comply with all laws. They can't do anything illegal.
  • Programs cannot support the election of a candidate for any office.
  • Programs must publish an itemized budget each year of where funding went.
  • Programs are limited to spending 5% of their budget on advertising or external communication.
  • If benefits are limited to a group (for example low income individuals) this must be stated up-front. No part of the criteria to receive benefits can be based on the perceived support for any program.
  • There is a 6 month period from public announcement of the program where a veto is allowed to be established. If the veto is able to get twice as many signatures as the proposed program, then the proposed program will be cancelled.

Assume the following systems are established and working properly under this model:

  • The country has a democratically elected leader and functional court/law system.
  • Each citizen has full free speech and there is a functional news media. This news media is actively engaged in reporting underfunded issues, programs that are improperly using funds, and expected allocation of funds based on pre-polling.
  • Each citizen's vote is verified properly to come from them.
  • Each citizen is able to vote securely and privately. Votes remain private.
  • All votes are counted properly. The system for counting the votes is properly verified at all stages.
  • Programs do not receive any information about who voted for them. There is no way for them to verify this information.

All citizens get control over an equal portion of taxes. In a nation with $2 billion in tax revenue and 1 million citizens, they'd each get to allocate $1000. They are allowed to split the vote between any number of programs with up to 0.1% precision.

As an example, someone might choose to put 50% on a program to cure cancer, 20% on a program that repairs a highway they use regularly, 7.5% on a space exploration program, and the remaining 22.5% on a project to end homelessness by providing affordable housing for people. In that case, 500 dollars of the tax revenue would go to the cancer program, 200 dollars to the highway repair program, 75 dollars to space exploration, and 225 dollars to the homelessness project.

Each year the projects would receive their budget based on the votes and the total taxes collected. Votes would be collected securely and projects would not be told who had voted for them.

Given the above system and assuming that the above systems established function correctly (as listed above), what are the biggest problems that would or could occur with this system? By biggest problems, I mean the worst possible mis-allocation of funds that could reasonable occur, which results in the most undesirable outcome for the citizens of this country.

  • 14
    The biggest problem is that you are relying on people who don't know what they are doing, don't know how other funds are being allocated, and don't know funding requirements. So you end up with a highway that could be plated in gold, and a space program that can't even afford pencils, because everyone wants their highway fixed and no one cares about space. You also run into the problem that giving everyone equal control over the same amount, when they didn't put the same amount in, is going to upset a lot of people.
    – Jonathan
    Commented Nov 30, 2017 at 22:07
  • 4
    I'm allocating 50% of my taxes to space pencils!
    – user1530
    Commented Nov 30, 2017 at 22:17
  • 4
    @blip don't knock the space pencil. It was rather brilliant. It's the million dollar space pen that might not have been. scientificamerican.com/article/fact-or-fiction-nasa-spen
    – userLTK
    Commented Nov 30, 2017 at 22:33
  • 2
    It's worth noting that this is cross-posted from Worldbuilding.
    – Brythan
    Commented Dec 1, 2017 at 0:47
  • 5
    This question seems like it'd be more appropriate on Worlbuilding SE.
    – Wes Sayeed
    Commented Dec 1, 2017 at 3:33

4 Answers 4


The 2 major issues I see arising:

  1. Rule by special interest. Most people take many gov't functions for granted, preferring to rise to a single cause instead. This ends up creating a group of special interests that end up with disproportionate funding compared to core services that people actively depend on that get horribly underfunded. You can also get very contradictory funding proposals (I'd imagine a project like the Keystone pipeline is a good example here...people opposed to it are opposed to it enough to suggest 100% of their tax dollars go to opposing it while those that support it may only give 2% of their funding to supporting it because they have a different cause to allocate to instead).

    In the US in particular, there are some very partisan components currently at play with one side very harshly attacking the other and I wouldn't put it past either side to use this to attack the other. How exactly would the American gov't function if 10% of their budget was going to investigate the clintons and another 10% of tax budget getting scooped up by the 'impeach the peach' project?

  2. People are going to screw with the model. Have we learned our lesson with Boaty McBoatFace (A UK attempt to get the public to name a new science vessel ended up with the top choices being "boaty mcboatface", "ss Hitler was right", and 'the gushin granny')? Mandatory voting in places like Brazil has caused some pretty humorous candidates (http://www.cnn.com/2014/10/02/world/americas/brazil-election-wacky-candidates/index.html) that include several Obama impersonators, wonder woman, and even a couple Bin Laden's.

From the cnn article

But some Brazilians don't share this sense of humor. One voter blames the proliferation of the humorous ads on "low education levels" and voters who think "elections are just a joke."

For this truly to work, you need a very informed and educated populace. I'm decently sure if you tried to do this in today's world, 5% of your countries GDP could be focused getting the flying spaghetti monster into space https://en.wikipedia.org/wiki/Flying_Spaghetti_Monster or a country wide manhunt for the Sasquatch.

Edit: Here is what happens when Texans got to name their garbage dump

Fred Durst Society of the Humanities and Arts

When Austin, Texas, asked residents to help rename its Solid Waste Services Department, the majority polled were in favor of a submission inspired by the frontman of ’90s rap-rock band Limp Bizkit. Nearly 30,000 people voted for 24-year-old Kyle Hentges’ motion to rebrand the garbage dump the “Fred Durst Society of the Humanities and Arts,” a name which even Durst himself publicly supported. “I want to thank all of you who are helping me in Austin. I hope we win,” he tweeted.

Pre-internet years, this crowd-sourced naming model actually did decently well and was responsible for names like the Seattle Seahawks and Toronto Raptor team names. There is something about the anonymous internet that brings out the absurd in people...more and more, if you give the public the opportunity to screw with something, we are.

  • 4
    The Obama administration had to make a statement rejecting building a deathstar.
    – user9389
    Commented Nov 30, 2017 at 23:34
  • I had to google that to confirm. techcrunch.com/2013/01/11/… titled : "Government White House Responds To Death Star Petition: Obama “Does Not Support Blowing Up Planets”" . This isn't simply sub-internet culture like reddit either, alot is simply facebook arranged. Taylor swift's contest had a school for the deaf in Boston get top nods to 'hear' her. Pitbull launched a contest that saw him exiled to Kodiak Alaska (using the clever hashtag #ExilePitbull). Protest = Screwing with.
    – Twelfth
    Commented Nov 30, 2017 at 23:42
  • 1
    If people want Boaty McBoatface then why is this a problem?
    – mike3996
    Commented Dec 1, 2017 at 9:30
  • 1
    Perfect. Let them name the boat Boaty McBoatface. Either this is so embarrasing that next time people will be more serious, or it isn't that bad at all. (There even is a chance that people are thinking they are so seldom taken seriously concerning actual democratic decisions that they make fun of such decisions if they have the possibility.)
    – Thern
    Commented Dec 1, 2017 at 10:32
  • @Nebr - You really want to hinge fiscal decisions on the effectiveness of collective embarrassment? I'm allocating 50% of my tax dollars to the 'peoples pants pulled down in public' fund to commemorate embarrassment driven course correction ideas.
    – Twelfth
    Commented Dec 1, 2017 at 23:09

This suffers from the same advantages and disadvantages of most systems of direct democracy: People don't do details.

The government spends money on benefits, health, education, defence, etc, etc.... Now, I have certain priorities and values. And I want all these parts of government to work properly. I don't know how much each section "should" get in order to function. So I want to get someone who understands this well, and who shares my priorities to do the detailed work. To this end, I vote in elections for the party that I feel shares my values. That party then gets someone to work out the details for me.

The effect of assigning tax spend by direct democracy is most likely an unbalanced economy, with "popular" programs getting very large amount of money - leading to wastefulness. And necessary but unpopular programs being underresourced.

Consider "welfare". It is generally unpopular because it involves giving money to people who are not working. It may get very little money in your system. The people who lose their job end up dead on the street. Welfare is unpopular but necessary.

On the other hand "cure for cancer" sounds like a very popular program. But just giving this program 10% of GDP won't make cancer go away overnight. Cancer research needs brains and time as much as it needs money. And if I'm paying for cancer research, does that mean that less high profile research is ignored. AIDS would have got very little research funding in the political climate of the 1980s

The setting of a budget is complex and subtle, It's not something I want to entrust to me, or my fellow citizens.

  • 2
    The core of this answer is the assumption that "experts" can allocate resources better than a group of people. This is not an established fact. In fact, it's a heavily disputed one. Just ask any economist that advocates for free market policies. In computing, we see that AI algorithms like ant colony and genetic work vastly better on complex problems with lots of inputs and variables. Throwing an "expert" (a solution which attempts to weigh all factors an formulatically arrive at a solution) at such problems more often results in sub-optimal solutions.
    – jpmc26
    Commented Dec 1, 2017 at 0:45
  • 4
    @jpmc26 Ok, but somebody somewhere has to work out the details for a program, and without the details, what defines what the program even is? Say there's a program to do something universally popular, like giving little wool hats to orphans who live in cold places. Who decides exactly who qualifies as an orphan and what's a cold place and how many hats they get and where the hats come from and how they get distributed and how much of the budget is spent on administration? And that's simple; say, a space exploration program requires far more decisions about what to explore and how to do it. Commented Dec 1, 2017 at 5:29
  • 5
    @jpmc26: To the extent that free markets work well, it's because of the close connection between how people use their resources and and the benefit they derive: everyone is an "expert" in how to use their resources in a maximally beneficial way (provided that we ignore, or somehow address, externalities and other market failures). The OP's proposal, however, has people using others' resources to try to directly solve global problems. Due to their rational ignorance, they won't do as good a job as a high-quality bureaucracy.
    – ruakh
    Commented Dec 1, 2017 at 5:41
  • 1
    @ZachLipton You wouldn't even state the problem that way to an AI algorithm. You'd specify the goal to maximize: whether people are warm enough. Then you'd specify potential solutions and assign weights to them (e.g., costs and effectiveness): hats, sweaters, fireplaces, what have you. Such algorithms convert the process of evaluating the effectiveness of a single solution into a mechanism for finding near optimal solutions. You do need an expert to define what's warm enough and in estimating the costs and effectiveness, but this is simpler than what you describe.
    – jpmc26
    Commented Dec 1, 2017 at 9:30
  • 4
    @jpmc26 I realize that there's debate as to whether this question belongs on this SE site, but I'd wager that an answer that assumes a super-intelligent AI capable of designing government programs is entirely speculative in nature. In any case, you've just moved the responsibility of formulating the details from experts to an algorithm, which makes setting the algorithm's inputs and constraints all the more important (lest your AI start setting houses on fire to provide warmth to neighborhoods). All you've done is shift the blame to a computer; the program's rules still have to be set somehow. Commented Dec 1, 2017 at 10:37

This assumes that projects can do roughly equally useful things with any amount of money. Realistically, many projects can't do anything useful without at least a certain amount of money and can't usefully use more than some particular amount of money.

Taking this into account, it makes much more sense to assign a particular amount of money to each project and then have people vote yes or no, accepting those projects that receive enough votes. This is essentially what pretty much every Democracy does today.

  • 1
    So, Kickstarter?
    – ruakh
    Commented Dec 1, 2017 at 19:35
  • 0.5% of the citizen population must provide their verifiable signature to support a proposed program.
  • There is a 6 month period from public announcement of the program where a veto is allowed to be established. If the veto is able to get twice as many signatures as the proposed program, then the proposed program will be cancelled.

To me, regardless of the other points, is the easiest way to put forward scenarios that show problems.


You're requiring only one in two hundred people to support something in order to get it through, and one in a hundred + one to know that it is going through and be organized enough to cancel it.

You can overwhelm this system alone by pure volume of programs submitted.


How does a program for lobbying for an 0.8% reduction in tariffs for imported tomatoes sound? Is it better or worse than lobbying for a 0.4% reduction? What if it's part of a lobbying package that reduces all fruit tariffs by 0.6%?

And why would the average voter care?

Funding of Government initiatives

How does the Government use discretionary funding? Only for the specifics that voters voted for? What about mid-term policy changes, in response to external factors (crises etc.)? What about where action is immediately required, but requires a complex solution, like the GFC? Australia's treasurer won Euromoney's Finance Minister of the year award for implementing government spending, contrary to what the rest of the world did, to ride out the GFC


Once funding is allocated to programs, who runs them? Public servants? Politicians? You're limiting the money spent in these programs, but not the broad direction in which they operate. You're still letting the administrators run something in line with their own ideology to a certain extent.

Not the answer you're looking for? Browse other questions tagged .