The biggest problem, IMO, is that what people want to pay for which things is only loosely correlated with what makes sense in reality. You would overfund the sexy line items and underfund the boring-but-necessary ones.
Few people have the bandwidth to meaningfully participate in such a direct democracy.
Wealthier people have different concerns and interests than poor people. So any system making voting power proportional to wealth is liable to result in the upper class voting through changes that defund various forms of assistance/subsidy for low incomes. Including things that are broadly socially desirable, but the wealthy aren't using.
Like what already happens by way of representatives paying more attention to wealthy constituents, business owners, and donors—but even moreso by formalising their ability to simply directly vote with their money.
You raise some problems with the current system:
Who polices the police? Voters can be ignorant. Elections can be rigged.
None of them seem to be particularly bad, we do have standard ways to deal with them, but sure enough let's see your proposed solution.
But suppose that tax payers were able to vote for their own tax rates, and voting power was proportionate to taxes paid? And suppose the vote took place via crypto, so that the results of the vote could neither be rigged nor ignored?
I don't see how this proposal helps us with police taking protection money from the people. Or with voters being ignorant. And what's the point of weighting the power of vote to the tax money? It seems that you want this for reasons completely unrelated to what you've mentioned in the post beforehand.
I can see how crypto theoretically solves the issue with rigging the votes, at the cost of it's own bunch of problems. For instance, it complitely obliterates the secrecy of voting and makes the whole system more expensive and slow.
As usual, cryptocurrency is a solution in a search for a problem. For any its application, you can most likely achieve the same benefits with less downsides without using crypto.
A problem with all systems of government is that whenever a person is given power over something, he can do whatever he wants with that power. An example might be that citizens vote to fund a police force, but then the police extract protection money from all the local businesses in excess of what the citizens were intending to pay in taxes when they set up the police force. If you set up a higher government over the police to police the police, then what stops that higher government from doing something similar to what the police were doing?
Voting causes another set of problems, in that votes can be rigged, and voters can be ignorant.
But suppose that tax payers were able to vote for their own tax rates, and voting power was proportionate to taxes paid? And suppose the vote took place via crypto, so that the results of the vote could neither be rigged nor ignored?
In order to avoid freeloaders, each tax payer would not individually pay what he thought he should pay, but all the tax payers collectively would pay whatever the consensus rate was. In this case, theoretically, about 50% of the taxes paid would be voluntary. This is because if more than half the taxes paid were involuntary, then the tax payers would have the power to lower their tax rate.
I think this system would not work until we already had a world where it was normal to pay for things via crypto. Then a government could set up a system for their own country within some preexisting crypto infrastructure (like ADA, or ICP).
Practical details about how it could work:
Crypto cannot fully track everything that happens in the real world. So, I think in order to do this, people would have to do something like declare an "income wallet" and an "expense wallet". Taxes would be extracted from the difference between the income going into the "income wallet" and expenses leaving the "expense wallet." There would have to be occasional audits to punish people for cheating. The government could optionally declare some items as tax deductible, such as items flagged with "groceries", "healthcare", or possibly "rent", so that people could pay for these through tax deductible expense wallets.
Examples:
Theoretically, ALL forms of taxation could be handled through this one system. I think it would be harder to cheat than one might think, because crypto is a public ledger. For instance, if Steam sold a game which they flagged as entertainment, but the buyer tried to flag it as groceries to have it be tax deductible, then there would be a discontinuity which an automated system could easily pick up on. Both sides of the transaction would have to lie in order for an item to be flagged incorrectly.
This system would work well with largely autonomous government agencies. For instance, the local government, state government, the monarch, and the military could all have separate wallets with their own tax rate. If a citizen, for instance, loved the military, loved his local government, but hated the monarch and the state, he could vote for taxes for the military and his local government, but vote for a tax rate of 0% for the monarch and state. Each branch of government then would be individually accountable to the tax paying public.
If each branch of government is independently accountable to the public for their funding, then it seems to me that no other system of accountability would be necessary. Each branch of government could be largely autonomous. For instance, the police could recruit their own members according to their own preferences without interference from an elected official, and likewise with the military. If they stepped out of line, either they would be punished by higher branches of government, or defunded by the public.
One problem would be that a government agency could choose to pay taxes. They would do this so that they would get voting power to raise taxes, without losing anything in the process, because whatever they pay through taxes would come back to themselves. This behavior would be punished if there were many autonomous branches of government, so that most of whatever a government agency would pay in taxes would go to some other agency.
There could also be a problem that some actors could vote for a billion percent tax rate in order to skew the average. I have thought of two solutions to this problem:
It would be cool if the system were set up so that government agencies could be added or removed. One possible way this could work would be that the sovereign declares a new wallet that he'd like to use to fund a new government agency. If more than 50% of the voting power votes for a non-zero tax rate for this new wallet, then it becomes an official new government agency for which people are obligated to pay taxes. Likewise, if at any time, 2/3 of the voting power votes for a tax rate of 0 for any government agency, that agency is abolished and no one has to pay tax to it anymore. In this case, if a citizen thinks that his tax rate is way too high, but he doesn't want the agency to be entirely abolished, it would be rational for him to vote for a 0.00001% tax rate. A tax rate of 0% would mean that the tax payer does not want that government agency to exist at all.
According to Goedel's theorem, there is no logical system which is both consistent and complete. As this applies to human organizations, I think it means that there can be no system of rules that can adequately address all possible circumstances. I think this means in practice, some degree of arbitrary power is necessary to be able to respond effectively to novel situations.
So I was thinking it might be best if the sovereign in theory has absolute power, but with a couple caveats. He should by default promise to uphold several basic standards (like what's in the Bill of Rights). Also, if he were part of this crypto voter based voting system, then his funding would depend on popularity among the tax payers.
So, by default, the sovereign should act according to widely known traditions (like the Bill of Rights), but in novel situations, he would have the right to act however he pleased, but he would be incentivized to try to justify his actions to the tax payers, or else his funding would be cut.
An example of how no system can adequately protect against all circumstances: The founding fathers were familiar with religious wars and religious persecution, so they established a separation of religion and state. But they were not familiar with plutocracy as it exists today. So, we have a state with property rights, but no checks on wealth being used to subvert the government. We have a central bank with power to print unlimited money with no oversight from congress and a government with legalized bribery (called lobbying). This seems like a huge issue to me. But according to Goedels' theorem, even if the US constitution had some anti-corruption measures in it, there would still be some other way that was unaccounted for which could be used to subvert the government. I think in theory, whenever constraints are placed on the government to avoid abuse, those constraints can be used as an opening through which one could attack the government. I think my idea of the sovereign obeying constraints by default, but in theory having unlimited power, but being dependent upon funding by tax payer vote, is an elegant solution to this problem. Hopefully when some conspiracy arose in order to subvert the government, the sovereign would exercise arbitrary power to put it down, and attempt to publicly justify what he had done. If the tax payers were wise, then they would reward or punish the sovereign according to his behavior.
Another issue with voting: a person cannot give informed consent to an idea which he does not understand. I think this means that the ideas that will be implemented tend to be understood by only about half of the ruling class. Suppose there were an idea which required one to be in the 80th percentile of political wisdom to understand, and another idea which would not work as well, but which could be understood by people in the 40th percentile. Then the idea in the 80th percentile would get 20% of the vote, and the idea in the 40th percentile would get 40% of the vote (everyone in between the 40th and 80th percentiles). I think an idea is stable at around the 50th percentile. Any idea significantly more complicated, and too few people understand it to support it. Any idea significantly dumber, and too many people will realize that the idea is dumb for the idea to get any support. So, a restricted voting class can result in superior policy if the voting class is wiser than the general public. I would think if the society were at least partially meritocratic, then that means richer people would on average be more qualified to vote than poorer people, and so this system would have a wiser voting class than in the case of unlimited suffrage.
If I were dictator, I would not immediately implement such a system. It is complicated, so there may be means of subverting it that would be hard to discern ahead of time. I would declare my intention to set up such a system a few years in the future, so that smart people could try to find ways of attacking the system. If it seemed to people in-the-know that a way was found to keep the system secure from subversion, then a public announcement would be made that the system would be implemented on a certain date a few years into the future. If the system were set up correctly to start with, I think it would be extremely hard to subvert.