For example, I'd love to see a progressive tax rate for corporations: companies' tax rates would be based on total revenue or profits. ex. If your company earns less than $50k/year, no taxes, 10% tax rate up to 250k, 20% up to $1M, 40% up to 10M, 50% on earnings over $10M.
This would provide a strong incentive to only maintain a large corporation if the value of your economy of scale exceeds the additional tax burden. Most companies would have an incentive to break up into many smaller subsidiaries. This would provide additional opportunities for competition, prevent profitable departments from being dragged down by executive mismanagement, and minimize the ability for power and wealth to be concentrated.
Would probably need some anti-collusion laws to make sure entities are truly 'separate' to prevent shell games. But I think it would be pretty useful if it could be implemented properly.
The US's largest chemical company, fairly close in size to BASF, is Dow Chemical, also more than a century old
> For example, I'd love to see a progressive tax rate for corporations: companies' tax rates would be based on total revenue or profits. ex. If your company earns less than $50k/year, no taxes, 10% tax rate up to 250k, 20% up to $1M, 40% up to 10M, 50% on earnings over $10M.
What sort of politicians would actually push this through Congress? What stops it from being repealed with the next cycle? Without a culture of socialism and understanding that corporate/owner and individual interests are not incentive compatible, none of this would work.
I think the only practical way to fix this problem is to remove the two party system which allows representatives to vote against the interests of their constituents except on the issues that strongly separate left and right. And the only way to remove the two party system is to remove the first past the post and move into some kind of proportional representation system. And that only takes constitutional changes...
"some kind of proportional system"
You can't implement a real government with a metasyntactic electoral system, obviously. You need a specific system. I think the answer is Schulze STV, but most people don't know what that is.
What's the problem with just assigning a number 0-10 for every candidate that you care to, and summing up the points?
Or just instant runoff?
That's Range Voting, and it's my preference, but the main problem is that it's complicated both for the voters, and in implementation. Approval Voting is a simplified form (reducing the range to just 0 or 1 vote per candidate) that has many of the same advantages, but has the benefit that people understand it more easily.
The problem with this (Range Voting) that there is no consistent mapping from preferences to 1-10 ratings, so that equivalent ballots have different meaning and equivalent preferences result in different ballots from different voters.
Approval has the same problem for normal elections.
This issue is overlooked in most analysis, which generally presume the existence of a simple quantifiable utility function and a consistent mapping from that function to ballot markings for these non-ranked-preference methods.
In social choice scenarios without secret ballots, you can tie concrete consequences to voting which eliminate this issue and make either method sensible.
Range and Approval (and IRV) are also single-winner systems and don't deal with the general problems with single-member district systems, as much as they may be marginally better means of electing in such districts than FPTP.
Schulze STV is a bit opaque, you are correct, and traditional STV (or STV without loser elimination) is more readily understandable and possibly practically better despite some theoretical deficiencies compared to the more opaque Schulze variantes, simply because it is more comprehensible.
Isn't this just a fundamental problem with voting? In FPTP, one voter might not like the candidate they vote for as much as someone else, or in a ranked system, someone might like their 1st and 2nd choices a lot more than their 3rd choice.
We give everyone's vote equal weight, and they can decide how to use that. Within that constraint Range Voting gives people the most freedom to express their true preferences.
No, it's not.
> In FPTP, one voter might not like the candidate they vote for as much as someone else,
Yes, but the ballot doesn't carry and the voting method doesn't use information about how much the voter likes the candidate; for honest ballots, the information contained in the ballot (limited though it is) has the same meaning.
> Within that constraint Range Voting gives people the most freedom to express their true preferences
No, it doesn't, because range ratings have no consistent, even in theory, mapping from internal preferences. They don't represent any answer to a consistently meaningful question about internal preferences (again, approval has the same problem.) Both bullet (as used in FPTP) and ranked ballots do (in the former case, the question is "if you could dictate who wins the election who would that be"; in the latter, the same question at first, with "and if they were not an option, who would you choose?" repeated until the ranking is complete.)
As noted before, the lack of consistent information problem goes away in social choice scenarios where a concrete meaning can be attached to the markings. examples:
In hypothetical non-secret ballot system where a range voting ballot represents exactly how much the voter commits, in some unit of currency, to contribute to the group if their chosen outcome is selected, the lack of consistent meaning is resolved.
Likewise for approval with a non-secret ballot, in the case where a an activity is chosen by ballot, and an "approve" mark is a binding commitment to participate in the activity if it is selected (or a non-approval is a binding commitment to opt-out.)