By way of pointing out the difficulties (and wondering about the "conscience" of the very wealthy), here are the results of a poll of the US 1% (pdf) (compared to the rest of the population):

h/t naked capitalism.

A yearly tax on assets over a relatively high threshold, single digit millions, per year would be harder to avoid. Include a 300% penalty for undeclared assets with the ability of private individuals to win a substantial percentage of the take from successful prosecutions based on efforts they undertook and there would be a strong disincentive to conceal assets. And the tax should be graduated, increasing at each order of magnitude increase in wealth. At least this would capture real property owned in the country of domicile and the stocks and bonds of publicly listed corporations.

Compared to the efforts required to pass such a law, enforcement would be a breeze. A simpler approach would be to tax exclusive use of commons in a Georgist fashion. This would fall on real estate and intellectual property and on the privilege of incorporation. A comparable tax on the domestic earnings of foreign corporations and individuals would round out the system in either approach.

Given the difference between income and assets the rate could be low. If effectively enforced less than a 1% rate should suffice.

"It is not necessary to have hope in order to persevere."

the fact that tax arbitrage is a survival strategy for small countries.

Maybe we need to get rid of the small countries.

AFP via Google: Liechtenstein accuses Germany of attacking its sovereignty (February 19, 2008)

Liechtenstein on Tuesday launched a withering attack on the German government for paying an informer in the Alpine state for bank secrets to launch a tax probe targeting hundreds of wealthy Germans.

Liechtenstein's Crown Prince Alois said Berlin had violated the sovereignty of the tiny principality and broken the law by buying secret data and sending in spies to uncover the scandal.

...

The investigation is said to involve four billion euros (5.8 billion dollars) hidden in secret bank accounts in Liechtenstein, a tax haven nestled between Switzerland and Austria, by Germany's business elite.

Cry me a fucking river, Alois of Liechtenstein.

A society committed to the notion that government is always bad will have bad government. And it doesn't have to be that way. — Paul Krugman

The fundamental challenge in developing a fair and effective tax collection system which would enable states to pursue basic public purpose is doing so before the current system crashes the world climate and ecology. The former is a prerequisite to the latter. But I don't know if even a wider appreciation of that issue will matter. If we are a social species it would seem that we are one hardwired to self destruct.

"It is not necessary to have hope in order to persevere."

Here is a game-theoretic perspective.

Nash equilibria are supposed to describe stable configurations of player's preferences, where no one can improve his payoff without the others changing their behavior. Here is a dirty secret: There might be plenty of Nash equilibria (exponentially in the number of players and their options) or various quality (by any statistics), and they are hard to compute. For example, telling if a given game has a Nash equilibrium with a guaranteed payoff for a certain player (or any average or extreme payoff that good enough, etc) is an NP-hard problem. A Nash equilibrium is basically a way for a game to get stuck at some outcome. Even if we define stronger equilibria for practical applications, the computational picture should be the same.

A simple example is a coordination game. Say, two players have to tell a number from 1 to N, and they both get a prize if they tell the same number. Otherwise they get nothing. Agreeing on a particular number is a Nash equilibrium clearly. But playing the same subset of {1,2,...,N} with uniform probability is a Nash equilibrium as well - you have to try that without communication (with far less expected payoff). The number of Nash equilibria equals the number of subsets of {1,2,...,N}, which is 2^N (because each number is either in the subset or not) minus 1 (because the empty subset is not allowed).

Other example is a congestion game. A cargo ferry leaves every hour, and K players want to transport their goods. The cost for a player is H+P*X, where H is the number of hours (s)he waits, X is the number of players using the same ferry, and P>1 is a constant ("latency"). For a Nash equilibrium, you can order the players and let them use the ferry one by one. That is pretty effective in the total or average cost. This gives K! equilibria. More democratic equilibria are obtained by randomizing ferry choices for players, or mixing player ordering and randomization. There is an overall cost for the equal treatment of players, however.

Today's political economics is clearly stuck at some nasty equilibrium, where most of the players do not have much choice but to suffer high rents, low wages or no income at all. Intentionally or not, the rich were seeking this equilibrium consistently in the last decades. Since the rich control the rules as well, the system is stuck for a long time. What would it take to escape this equilibrium? Serious planning and action will certainly be required. And by definition, non-optimal choices (aka sacrifices) at each step, for you and for many others, will also be a must.

The Hun is always either at your throat or at your feet. Winston Churchill