Aretae raises the question with respect to formalism: Doesn't it depend on the interests of the ruler and the ruled being aligned?
The justification of democracy is that by making the rulers answerable to the population, it prevents the rulers from acting in a manner that is good for them and bad for the population — such as spending all the money on themselves.
Formalism in the true Moldbuggian sense has an answer to that: If a voter has actual influence over the government, that should be recognized alongside whatever other actual influences exist, and turned into a shareholding in the government. That makes the value of the influence more predictable, which makes everything more efficient. Every share in the government is the same as every other, so there is no more need for battle between newspapers and civil service departments, unions and universities, to make one group's influence more than another's. Everything runs much more smoothly, and everyone is better off.
I am not a true formalist, however. I see the joint-stock sovcorp as highly desirable but quite impossible. The enforcement of shareholder rights depends on the cryptographic protocols which link shareholdings to the ability to activate or deactivate the security force's weapons. Without disputing the existence of protocols with the correct theoretical properties, I am utterly unable to imagine them being implemented successfully. It is amusing to contemplate control of the world's armaments falling into the hands of Anonymous, but nobody is ever really going to risk it.
So, without formalism, what is my own response to the conflict of interest between ruler and ruled? It is to live with it. An absolute ruler will rule in his interest and not mine, and will raise money from taxes for his own use.
The ruler will be in the position of the proprietor of a firm; he is in a position to take any spare cash in the economy for himself. Like any government, he can levy taxes on anything he wants, and like any proprietor he can use the revenue raised to invest in the firm, or withdraw it from the firm as a dividend.
That brings us to the Laffer Curve. Everyone but the dimmest of left-wingers accepts that at some point, increasing a rate of tax decreases the revenue raised by the tax. However, the normal discussions of this miss a whole dimension, of time. Tax rates today affect not only the size of the tax base today, but also the size of the tax base tomorrow and into the future. The tax rate that maximises tax receipts over the next 12 months will not be the same as the tax rates that maximises receipts over the next 10 years, or the next 25 years.
In an idealised model of a proprietor of a state, with perfect foresight and perfect security, any extraction of tax that reduces economic growth would reduce the NPV of the proprietor's interest. In more realistic situations, that would not hold; the rational proprietor would seek to diversify by taking profits out of the state and moving them into other investments, even at the cost of some impact on the profitability of the state.
My support for the idea of a secure, absolute ruler is motivated by the expectation that the cost of what the ruler takes would be smaller than the cost of the deadweight loss imposed by a government in which nobody has a significant interest in overall long-term growth, but which depends for short-term survival on appeasing large and changing interest groups — whether organised voter blocs, civil service departments, the military, or any other party on which an insecure government relies for survival.
I am much less worried about a proprietor's extraction of profit from a country than I am about how much he will have to do to stay in power. That is the most important divergence of interest: he has an overwhelming interest in preserving his rule, whereas I am much less concerned.
All but one of Aretae's examples of bad rulers caused damage not to gain wealth from the country, but in the course of maintaining power. The exception is King Leopold's rule of the Belgian Congo, which was not in any sense a productive economy, but merely a pile of valuable ivory over which ran wild animals and (in the circumstances) uncivilisable natives — the experience does not extend to any country which is not a backward colony of a more advanced civilisation.
The example that is most troubling for me is not Stalin or Leopold but Kim Jong-Il. The same family has ruled there for 60 years, and secure rule in my theory should have produced good government. My assumption is that, while Kim Il-Sung and his successor have succeeded in retaining power, the power of the ruler is neither complete nor secure, and they are in a constant struggle with rivals within the regime. However, the lack of information about the internal politics of North Korea means that there is little evidence for or against this assumption.