If your point is that "monopoly" has a different meaning in a colloquial/political context and in a legal context, that it relies on non-obvious definitions of what the relevant market, what dominant is, and that it varies by country, I do not think it strongly supports the assertion that it does not have a loose definition.
Because it is generally accepted that monopolies are bad and that not-monopolies are ok. It is much easier to sell the simple idea that "X is a monopoly, so split X", than to make a long case that "X does Y and Z which meet ABC definition of unfair use of market position, bla bla bla, so split X". Conversely, for a company in the wrong, it is easier to say "Of course monopolies are bad, but we are not a monopoly!".
State-owned monopolies exist in most capitalist countries, from operating prisons, to managing railroad or telecom infrastructure, to nuclear energy. The US may have fewer of those than say, the UK or France or Singapore, but I would not call those countries _communist_.
I think you missed the point - representatives should be expert (at a minimum) in HN interests or come from the same demographics (better) as HN commentators. Anything else means they are "morons", to use the word of the previous commentator.
Words get abused a lot, and for some of them, they do not have a clear-cut, universally agreed upon meaning. "Monopoly" is one of such words (others candidate include: freedom, democracy, justice).
When random person X complains about, say, Google having a monopoly, there are two possibilities:
- Person X means something along "Google has a dominant market position and abuses its power", slightly abusing the meaning of "monopoly".
- Person X means Google is literally a monopoly, that it is not possible to get online ads otherwise and does not know that firms such as Facebook exist.
Somehow a lot of people choose to believe interpretation #2 is true, and spend a lot of time debating whether this or that company is a "monopoly" as if it is somehow more important than the substantive issues.
> Ultimately, this isn't a step towards more government surveillance.
Architect of (illegal) government surveillance gets influential role in a major Internet company.
I do not expect him to strongly criticize Amazon's privacy violations or questionable data collection. I fully expect him to be supportive of Amazon initiatives that will make for an easier government surveillance (even if only for improving their chances at getting government contracts).
Of course the US government cannot coerce Amazon any more with Alexander on board. Yet having powerful, government surveillance supporting, well versed in building systems of government surveillance, individuals on the board will surely have a very different impact than having a privacy rights activists on the board.
I don't think such calculations are particularly useful, maybe about as much as Paul Graham's wealth tax "model". The reason is that tax codes are way more complicated than just nominal tax rates.
Did you know, for instance, that you taxable income is 90% of your real income [0]? So the 45% rate kicks in at a 175,340€ wage, not actually 157,806€.
Anyway, it does not matter because of the unusually large income splitting [1]. If both adults have a 157,806€ wage and say, two kids, the total income would be 315,612 with three fiscal shares, and thus would pay 3 times the amount of taxes owed for a 315,612/3 income (i.e. 105204€), where the marginal tax rate is 30%, not 45% [2].
Anyway, it does not matter either because the main income tax in France is not the "income tax", but the "generalized social contribution" (flat rate).
My point is not to write an essay on French taxation, but to show that simply comparing tax brackets and rates is useless, since the definition of "taxable income" is not the same between different countries, how brackets, rates and taxable income are used to actually compute the tax amount is not straightforward, there are many others taxes, and so forth.
[2] Amount of tax is number of n T(i/n), where i is income, T is the function which maps income to taxes owed and n is the number of fiscal shares. Because T is convex, n T(i/n) is less than T(i).
All this feels so wasteful, especially the CIR and formerly the CICE. Instead of having low taxes, you have nominally high taxes and a bazillion paperwork-heavy to make them low. Might as well just make them low in the first place.
> In contrast to Chamley-Judd, the optimal tax on capital is positive in our model because we have finite long run elasticities of inheritance to tax rate
> I actually worry a lot that as I get "popular" I'll be able to get away with saying stupider stuff than I would have dared say before. This sort of thing happens to a lot of people, and I would really like to avoid it
Paul Graham, as quoted in Maciej Cegłowski's blog post "Dabblers and blowhards"
Property taxes are a wealth tax, specifically a tax on real estate wealth. It's hard to see why taxing this form of wealth is so great, but other forms of wealth is so bad.
As for the argument that the US should be more decentralized - less money goes to the federal government, more to the states - this may or may not be true, but this applies equally to all taxes, not wealth taxes in particular.
The extreme case of this is France, where the "income tax" (90B€/year, progressive) is not the largest income tax. Rather, it is the "generalized social contribution" (124B€/year, flat rate). Of course, public discourse is focused on the income tax (50% of households don't pay any tax whatsoever!), not the main tax on income.
An important point here would the magnitude of those effects: how much a 1% wealth tax would reduce wealth creation? Whether it is by .0000001% or by 99%, it would be "incentives to create less wealth" but in the former case it is all but negligible and in the latter case it is a catastrophe.
Capital tax opponents seem to always use the elasticity of wealth creation with respect to the wealth tax rate is extremely high, but I do not remember seeing any evidence on this.
